University of Southern California Dissertations and Theses

Conservation and reconstruction of textile blocks: an investigation of treatment and replacement options at the Frank Lloyd Wright Freeman House

(USC Thesis Other)

Conservation and reconstruction of textile blocks: an investigation of treatment and replacement options at the Frank Lloyd Wright Freeman House

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content CONSERVATION AND RECONSTRUCTION OF TEXTILE BLOCKS
AN INVESTIGATION OF TREATMENT AND REPLACEMENT OPTIONS AT
THE FRANK LLOYD WRIGHT FREEMAN HOUSE
by
Benjamin McAlister
__________________________________________________________________
A Thesis Presented to the
FACULTY OF THE SCHOOL OF ARCHITECTURE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF BUILDING SCIENCE
August 2009
Copyright 2009 Benjamin McAlister
ii
DEDICATION
To my family and friends who have continued to stand by me and support me in all
my educational endeavors. Thank you so much for all that you have done for me
and for the opportunities you have provided to me. You are forever in my gratitude.
iii
ACKNOWLEDGEMENTS
I would like to acknowledge the following individuals for their extensive help and
invaluable advice in the course of my research and during the execution of the
experimentation phases of my thesis. Without them, the completeness of this thesis
would not have been possible and the exploration of issues would have been far
less advanced. All these individuals unselfishly offered their time and knowledge in
support of further education and awareness about the Freeman House.
First and foremost, I would like to express my gratitude for the unwavering support
and guidance provided by my committee members – Doug Noble, FAIA, Ph.D.;
Ken Breisch, Ph.D.; and Gail Peter Borden, RA, AIA. I would also like to thank
my many trusted advisors who each contributed their advice and expertise to my
research – Marc Schiler, University of Southern California; Thomas Spiegelhalter,
University of Southern California; Jeffrey Chusid, Cornell University; Peyton Hall,
Historic Resources Group; Kathryn Smith, author of numerous works on Frank
Lloyd Wright; Beril Bicer-Simsir, Assistant Scientist for The Getty Museum Science
Department; Virginia Toledo, The Getty Research Institute; and Thomas Roby; The
Getty Conservation Institute.
For their advice and guidance on block manufacturing and helping to recover
undocumented information, I am extremely grateful to Dana Smith, former Assistant
to the Dean for Special Programs & Projects at USC, as well as students Michael
Curiel, and Michael Villaseñor, who offered their first-hand experience with block
iv
making. I also want to thank Jeff Keenan, President of Moonlight Molds, for sharing
his advice and experience about the process of manufacturing replacement textile
blocks for the Ennis-Brown House.
Many thanks go to Len Marvin of AC Martin Partners for arranging the use of
compression testing facilities for my thesis at Smith-Emery Laboratories in Los
Angeles, California. To John Latiolait of Smith-Emery Laboratories and his staff, I
extend my sincerest gratitude for allowing me access to your facility and teaching me
about the process of testing concrete.
I would also like to thank the facilities coordinators, Ian McCully and Raul Lopez,
for always being willing to lend a hand, for expediting repairs to the block making
facilities, and for their sincere interest in my work. Ruth Wallach, Librarian for the
USC Architecture and Fine Arts Library, was instrumental in helping to acquire
research materials.
To my colleagues in the Chase L. Leavitt Building Science Program, thank you for
your interest in my thesis work and for your hours of hard work helping me to make
blocks.
v
ii
iii
ix
x
xvii
1
2
4
10
17
19
20
20
23
25
25
29
33
35
35
37
37
40
40
47
53
53
56
60
63
TABLE OF CONTENTS
Dedication
Acknowledgements
List of Tables
List of Figures
Abstract
Chapter 1: Investigating the Freeman House and Textile Block Deterioration
1.1 Cultural and historic significance
1.1.1 The Arts and Crafts Movement and textile block homes
1.1.2 A notable work by Frank Lloyd Wright
1.1.3 A gathering place for friends of the Freemans
1.2 Harriet Freeman’s gift and instructions for use
1.3 The textile block system
1.3.1 Function of the construction system
1.3.2 Materials
1.4 Block deterioration
1.4.1 System failure points
1.4.2 Previous retrofitting work
1.4.3 Looking forward
1.4.4 Possible reuse schemes
1.5 Scope of work
1.5.1 Hypothesis statement
1.6 Chapter overview
Chapter 2: History of the Freeman House
2.1 Early 20th century explorations in concrete construction
2.2 Wright’s period of transition in Southern California
2.3 Freeman House architectural history
2.3.1 The family and their home
2.3.2 The design and construction
2.3.3 Architectural modifications
2.4 Previous studies of the Freeman House
vi
64
67
67
69
71
72
73
75
75
78
80
81
82
83
84
85
86
87
88
88
89
90
90
91
91
91
92
92
93
93
95
95
96
100
100
101
102
104
2.4.1 Wank Adams Slavin Associates proposal and report
2.4.2 Jeffrey Chusid deterioration survey drawings
2.4.3 Nabih Youssef structural report
2.4.4 Smith-Emery/American Petrographic Services block analysis
2.4.5 Sumit Brahmbhatt’s thesis
2.4.6 Alice Ormsbee’s thesis
2.4.7 Peyton Hall’s USC class update of Chusid’s drawings
Chapter 3: Concrete Block and Masonry Conservation
3.1 Secretary of the Interior’s standards
3.2 International standards
3.3 Block deterioration found at the Freeman House
3.3.1 Chipping
3.3.2 Cracking
3.3.3 Efflorescence
3.3.4 Erosion
3.3.5 Spalling
3.3.6 Biological growth damage
3.4 Block conservation and repair techniques
3.4.1 Waterproof coating
3.4.2 Water repellent coating
3.4.3 Water washing
3.4.4 Caulking
3.4.5 Chemical cleaning
3.4.6 Poulticing
3.4.7 Composite patching / plastic repair
3.4.8 Consolidation
3.4.9 Mechanical repair
3.4.10 Paint and stucco
3.5 Conservation options summary
3.5.1 Block replacement principle
Chapter 4: Textile Block Research Methods
4.1 Methodology
4.2 Block manufacturing
4.3 Block testing
4.3.1 Slump test
4.3.2 Karsten pipe testing
4.3.3 Compressive strength of block material
4.3.4 Color and texture comparison
vii
106
108
108
112
118
120
122
136
136
140
144
148
149
151
151
152
154
158
165
172
172
176
176
177
179
184
186
187
190
191
191
192
193
195
195
4.4 Data collection and presentation
Chapter 5: Textile Block Manufacturing
5.1 Textile block composition
5.2 Tools and formwork used in manufacturing replacement blocks
5.3 Method for manufacturing textile blocks and tiles
5.3.1 Block manufacturing steps
5.4 Mix designs
5.5 Challenges encountered during manufacturing
5.5.1 First attempts
5.5.2 Improvements in the process through experience
5.5.3 Refining the method
5.5.4 Considerations
5.6 Final recommendations on the manufacturing process
Chapter 6: Testing of Concrete Textile Blocks
6.1 Testing choices
6.2 Slump testing
6.3 Karsten pipe testing
6.4 Compression testing of cylinders
6.5 Color and texture comparison
Chapter 7: Analysis and Evaluation of Testing Results
7.1 Block formation and testing analysis
7.2 Density, hydration, and capillary action
7.2.1 Karsten pipe testing revelations
7.2.2 Dry versus wet
7.2.3 The effect of Karsten pipe testing on mixture designs
7.3 Color comparison findings
7.4 Slump and its effect on block making
7.5 Importance of compressive strength
7.6 Overall evaluation
Chapter 8: Conclusions
8.1 Block manufacturing
8.2 Cement hydration
8.3 Making repairs to the house
Chapter 9: Future Work
9.1 Performance of a replacement block next to an existing block
viii
196
197
198
198
199
200
200
201
202
208
208
215
217
228
229
235
235
238
241
245
251
252
252
259
261
9.2 Testing of other additives or sealants
9.3 Innovative ways to form a block
9.4 Study of the viability of the current structure in an earthquake
9.5 Method for replacing a textile block in the center of the wall
9.6 Method for tiling the cast-in-place concrete walls
9.7 Method for insulating the house
9.8 Methodology for determining which blocks to replace
9.9 Future work summary
Bibliography
Appendix A: Freeman House Drawings
A.1 Historic American Building Survey (HABS) drawings
A.2 Jeffrey Chusid 1988 CAD translation of FLW original drawings
A.3 Jeffrey Chusid 1991-1992 survey of the Freeman House exterior
Appendix B: Note From Harriet Freeman
Appendix C: Charter of Athens
Appendix D: Block Manufacturing Photo Journal and Process Video
D.1 October 8, 2008 block manufacturing
D.2 November 2, 2008 photos of cured blocks
D.3 December 3, 2008 block manufacturing
D.4 Catalog of blocks manufactured
D.5 Video of block manufacturing process
Appendix E: Karsten Pipe Testing Photos and Videos
E.1 Photos
E.2 Videos
Appendix F: Compression Testing Videos
ix
LIST OF TABLES
Table 5.1: Test samples cast on October 30, 2008 and November 19, 2008.
Table 5.2: Blocks cast on December 3, 2008.
Table 5.3: Blocks cast on January 6, 2009.
Table 5.4: Blocks cast on January 7, 2009.
Table 5.5: Blocks cast on January 8, 2009.
Table 5.6: Blocks cast on January 9, 2009.
Table 5.7: Blocks cast on January 16, 2009.
Table 5.8: Blocks cast on February 12, 2009.
Table 5.9: Blocks cast on February 19, 2009.
Table 5.10: Blocks cast on February 28, 2009.
Table 5.11: Blocks cast on March 1, 2009.
Table 6.1: Karsten pipe test results.
Table 6.2: Compression test results.
Table 7.1: Block properties matrix.
124
125
126
127
128
129
130
131
132
133
135
156
161
173
x
LIST OF FIGURES
Figure 1.1: View of the garage and front entrance of the home.
Figure 1.2: A Freeman House pattern-face textile block.
Figure 1.3: The original Freeman House pattern-face mold found in the
garage.
Figure 1.4: Block patterns from Wright’s block houses drawn by artist Charles
Calvo.
Figure 1.5: The Gustav Stickley-designed Spindled High-Back Armchair and
a Frank Lloyd Wright chair for the William R. Heath House in
Buffalo, New York.
Figure 1.6: The Gamble House in Pasadena, California.
Figure 1.7: Alice Millard Residence details of the concrete block assembly.
Figure 1.8: Millard House interlocking blocks exposed during restoration
work.
Figure 1.9: Frank Lloyd Wright’s 1923 patent drawing.
Figure 1.10: Aerial isometric view of the Freeman house.
Figure 1.11: Freeman House living room with view down Highland Avenue.
Figure 1.12: Diagram of a typical Freeman House textile block wall.
Figure 1.13: Detail of a construction photo showing exposed steel reinforcing
rod protruding from blocks.
Figure 1.14: The three main types of block found at the house: left and right
hand pattern-face, perforated pattern-face, and flat-face.
Figure 1.15: Materials piled up on Glencoe Way during construction.
Figure 1.16: Measures to reinforce the walls after the Northridge earthquake.
Figure 1.17: Block cracking and spalling on the south porch column.
Figure 1.18: Efflorescence on the lower floor interior hallway.
1
4
5
6
8
9
11
12
13
15
17
21
22
23
24
26
26
27
xi
Figure 1.19: Exposed steel tie connecting the two wythes of the wall near a
window.
Figure 1.20: Installing support piers along Glencoe Way.
Figure 1.21: New concrete columns on either side of the living room fireplace.
Figure 1.22: Vierendeel girder installed at the living room clerestory.
Figure 1.23: Living room balcony reconstruction using cast-in-place concrete
walls.
Figure 1.24: South sleeping porch reconstruction using cast-in-place concrete
walls.
Figure 2.1: One of six apartment units at Horatio West Court.
Figure 2.2: Walter Burley Griffin’s Knitlock construction system.
Figure 2.3: A Nel-Stone advertisement and wall construction diagram.
Figure 2.4: The Henry Bollman House exterior and view of the block
columns.
Figure 2.5: Lloyd Wright’s Sowden, Derby, and Samuel-Novarro houses.
Figure 2.6: Lloyd Wright’s design for an improved concrete block wall.
Figure 2.7: Exterior of Hollyhock House with canted walls and cast
ornaments.
Figure 2.8: The Little Dipper, northwest elevation and block offset details.
Figure 2.9: Rendering and partial plan of Doheny Ranch, House C.
Figure 2.10: Perspective drawing of Doheny Ranch project.
Figure 2.11: A guest cottage at the Arizona Biltmore Hotel
Figure 2.12: Timeline of the Wright’s Southern California cast concrete
projects starting with Hollyhock House, followed by the Millard
House, the Playhouse for Aline Barnsdall, then the Storer House,
Freeman House, and finally the Ennis House.
28
30
30
31
32
32
41
42
42
44
45
46
48
49
50
51
51
52
xii
Figure 2.13: Harriet and Samuel Freeman.
Figure 2.14: Three generations of Wright family architects contributed to the
Freeman House (left to right: Frank Lloyd Wright, Lloyd Wright,
and Eric Lloyd Wright).
Figure 2.15: Sketch of the Freeman House by Frank Lloyd Wright.
Figure 2.16: Drawings of a stone carving design for the Imperial Hotel and
the design of the Freeman House textile block.
Figure 2.17: The Freeman living room decorated with furniture by Schindler.
Figure 2.18: Aluminum mullions added by John Lautner.
Figure 2.19: Image of Sample 2 at 15x magnification.
Figure 2.20: Image of poorly hydrated belite clusters in Sample 3.
Figure 2.21: Sensors and output data from Sumit Brahmbhatt’s thesis.
Figure 2.22: Karsten pipe testing on the wall of the Freeman House.
Figure 2.23: Photos included in a student’s survey of block conditions.
Figure 3.1: Chipping on the exterior roof parapet and on the east wall.
Figure 3.2: Cracking found on an exterior retaining wall.
Figure 3.3: Efflorescence found in the living room and hallway.
Figure 3.4: Erosion of block faces.
Figure 3.5: Block spalling on a retaining wall near the home’s entry.
Figure 3.6: Ivy growth on the walls and damage resulting from biological
growth.
Figure 4.1: The process of making a Freeman House textile block.
Figure 4.2: Peter Purens making a Freeman House block using the original
mold.
54
56
57
59
60
62
70
70
71
72
73
81
82
83
84
85
86
96
98
xiii
Figure 4.3: Diagram of the Karsten pipe test on horizontal and vertical
surfaces.
Figure 4.4: Compressive strength test stand at Smith-Emery Laboratories.
Figure 4.5: The Munsell ColorChecker® card.
Figure 4.6: Example of histogram sampling in Adobe Photoshop.
Figure 5.1: Block 32, which failed within hours after its casting.
Figure 5.2: Concrete mixing machine.
Figure 5.3: Wooden sifting box with metal mesh screen.
Figure 5.4: The original Freeman House form box open (left) and closed
(right).
Figure 5.5: The reproduction form box open (left) and closed (right).
Figure 5.6: Coffered aluminum back support.
Figure 5.7: Coffered steel backing plate placed on top of aluminum back
support.
Figure 5.8: Form box with coffered back support and backing plate in place.
Figure 5.9: Original Freeman House pattern-face mold.
Figure 5.10: Reproduction flat-face mold (impression surface shown at left; top
surface with handles shown at right).
Figure 5.11: The vibration device and speed control located below the
hydraulic hammer press.
Figure 5.12: Pressure distribution plate used for machine compaction.
Figure 5.13: Hydraulic hammer press and compressor used for machine
compaction of the replacement textile blocks.
Figure 5.14: The block manufacturing process.
101
103
104
105
109
112
113
113
114
114
114
115
115
115
116
117
117
119
98
xiv
Figure 5.15: Block 31 cast with a wet mix exhibits a smooth surface in contrast
to the texture of the existing Freeman House blocks.
Figure 5.16: Checking the consistency of the mixture.
Figure 5.17: The first attempts at making Freeman House tiles showing a
broken corner (left), a stress fracture (center), and a gouge in the
edge (right).
Figure 5.18: A successfully formed tile on top of the wooden support block.
Figure 5.19: Unsuccessful pattern-face block with cracks and broken corners.
Figure 5.20: Pattern-face block with lower channels uncompacted.
Figure 5.21: Misting tiles with water on the curing rack.
Figure 5.22: Block 10 perfectly formed with flat-face mold and 40:20:10 mix.
Figure 5.23: Block 8 in the form box (left) and after removal (right).
Figure 5.24: The additives tried included acrylic fortifier (left), water-stop
cement (center), and clear sealer (right).
Figure 5.25: Block 9 with plastic wrap to trap moisture during curing process.
Figure 5.26: Block 14 cured to a more gray finish than other blocks.
Figure 5.27: Face-up and face-down casting methods that were attempted.
Figure 5.28: The hybrid block process used in Blocks 38 and 39.
Figure 5.29: Block 38 successfully cured.
Figure 6.1: Preparing the informal slump test.
Figure 6.2: Completed slump test showing virtually zero slump of the
material.
Figure 6.3: Clay putty and Karsten pipe used for this testing.
Figure 6.4: Placing the concrete cylinder sample in the testing apparatus.
Figure 6.5: Test samples that were cast in PVC tubes.
123
136
137
138
138
139
139
140
141
142
143
144
145
147
147
152
153
155
159
160
xv
Figure 6.6: Test cylinders made ready with a sulfur cement cap on both ends.
Figure 6.7: Test cylinder S1 before (left), during (center), and after testing
(right).
Figure 6.8: Test cylinder S2 during (left) and after testing (right).
Figure 6.9: Test cylinder S3 after testing.
Figure 6.10: Test cylinder S4 during (left) and after testing (right).
Figure 6.11: Test cylinder 23 before (left) and after testing (right).
Figure 6.12: Test cylinder 24 before (left) and after testing (right).
Figure 6.13: Original pattern-face Freeman House block (no number).
Figure 6.14: Original pattern-face Freeman House block (370 B-8).
Figure 6.15: Original flat-face Freeman House block (55 B-6).
Figure 6.16: Original flat-face Freeman House block (no number).
Figure 6.17: Original block compared to new block #2.
Figure 6.18: Original block compared to new block #14.
Figure 6.19: Original block compared to new block #16.
Figure 6.20: Original block compared to new block #17.
Figure 6.21: Original block compared to new block #21.
Figure 6.22: Original block compared to new block #31.
Figure 6.23: Original block compared to new block #36, parts a and b.
Figure 6.22: Original block compared to new block #38.
Figure 7.1: Block 36 after the addition of the pattern-face topping.
Figure 7.2: Side view of Block 36 showing a very dry mixture topping added
to a cured wet mix block.
161
162
162
163
163
164
164
166
166
167
167
168
168
168
169
169
169
170
170
180
181
xvi
Figure 7.3: Diagram of the hybrid block casting method for Blocks 38
and 39.
Figure 7.4: Block 38 front, side, and back faces.
Figure 7.5: An original Freeman House block compared to Block 31.
182
183
185
xvii
ABSTRACT
This primary goal of this thesis was to develop an improved concrete block mix
and associated fabrication process for replacement textile blocks at the Frank Lloyd
Wright Freeman House. Desired characteristics for the replacement mix were water
resistance and consistency with the home’s historic fabric and historic preservation
standards. Issues explored include the use of modern-day mixture materials, block
formation methods, compressive strength, material slump, water permeability, and
color comparison.
The house is one of four Wright-designed concrete block homes, using pre-
cast sixteen-inch square blocks for both enclosure and structure. The house
has experienced significant damage to its exterior from airborne pollutants and
earthquakes. The porous blocks allow water penetration, rusting the steel rods woven
between the blocks. This has led to block cracking and, consequently, weakened the
system’s structural stability.
Appendices include photos and videos of the block manufacturing process and the
concrete material testing.
1
CHAPTER 1: Inv Es TIg ATIng THE FREEmAn Hous E And T Ex TIl E
BloCk dETERIoRATIon
View of the garage and front entrance of the home. Figure 1.1:
1
In 1923, Samuel and Harriet Freeman commissioned Frank Lloyd Wright to design a
modest house for them in Hollywood, California. Completed in early 1925, Wright’s
prolonged experiment with standardized concrete block units is still standing, but
in great disrepair. The research presented here encompasses three primary motives:
to explore the damage that has occurred in the blocks that constitute this innovative
construction, to investigate treatment options, and to propose a replacement block
mixture that would allow visual continuity to be restored in areas that have lost
original blocks. The ultimate objective was to create a replacement block prototype
with better performance characteristics than the existing, especially in its ability to
resist water penetration.
1 Courtesy of the USC Freeman House Archives, CD 1. Photo by Michael Arden in May 1999.
2
Considerable effort has also been invested in documenting previous research efforts
in order to comprehend the breadth of work that has been done at the Freeman
House and thus carry out this thesis, building on the collected knowledge of previous
studies. Following the understanding that Harriet Freeman wanted this to be “a
practical house for [the] work [of] students [and] teachers,”
2
the replacement block
solution recommended in this research is both performance-driven and intended to
be compatible with the historic character of the existing blocks. Newly manufactured
blocks have been evaluated for their compatibility with existing blocks in terms of
water resistance, material slump, color, texture, and compressive strength.
In this chapter, the significance of the Freeman House is explored along with a
description of Wright’s innovative textile block system. Also included is an overview
of the deterioration of the blocks, attempts at stabilizing the structure, and goals for
rehabilitating the house to a new use that retains the historic character of the property.
Additionally, the scope of work for this thesis research is established and a hypothesis
about anticipated outcomes is proposed.
1.1 Cultural and historic significance
The Freeman House is a historic and cultural resource that has fallen into significant
disrepair. The house was listed on the National Register of Historic Places in 1971.
Also, the State of California designated the four Frank Lloyd Wright-designed block
2 Harriet wrote this in a note concerning her gift of the house to the USC School of Architecture. A
reproduction of this note is available in the USC Freeman House Archives.
3
houses of Southern California as Historical Landmark No. 1011 in 1993. According
to Jeffrey Chusid, “This multiple-property designation recognized the buildings as a
single historical phenomenon with four distinct elements [Millard, Storer, Freeman,
and Ennis].”
3
Additionally, the Freeman House was designated a City of Los Angeles
Historic-Cultural Monument in November 1981.
4
Faced with the potential for
further losses to its historic character, it is essential that the importance of these
designations be heeded. The significance of this house must be realized in order
to rehabilitate it as a functional teaching tool for the School of Architecture at the
University of Southern California. Also, it is important to preserve the house as a
record of the experimentation with concrete that was ongoing in Southern California
at the turn of the century. The home derives its principal significance from three
historic and culturally valuable sources. First is its intrinsic relation to the Arts and
Crafts Movement and the ideals that Wright followed in developing the textile block.
Second is the revolutionary design by Frank Lloyd Wright during a time that others
were experimenting with concrete as well. Finally, the presence and influence of the
Freemans in the history of this house gives the structure a vital role in the cultural
atmosphere of Los Angeles. These topics are examined in the following sections.
3 CHUSID, J. M. 2004. Modernist Threads: The Life, Death and Reconstruction of Frank Lloyd Wright’s
Freeman House. Los Angeles: unpublished, p.203. This manuscript was found in the USC Freeman
House Archives in a draft format. The date given is speculative based on the text indicating he was
writing in 2004 (“At the end of 2004, the Freeman House has been stripped of most of the work of
Schindler....” p.237). His book is forthcoming but not complete prior to publication of this thesis.
4 Dates for National and Local designations were found in the application form for The Getty Grant
Program, dated March 31,2000, located in the USC Freeman House Archives. The date for the State
of California designation was found in CHUSID, J. M. 2004. Modernist Threads, p.203.
4
1.1.1 The Arts and Crafts Movement and textile block homes
In late nineteenth century England, intellectuals such as William Morris and John
Ruskin sparked the Arts and Crafts Movement through their writing and buildings.
Recognized for reinvigorating an appreciation of the hand-made object, they sought
also to restore the dignity of labor by combating the oppressiveness and inferiority
of machine-made items. Moreover, this movement focused on reestablishing “a
harmony between architect, designer and craftsman and to bring handcraftsmanship
to the production of well-designed, affordable, everyday objects.”
5

A Freeman House pattern-face textile block. Figure 1.2:
6
5 CUMMING, E. AND KAPLAN, W. 1991. The Arts and Crafts Movement. New York: Thames and
Hudson, p.6.
6 WRIGHT, E. L. 2008. Frank Lloyd Wright and the Freemans. Lecture. University of Southern
California. 20 November 2008.
5
Frank Lloyd Wright’s designs for the textile block homes are clear indications of his
support for the ideals of the Arts and Crafts Movement. He envisioned his textile
block designs for Southern California as an exploration of a way that hand-craft
and the machine could be melded into one entity. This exploration took advantage
of the best aspect of hand-craft - the quality of the product - and the best aspect
of machine-made - repetitious efficiency and affordability. The material for this
exploration was concrete block, specifically the “textile block” as Wright referred to
it. Each of his four block homes utilized individual patterns, reminiscent of textiles.
The organic patterns were formed by aluminum molds, products of the industrial
era, successfully linking hand-craft with the machine. The literal weaving of steel
reinforcing in the channels along the edges of the blocks reinforces Wright’s reference
to the system as a textile, a craft product reinvigorated during the Arts and Crafts
Movement both in England and the United States.
The original Freeman House pattern-face mold found in the garage. Figure 1.3:
7
7 Courtesy of the USC Freeman House Archives, CD1. Date unknown. This last remaining original
mold was found in a cardboard box inside a locked cabinet in the garage by Jeffrey Chusid during the
spring of 1986 as he notes on page 193 of his manuscript for Modernist Threads.
6
An important idea that Wright explored in this series of homes was the notion
that the materials came from the site and therefore the house would literally grow
up out of the ground, as if it had been naturally formed there. This is alluded to
by the organic nature of the block patterns and the texture found on the block
faces. Speculations on the imagery seen in the Freeman House pattern include the
eucalyptus trees found on the site and an abstracted plan of the house placed in its
distinctively trapezoidal, hillside site. Neither in his writing nor his lectures did
Wright reveal his intentions behind the imagery of the Freeman House pattern.
Block patterns from Wright’s block houses drawn by artist Charles Calvo. Figure 1.4:
8
8 ALOFSIN, A. 1993. Frank Lloyd Wright: The Lost Years, 1910-1922: A study of influence. Chicago:
University of Chicago Press, p.297.
7
Wright’s notion of creating an affordable, Southern California home has deep roots
in the ideal of diversity championed during the Arts and Crafts Movement, which
said that art should be regionally designed and reflect the nature of its region, in this
case, the American Southwest. Instead of creating “a homogenous style applicable to
all locations,”
9
Wright sought to craft a style that drew on the characteristics inherent
to the region. On the Freeman House construction site, workers crafted the concrete
blocks by hand inside aluminum molds. The craft of these workers is prominently
displayed as both art and structure on the exterior and interior of Wright’s block
homes. His designs celebrate the fine skill of the men who built the four block
houses. This flows quite nicely with Wright’s body of work as he is known for
celebrating a material’s natural characteristics. Wright likely acquired this penchant
for combing craft and the natural form with modern advancements during the early
years of his career working in the office of Adler and Sullivan. This office was known
for their work including the Wainwright Building in Saint Louis, Missouri, and the
T ransportation Building at the World’s Columbian Exhibition in Chicago in 1893.
These structures evoked a sense of connection to the natural world through their
biomorphic ornamentation while still embracing modern technology.
Around this same period, Wright’s theory of blending the hand-craft with the
machine was echoed by Gustav Stickley, a furniture maker in New York. He
attempted to create furniture with a hand-crafted quality but produced with factory
9 HIRSCHL & ADLER GALLERIES. 1989. From Architecture to Object. New York: Hirschl &
Adler Galleries, p.16.
8
efficiency and priced affordably. “By using factory methods to produce basic
components, and utilizing craftsmen to finish and assemble, he was able to produce
sturdy, serviceable furniture which was sold in vast quantities, and still survives.”
10

Between 1901 and 1916, Stickley also published floor plans for homes along with full
specifications for free in his magazine called The Craftsman. In doing so, he hoped to
promote his furniture business along with Arts and Crafts ideals.
11
The Gustav Stickley-designed Spindled High-Back Armchair Figure 1.5:
12
and a
Frank Lloyd Wright chair for the William R. Heath House in Buffalo, New York.
13
Expressions of the Arts and Crafts Movement in American architecture included
what is commonly referred to as the Craftsman style house. One of the best-
known examples of the Craftsman style home is the Gamble House in Pasadena,
California. Designed by Charles and Henry Greene in 1908, the house is an exquisite
10 JIROUSEK, C. 1995. Art, Design, and Visual Thinking: The Arts and Crafts Movement. Available
at http://char.txa.cornell.edu/art/decart/artcraft/artcraft.htm. [Accessed 27 January 2009].
11 CUMMING, E. AND KAPLAN, W. 1991, p.122; 141.
12 HIRSCHL & ADLER GALLERIES. 1989, p.45.
13 HIRSCHL & ADLER GALLERIES. 1989, p.90.
9
exhibition of wood joinery and stained glass windows. This was directly in line with
the thinking of the Arts and Crafts Movement, yet different from both Wright and
Stickley in that it did not consider the machine, instead emphasizing the hand-tooling
required to produce the wood joinery and stained glass for the home. While both
Wright and the Greene brothers could be called Arts and Crafts architects, Wright was
not designing in the typical Craftsman style.
The Gamble House in Pasadena, California. Figure 1.6:
14
With the Freeman House, he was instead pushing into the Modern era of
architecture, exhibiting a modular form that broke traditional rules of design.
Mitered glass at the corners of the living room foreshadowed Modernists that would
follow years later. While still celebrating the art of hand-crafted concrete blocks, the
modularity of such a system is later seen in noted Modernists such as Ludwig Mies
van der Rohe and Richard Meier, who still practices today using modularized panels.
According to the Historic Structure Report for the Freeman House, prepared by Jeffrey
14 BOSLEY, E. R. 1992. Gamble House: Greene and Greene. London: Phaidon Press Limited, p.29.
10
Chusid, the house is “part of the foundation for the explosion of modern architecture
which followed in Los Angeles”
15
and its “living room is one of Wright’s great small
spaces with two-story, mitered corner glass windows that touch the glass edge-to-
edge and signal Wright’s transition from the Arts and Crafts style to the Modern
movement.”
16

1.1.2 A notable work by Frank Lloyd Wright
The Freeman House is one of four homes in Southern California where Wright used
his innovative concrete block construction method. Leading up to the block houses,
Wright undertook a series of projects for Aline Barnsdall, including her residence,
commonly known as Hollyhock House. Completed in 1921, this design utilized cast
concrete decorative elements that resembled the hollyhock flower. Following from
these cast concrete adornments came the Alice Millard Residence in 1923, the first of
four experimental houses in which Wright used a cast concrete block to construct the
entirety of the walls. The blocks at the Millard House, also commonly known as La
Miniatura, are 15 1/2” square and cast with an interlocking shape on the back side of
the patterned and plain blocks. Only the patterned blocks are used on the exterior,
but plain blocks can be seen mixed with the patterned blocks on the interior. There
are also some areas where a foundation wall is stuccoed concrete rather than using the
15 CHUSID, J. M. 1989. Historic Structure Report: Samuel and Harriet Freeman House, Hollywood,
California, Frank Lloyd Wright, 1924. Los Angeles: School of Architecture, University of Southern
California, p.9.
16 CHUSID, J. M. 1989, p.12.
11
Alice Millard Residence details of the concrete block assembly. Figure 1.7:
17
concrete block elements. The blocks are held together using expanded metal mesh
and a conventional mortar bed.
18
Wright included a sketch of this system in his
patent drawing along with drawings for the well-documented textile block system.
17 SWEENEY, R. L. 1994, p.21.
18 SWEENEY, R. L. 1994, p.20.
12
Millard House interlocking blocks exposed during restoration work. Figure 1.8:
19
Wright described his reasoning for the system of block construction as follows:
This type is made from the gravel of decayed granite of the hills easily
obtained there and mixed with cement and sand in molds or forms to make
a fairly solid mass either used in small units or monolithic construction, or in
combination. This is the beginning of a constructive effort to produce a type
that would fully utilize standardization and the repetition of appropriate units.
This standardization and repetition are essential values in the service rendered
by the Machine. They should be employed as elements in any architecture
modeled by the “third dimension”. I am still engaged in this effort to produce
an integral Architecture suited to the climatic needs of California.
20
It is clear from this excerpt that Wright intended to celebrate both nature and the
machine in built form. In fulfillment of this goal, Wright sought to experiment
further with a structural idea that followed from the Millard House’s block design.
The setting for this next experiment was to occur on Olive Hill, with the sponsorship
of Aline Barnsdall. Intended as a private school building that emphasized artistic
expression through acting, the Community Playhouse design utilized 16” square
19 SWEENEY, R. L. 1994, p.22.
20 SMITH, K. 1992, p.166.
13
Frank Lloyd Wright’s 1923 patent drawing. Figure 1.9:
21
21 SWEENEY, R. L. 1994, p.44.
14
blocks following Wright’s new block wall system (Figure 1.9) that used interlocking
steel rods inside channels between the blocks, which were filled with a wet grout
mix. Wright also included stepped block columns that are not found in any of the
other block home designs. This building, also commonly referred to as The Little
Dipper, secured a building permit on November 7, 1923, and was contracted to be
constructed at a cost of $12,500.
22
Unfortunately, after only 226 blocks had been
set in place, a building inspector called for changes that Barnsdall refused to pay for,
permanently ending construction of what would have been Wright’s first textile block
building.
23
The three subsequent block houses designed by Wright are known as “textile block”
homes because the 16” square blocks have channels on all four edges through which
steel reinforcing rods tie them together, similar to the weaving of a textile. The
first house to be successfully built using this system was the John Storer Residence,
constructed in 1923. The Freeman House followed, beginning construction in
1924. Soon after the Freeman House was finished, the Charles Ennis Residence was
completed in 1925. Both the Ennis and Storer residences encompass a much larger
square footage than the Freeman’s home, but the Freeman house exhibits Wright’s
textile block intentions most clearly. All the block homes were constructed nearly
simultaneously, allowing one project to learn from the lessons of another. Wright’s
lack of oversight during much of the construction and unresponsiveness to his son’s
22 SMITH, K. 1992, p.186.
23 SMITH, K. 1992, p.187.
15
telegrams forced Lloyd Wright to make many decisions on-site. Frank Lloyd Wright’s
failure to respond to problems encountered at the job site can be attributed to the
serious deterioration these houses have suffered. While the four block homes used a
similar construction method, they are readily distinguished by their block patterns.
Aerial isometric view of the Freeman house. Figure 1.10:
24
At the Freeman House, many noted architects, most of whom were former
apprentices of Frank Lloyd Wright, have worked on repairs and renovations for the
Freemans. These men were familiar with the textile block construction system and
include such names as Frank Lloyd Wright, Jr. (commonly referred to as Lloyd Wright
to distinguish him from his father); Rudolph Schindler; Robert Clark; John Lautner;
and Eric Lloyd Wright. This impressive list of contributing architects is another of the
many factors that make this house worthy of preserving. From the furniture added
by Rudolph Schindler to the aluminum window muntins installed by John Lautner,
24 LENTZ, J. B. 1969. Aerial Isometric From Southeast: Sheet 2 of 7. Historic American Building
Survey No. CA-1989. U.S. Department of the Interior, National Park Service. Washington,
D.C.: Library of Congress. Available at http://memory.loc.gov/cgi-bin/displayPhoto.pl?path=/
pnp/habshaer/ca/ca0200/ca0228/sheet&topImages=00002a.gif&topLinks=00002r.tif,00002a.
tif&title=&displayProfile=0. [Accessed 6 February 2009].
16
the mark of many great architects has been left on the Freeman House, in most cases
adding to the significance of this architectural treasure envisioned by Wright.
Wright further conceived of this house as a prototype small home, which would lead
to the Usonian house that became a focus of his later work. While the Freeman
House is dwarfed by the other textile block homes at approximately 2,000 square
feet of interior space, Chusid contends that this is one of Wright’s seminal works
because of its refinement, in part due to the limited budget that the Freemans offered
Wright.
25
This deserves some skepticism as the budget of $9,100 was overspent by
$13,900 to a total cost of $23,000 by the time the Freemans occupied the home in
March of 1925. Nonetheless, Wright’s design ingeniously nestles this multi-level
home on a hillside that was likely considered unsuitable for development by other
architects and his system of construction helped to define a new age of architecture
that would be cost effective, structurally efficient, and climatically responsive, leading
to the Modern architecture movement in America. As Wright described in his article
for Architectural Record:
I finally had found a simple mechanical means to produce a complete
cladding that looks the way the machine made it, as much at least as any
fabric need look. Tough, light, but not “thin,” imperishable, plastic; no
unnecessary lie about it anywhere and yet machine made, mechanically perfect
Standardization as the soul of the machine here for the first time may be seen
in the hand of the architect, put squarely up to imagination, the limitations of
imagination the only limitation of building.
26
25 CHUSID, J. M. 1989, p.9.
26 KAUFMANN, E. AND RAEBURN, B. eds. 1965. Frank Lloyd Wright: Writings and Buildings.
New York: Meridian Books, p.225.
17
Freeman House living room with view down Highland Avenue. Figure 1.11:
27
1.1.3 A gathering place for friends of the Freemans
The Freemans utilized their home as a gathering place and sometimes as a hotel for
their friends. Chusid describes the home as the site for salons that made significant
contributions to the history of the avant-garde in Los Angeles by involving a variety
of artists, architects, actors, scientists and political figures.
28
Some of their numerous
guests included Albert and Esther Dekker, who lived in the downstairs apartment;
Rudolph Schindler, who would later make new furniture at Harriet’s request and
make alterations to their living arrangements; Lester Horton, founder of the Lester
Horton Dance T roupe; Bella Lewitsky, a dancer and choreographer; Xavier Cugat,
a Latin band leader who was gaining in popularity in 1928 in performances at the
27 Courtesy of the USC Freeman House Archives, CD1. Photo by Julius Shulman, date unknown.
28 CHUSID, J. M. 1989, p.13.
18
Coconut Grove (The Freemans were so fond of their friendship with Cugat that they
named their fox terrier Rhumba.);
29
Helen Walker, a film actress in the 1930s and
early 1940s; Edward Weston, a noted photographer; Fritz Zwicky, a scientist; Rudi
Gernreich, a fashion designer; Jean Negulesco, a film director; and Wynn Ritchie
Evans, a dancer.
30

Harriet was an avid dancer and involved herself a great deal with the Lester Horton
Dance T roupe, where she met dancer and choreographer Bella Lewitzky and her
husband Newell Reynolds. This couple would later have their wedding reception in
the Freeman’s living room (Figure 1.11) in 1940. Harriet also used the living room
as a dance studio for teaching her weekly dance exercise class. She was proud of the
living room, so she wanted to show it off to her students.
31
Also, Wynn Evans, whom
she met through contacts at the Lester Horton Dance T roupe, lived on the Freeman’s
couch for about a year starting in 1928.
32
Evans, along with the other house guests
(the Dekkers, Cugat, and Walker), experienced a house that was an exciting place to
live with a great deal of unique aspects that made it different from any other house
they had ever been inside. Most disruptive for the guests were the leaking walls and
the drafty doors, but the Freemans seemed to ignore these things.
33
29 CHUSID, J. M. 2004, p.161.
30 CHUSID, J. M. 1989, p.19.
31 CHUSID, J. M. 1989, p.22.
32 CHUSID, J. M. 1989, p.24.
33 CHUSID, J. M. 1989, p.24.
19
1.2 Harriet Freeman’s gift and instructions for use
Sam Freeman passed away in 1981 inside the house. Even though the two basically
had separate lives, involved themselves in extramarital affairs, and lived in rooms
of their own, neither wanted to lose possession of the house.
34
Following Sam’s
death, Harriet Freeman made arrangements in 1983 for her home to be donated to
the School of Architecture at the University of Southern California for its use and
continued upkeep. When Harriet died inside the house in 1986, the house came
under the ownership of the School of Architecture.
35
In a handwritten note, she gave
the following instructions:
1. No connection or similarity to Gambell[sic] H
2. Not to be thot[sic] of in the same breath
3. Compare silver & platinum.
4. Not a complement to Gambell[sic] House.
5. Don’t even think of the 2 together.
6. Think of our house as a single gift.
7. A practical house for work - students - teachers etc.
8. All the house needs is what is necessary to be made efficient not a rebuilt
house.
9. Not large house like Gambel[sic] or gigantic Ennis house just a private
house of 4 rooms which Mr. Wright said would cost us between 8 &
$10,000. If it would cost us more he would be responsible.
36
It is with Mrs. Freeman’s intentions in mind that this thesis work has been conducted,
using the house as a laboratory for research in issues of material science and historic
34 CHUSID, J. M. 1989, p.9.
35 CHUSID, J. M. 1989, p.9.
36 See Appendix B for a reproduction of Harriet’s note, courtesy of the USC Freeman House
Archives.
20
preservation. Since Harriet specifically stated that the house not become a museum,
the preservationist’s approach to the house has been only considered to a degree,
allowing the viability of the structure and an interest in construction techniques to
come to the forefront. An attempt has been made, however, to be compatible with
the historic fabric of the house as much as is possible. With the best intentions for
the future of the house in mind and following Mrs. Freeman’s wishes for her former
home, this thesis explores the options for repairing and replacing the concrete block
walls of the house.
1.3 The textile block system
Frank Lloyd Wright developed this revolutionary structural construction concept as a
way for homeowners to construct their own residence without any major involvement
from the traditional construction trades. The reality of the system is that it took
more skill and effort to create perfect blocks than would be necessary with traditional
construction materials and methods. Not only did this system cost more than double
what Wright had estimated, but its structural integrity was severely undermined by
the factors explored in this section.
1.3.1 Function of the construction system
A textile block is the name given to the prefabricated, sixteen inch square, concrete
blocks designed by Frank Lloyd Wright, which have channels on all four sides
allowing steel reinforcing rods to be woven through to form the walls of the house.
21
Diagram of a typical Freeman House textile block wall. Figure 1.12:
37
Grout was then poured in around the steel reinforcing to create a flexible wall that
would, in theory, move with any seismic activity. This system was different than
traditional masonry in that no grout lines are exposed on the finish face of the wall.
As mentioned previously, this is the construction method used in three of the four
Southern California block homes. The Millard House is not a true textile block
37 Author diagram.
22
construction, but its blocks were cast using the same method, possibly with redwood
molds instead of aluminum.
Detail of a construction photo showing exposed steel reinforcing rod Figure 1.13:
protruding from blocks.
38
The mixture of the concrete for all the block homes is very dry in comparison to a
typical concrete mixture. In the specifications that Wright gave the contractor, he
prescribed a mixture of one part Portland cement to four parts clean sand to be mixed
in a mechanical mixer and water added to the point that the mix would stand up
when squeezed by hand.
39
These homes were intended to be affordable and would be
made by the homeowner-cum-builder on-site. In reality, many construction workers
were needed to produce the approximately 11,000 blocks that make up the walls of
the house. The method for making blocks, while straightforward, is time consuming
38 WRIGHT, E. L. 2008.
39 CHUSID, J. M. 1989, p.81.
23
and labor intensive. Great skill is involved in the critical compaction process that
imprints the block with the “textile” pattern designed by Wright.
In the Freeman House, there are three main versions of block: a pattern-face block
(with left and right hand versions), a pattern-face block with perforations to permit
light through, and a flat-face block. In order to properly construct the house,
however, there were many variations made in order to form corners, chamfers, and
other difficult details not able to be accomplished with full-size blocks. At the
Freeman House, Jeffrey Chusid has documented 74 distinct variations in block
types, including some that were cut after casting to fit special conditions.
40
This
straightforward, simple system of blocks suddenly became much more complicated.
The three main types of block found at the house: left and right hand Figure 1.14:
pattern-face
41
, perforated pattern-face
42
, and flat-face.
43
1.3.2 Materials
The materials used in the block mix are sand, Portland cement, and water. Based
on a previous study, the proportions of these materials were determined to be
40 CHUSID, J. M. 1989, p.83.
41 Courtesy of the USC Freeman House Archives, CD3. Photo dated 18 May 2001.
42 Author photo.
43 Courtesy of the USC Freeman House Archives. Photo dated January 1995.
24
approximately one part cement to two parts sand and aggregate. Previous attempts
to make replacement blocks at USC have used this mixture formula. However, this is
different from the mixture that Wright had specified in the construction documents.
All four of the block houses seem to have used different materials in their mixes. The
Freeman House did not use on-site materials. Instead, the contractor ordered sand
and cement from his outside suppliers. The Ennis House is believed to be the only
one that actually used decomposed granite from its site. This ultimately led to the
premature deterioration of the blocks and the need to replace entire walls of the house
with newly manufactured blocks.
44

Materials piled up on Glencoe Way during construction. Figure 1.15:
45
During the restoration of the Ennis House in 2008, the mix used for new blocks was
developed by Moonlight Molds in Gardena, California. This mixture utilized several
44 WRIGHT, E. L. 2008.
45 WRIGHT, E. L. 2008.
25
different materials in the following proportions: 17.92% white cement, 22.03%
birdeye gravel, 22.03% Gillibrand washed concrete mix
46
, 17.92% Unamin #8,
8.96% bunker sand, and 11.15% water. Added to this mixture were two different
coloring dyes that produced the coloration determined to best match the existing
blocks of the Ennis House. The mix was poured into fiberglass molds formed on a
block made with the original aluminum mold. The new blocks were allowed to cure
inside these fiberglass molds, instead of removing them immediately as had been done
in the original construction.
47

1.4 Block deterioration
1.4.1 System failure points
The 1994 Northridge earthquake was a jarring event across the Los Angeles region.
Since that time, experts have learned a great deal about inadequate construction
methods and faulty assemblies that were once believed to be safe in case of a
seismic event. The Freeman House is no exception. Many aspects of the textile
block construction method have been scrutinized as being insufficient for envelope
enclosure, seismic stability, and durability of the facade. This is not to say that the
problems were not evident before the Northridge earthquake, but they were even
46 This is a coarse sand with all the fines washed out. It comes from Gillibrand’s Simi Valley,
California, pit. Information obtained from an e-mail conversation with Jeff Keenan, President of
Moonlight Molds, Incorporated. 23 February 2009.
47 Information obtained during a phone conversation with Jeff Keenan, President of Moonlight
Molds, Incorporated. 17 November 2008.
26
Measures to reinforce the walls after the Northridge earthquake. Figure 1.16:
48
Block cracking and spalling on a south porch column. Figure 1.17:
49
48 Courtesy of the USC Freeman House Archives, CD 3. Photo dated 22 November 2000.
49 Courtesy of the USC Freeman House Archives, CD4. Photo dated 14 September 2000.
27
more obvious after its occurrence. The deficiencies stem from Wright’s imperfect
design and from errors and omissions made during construction. One of the
main problems is that the blocks themselves lacked control in their production.
Throughout the course of the build, materials were obtained from various
commercially available sources, depending on the contractor. Different crews were
brought in to make the blocks and produced a range of colorations and textures from
the bottom to the top of the house. The pace of construction was quite fast and
because of this, the blocks varied considerably in their composition and durability.
50

It is also speculated that earth was added to the mixture in some cases to produce
a buff color. This would account for the brittle nature of the block face.
51
Many
of the blocks cracked or eroded as a result of their composition and in other places
efflorescence has been a major issue.
Efflorescence on the lower floor interior hallway. Figure 1.18:
52
50 CHUSID, J. M. 1989, p.91.
51 CHUSID, J. M. 1989, p.224.
52 Courtesy of the USC Freeman House Archives, CD3. Photo dated 3 October 2000.
28
The 1/4” steel reinforcing rods placed between the blocks were, in some instances, not
fully encased by the grout poured between the blocks since the grout had difficulty
flowing in the narrow channel and reaching all the way to the bottom. The exposed
portions of steel rod have been in contact with water either by not being fully encased
or through cracking of the grouting material, which has rusted the steel as a result
of this exposure. This condition was first discovered during restoration work at the
Storer House and was then seen more fully at the Freeman House following the 1994
earthquake, which forcibly exposed this problematic wall construction detail.
53
Exposed steel tie connecting the two wythes of the wall near a window. Figure 1.19:
54
\
Another place where rust and water transfer occurred was between the two wythes of
block that make up the walls. Designed by Wright to provide an insulating air space
between interior and exterior, the double wythe was supposed to be tied together at
every intersection of the reinforcing rods (Figure 1.19) and then that tie was supposed
53 CHUSID, J. M. 1989, p.91.
54 Author photo.
29
to be covered in two inches of concrete. The problem with this arose because the
blocks were cast as perfect squares, meaning the corners would have to be chipped
away to allow access to the steel rods. Possibly due to the difficulty of constructing
this detail, it appears that the encasement of concrete around the ties has either worn
away or was never there to begin with. Unfortunately, this allows any water that
enters the outside wall to transfer through to the inside wall and the interior of the
home by way of these ties, further rusting the reinforcing rods, cracking the grout,
and damaging the blocks.
1.4.2 Previous retrofitting work
After the Northridge earthquake in 1994, the Freeman house was in danger of
collapsing and needed critical repairs. With financial support from the Federal
Emergency Management Agency, the J. Paul Getty T rust, and others, significant
retrofitting was undertaken in 2000 and 2001 to stabilize the home and prevent
further damage. All work proposed in the Getty Grant application was to abide by
the Secretary of the Interior’s Standards for Rehabilitation.
55
The grant was awarded
on October 31, 2000 in the amount of $75,000.
56
As a key part of the repair work,
concrete piers were poured to bedrock on the street frontage to prevent the hillside
from pushing the house down. The textile block retaining walls along the street were
55 Response to Question 9 of the Getty Grant Program Proposal dated 31 March 2000, submitted
by Robert H. Timme, Dean of the School of Architecture. Courtesy of the USC Freeman House
Archives.
56 Correspondence from Deborah Marrow, Director of the Getty Grant Program, to Robert Timme,
Dean, dated 31 October 2000.
30
insufficient to support such a great load. The foundations were also reinforced in
several areas. The living room was strengthened with concrete columns on either side
of the fireplace (Figure 1.21) and a matching pair opposite the fireplace on the south
wall. These concrete columns have yet to receive any veneer of textile block tile.
Installing support piers along Glencoe Way. Figure 1.20:
57
New concrete columns on either side of the living room fireplace. Figure 1.21:
58
57 Courtesy of the USC Freeman House Archives, CD4. Photo dated 21 November 2000.
58 Courtesy of the USC Freeman House Archives, CD3. Photo dated 8 June 2001.
31
T wo steel Vierendeel girders (Figure 1.22) were installed to support the clerestory glass
window above the living room. The girders were then encased in restored clerestory
windows. The original sashes had decayed significantly and had to receive extensive
repairs. On top of these steel Vierendeel girders, wood beams and plywood were built
up to create a new roof with a slight pitch, allowing water to drain off. The original
roof was designed to be flat, which allowed water to pool on the roof and ultimately
led to water infiltration problems.
Vierendeel girder installed at the living room clerestory. Figure 1.22:
59
The weakened south porches were allowed to collapse and were replaced with cast-in-
place concrete walls (Figures 1.23 and 1.24), which were intended to be covered in
tiles to mimic the original appearance of the house. This veneer step, which would
restore visual continuity of the block facade, has yet to be completed.
The roof was recovered and sheet metal caps were applied to portions of the parapet
where water was penetrating through significantly deteriorated blocks. While it
59 Courtesy of the USC Freeman House Archives, CD3. Photo dated 1 June 2001.
32
can be said that this seismic retrofitting work has degraded the historic character
of the home and bypassed the structural function of the textile block system, it has
ultimately preserved the house from being entirely lost.
Living room balcony reconstruction using cast-in-place concrete walls. Figure 1.23:
60
South sleeping porch reconstruction using cast-in-place concrete walls. Figure 1.24:
(Note the vertical channels on the face intended to hold newly manufactured tiles.)
61
60 Author photo.
61 Courtesy of the USC Freeman House Archives, CD4. Photo dated 17 April 2001.
33
1.4.3 Looking forward
There is a strong need for the rehabilitation of the house as it continues to deteriorate.
Deferred maintenance and the persistence of water infiltration are ongoing problems
that put the historic textile blocks at extreme risk. Even with the retrofitting work
that has been completed, the original blocks are still in danger and there is a great
deal of work that still needs to be done. The most pressing issue for this investigation
and the continued viability of the home is the problem of water infiltration through
the blocks. If water is not prevented from entering the face of the block wall, the
home will continue to decay. Additionally, any recommended treatment will require
continual maintenance no matter what solutions are implemented to preserve or
replace the textile blocks. At the Storer House, previous owner Joel Silver ensured
that the blocks of his home were sealed every five to seven years to maintain water
resistance.
Ultimately, blocks will need to be replaced if the house is to survive and blocks will
continue to decay without any treatment of the surface exposed to weathering. An
informed decision will have to be made about what the best approach is for replacing
blocks or preserving the existing blocks in-situ. Due to the interwoven steel rods
between the blocks, it will be quite difficult to replace blocks in the center of a wall
without dismantling the entire wall. There are three options here: a new type of
block could be designed to fit around the existing reinforcement; the block face could
be removed and a veneer of the pattern or flat-face applied; or the entire wall could
34
be catalogued, dismantled and a newly manufactured block wall could be put in its
place. One of the first two options, or both, seem to be most likely the solution.
The proposal for block replacement in this research gives consideration to many
variables including historic accuracy in material selection, color and texture
consistency with the existing blocks, durability of the replacement options, and
manufacturing techniques for replacements. There is not one solution that is the
correct or comprehensive answer. There will be many different solutions that fit the
various problems encountered throughout the house. The most appropriate approach
can only be determined when the cost effectiveness of the solutions are weighed with
consideration for the intended use of the house in the future and human safety inside
the structure during future seismic activity.
Jeffrey Chusid, who was Director of the Freeman House and lived there for many
years while teaching in the USC School of Architecture, holds a similar opinion:
Almost since the project began in 1987, there has been a sense that the house
would be better served if projects were considered as a legible combination of
preservation, rehabilitation, restoration and reconstruction. Such an approach
could offer an extraordinary pedagogical opportunity, demonstrating the
differences between approaches as well as showcasing their results. Second,
the house has a peculiar history of continuous intervention by important
architects, which makes it not just difficult to identify a particular date as
a restoration objective, but even misleading and uninformative to do so.
Taking different approaches in different areas of the house could allow several
narratives to be interpreted at the same time: the original design of Frank
Lloyd Wright, the dense, layered additions by Rudolph Schindler and others;
even the rich social tapestry of a house filled with art and artists. Such an
approach could be used in the conservation of the historic fabric as well.
One part of the house could remain as an intact example of the textile-block
35
system, while another demonstrates contemporary experiments with new
block, and still a third showcases the various structural strategies required to
bring the system up to code.
62
1.4.4 Possible reuse schemes
In this research work, Harriet Freeman’s desire to use the house as a laboratory for the
work of students and teachers was the driving force. But after the house is repaired,
the question remains as to what purpose the house should serve. There is an ongoing
discussion as to whether the house should be restored to a period significance or if it
should be rehabilitated to a new use . One scheme that has been suggested by Chusid
and others is to use the house as a residence for visiting scholars or exemplary students
as well as a reception hall for special events and fund-raising. This is the model used
at the Gamble House, also owned by the USC School of Architecture. However,
Harriet would object to the house becoming a museum like the Gamble House. In
any scheme adopted in the future, it would make sense to keep a resident in the
house as a caretaker who could monitor the condition of the blocks and ensure that
maintenance is carried out on a regular basis.
1.5 scope of work
This study focused primarily on the concrete textile blocks that make up the walls of
the Freeman House. While there are issues that exist with the steel reinforcing rods
that tie the blocks together, this research effort did not consider these issues within its
62 CHUSID, J. M. 2004, p.237.
36
scope. Blocks were evaluated for their composition and for the qualities that make
them identifiable as part of the Freeman House. These factors dictated how the
research work was carried out and what recommendations were made for replacement
blocks.
The most appropriate replacement block was recommended in this study by
considering the Secretary of the Interior’s Standards for the Treatment of Historic
Properties with Guidelines for Preserving, Rehabilitating, Restoring & Reconstructing
Historic Buildings along with a strong interest in the continued survival of the home
as a laboratory for students and teachers. The determination of a specific replacement
block solution necessarily eliminated some of the considerations about historical
accuracy that have been expressed in the past and will not be considered “ideal” by all
people. The focus for this study was long-term durability and consistency with the
appearance and texture of the existing blocks at the house.
In the course of the research, sample blocks were made and tested through standard
concrete experimentation methods. The four types of tests completed were a slump
test, the Karsten pipe water penetration test, a compressive strength test, and a color
and texture comparison. Results from this testing led toward the recommendation
of a mixture and manufacturing technique for the replacement blocks. The final
product realized by this thesis was the concrete mixture and detailed procedure used
for creating this block mixture. Recommendations were also made as to the proper
use of this replacement block.
37
1.5.1 Hypothesis statement
The intent of Wright in 1923 to design a standardized textile block for his home
designs was flawed in his inability to maintain consistency between the mixtures and
create a watertight enclosure for the home. Therefore, if a replacement block is made
to repair the existing walls, it will need to utilize modern-day mixture materials and
techniques to create a new block that is both compatible with the existing structure
and superior in terms of its water resistance and durability.
1.6 Chapter overview
Chapter two focuses on the history of the Freeman House and the couple that called
it home for many decades. Included in this history is a description of the explorations
in concrete architecture in Southern California in the early 1900s and a brief synopsis
of the work of the many architects who have carried out renovations at the house
over the past 85 years. Also, this chapter looks at previous studies of the house and
material composition investigations.
Chapter three provides an overview of conservation standards both domestic and
international, a catalogue of typical block deterioration found at the Freeman
House, and techniques for repairing concrete block and masonry in keeping with the
Secretary of the Interior’s Standards.
38
Chapter four provides a detailed methodology for the block manufacturing process
and the testing techniques used in this thesis research, along with an explanation of
data presentation.
Chapter five details the process of manufacturing textile blocks and tiles for this study
and provides charts describing the block mixtures that were developed and what
admixtures, sealants, and manufacturing techniques were utilized in an attempt to
prevent or minimize water penetration through the blocks.
Chapter six describes the results from all the testing completed on the sample
replacement blocks. Charts, diagrams and photos are provided to better describe the
testing procedures implemented. Blocks made with the existing mixture and with
various new mixtures were tested to make a comparison between what currently exists
on-site and what might function as an appropriate replacement.
Chapter seven provides a thorough analysis and evaluation of the testing results
obtained during the block manufacturing and testing phases. Special consideration
was given to the effect that replacement blocks will have on the historic character
of the structure and their intended use for the home. Also discussed is the issue of
replacement block compatibility with the existing block and what a compatible choice
would be in terms of water content, porosity, and color matching.
Chapter eight offers a series of conclusions about the best method for manufacturing
a replacement block, the selection of an appropriate replacement block mixture, and
39
the ramifications of such selections for the future of the Freeman House. In many
ways, more questions have been found than answers, however, the development of
a replacement block prototype is a major step in the process to rehabilitating the
Freeman House.
Chapter nine explores topics that would naturally follow from and complement this
thesis research but are not covered within its scope. These issues are related to the
textile blocks and replacement or repair options and in some cases inextricably linked
to the survival of the house. Future research must address these topics in the process
of working towards a successful rehabilitation of the Freeman House.
40
CHAPTER 2: HISTORY OF THE FREEMAN HOUSE
2.1 Early 20th century explorations in concrete construction
At the beginning of the 20th century, there was a prevailing uncertainty among
architects about what the predominant style of the new century would look like. For
architects Irving Gill, Rudolph Schindler, Frank Lloyd Wright, and Lloyd Wright,
the style of 20th century architecture would be tied to advances in new materials,
namely concrete. Concrete was reintroduced to the building industry in the late
18th century in Europe. Previously, the knowledge had either been lost or forgotten
since the Romans used concrete in their prolific construction campaigns. Around
1840,
1
concrete had begun to be used in the United States for industrial buildings
and engineering projects like dams, canals, and bridges.
2
Gill, Schindler, and both
Wrights sought to experiment with this new material in innovative ways, hoping to
define the architecture of a new generation.
For Gill, who previously worked with Frank Lloyd Wright in the Chicago office of
Louis Sullivan, explorations were with poured-in-place buildings and systems between
1910 and 1919. One of his best known projects done in concrete was Horatio West
Court Apartments in 1919, located in Santa Monica, California. In 1921, Schindler
worked in concrete using pre-cast tilt-up slabs for the Schindler-Chase House and
1 SWEENEY, R. L. 1994. Wright in Hollywood: Visions of a new architecture. Cambridge, MA and
London: The MIT Press, p.216.
2 CHUSID, J. M. 2004. Modernist Threads: The Life, Death and Reconstruction of Frank Lloyd Wright’s
Freeman House. Los Angeles: unpublished, p.100.
41
One of six apartment units at Horatio West Court. Figure 2.1:
3
also used concrete at Pueblo Ribera Courts in 1924 and the Lovell Beach House
in 1926.
4
Other architects at the same time were utilizing concrete for homes and
businesses because of its ability to be formed in any shape and, most importantly, for
its fire resistance.
New systems that were constructed with standardized blocks had also been developed.
A former employee of Frank Lloyd Wright, Walter Burley Griffin, had designed a
system called Knitlock in 1916, which was based on a six inch module and comprised
two block components. One had vertical ribs that became structural support and the
other served as infill panels between the ribs. The L shape of the ribbed, or “vertebral”
segments allowed them to easily turn a corner. Metal reinforcing was placed in
between the vertical voids and this was filled with grout. Despite the similarities to
the later textile blocks, Griffin never accused Wright of stealing the Knitlock design.
5
3 Author photo.
4 CHUSID, J. M. 2004, p.101.
5 SWEENEY, R. L. 1994, p.207-208.
42
Walter Burley Griffin’s Knitlock construction system. Figure 2.2:
6
A Nel-Stone advertisement and wall construction diagram. Figure 2.3:
7
6 SWEENEY, R. L. 1994, p.207.
7 SWEENEY, R. L. 1994, p.209-210. Notice that the corners are finished with poured-in-place
concrete which was different from Wright’s textile block system that made blocks for the corners..
43
Another system of block construction being marketed in the early 1900s was from the
Nel-Stone Company, founded by inventor William E. Nelson. The mortarless block
wall system he designed (Figure 2.3) created circular channels between the blocks
when they were stacked edge-to-edge. In this channel, reinforcing bars and concrete
were filled in to tie the wall together and give it strength. This system, like Griffin’s
Knitlock, was introduced several years prior to Wright’s first concrete block designs.
Unlike the Wright’s textile block system, the Nel-Stone wall is a single vertical layer of
blocks that are twelve inches square instead of the double wythe wall of sixteen inch
blocks that Wright used. One of the franchise plants that Nelson opened was located
in Los Angeles, however, it is unclear to what extent Wright was aware of this system
when he was designing the Millard House in 1923.
8
In a more architecturally aesthetic sense, Lloyd Wright (LW) was exploring the use
of concrete blocks in his own practice. In his 1922 Henry Bollman House, built
in Hollywood, California, LW introduced “the ‘knit-block’ system in which a small
airspace separated double walls of four-inch-thick blocks that were tied together by
steel rods.”
9
The extent to which this design influenced his father’s four block houses
is not wholly clear. Sweeney writes that, “. . . Lloyd’s role as a catalyst for the textile
block system remains unconfirmed . . . . Lloyd’s description of his ‘core system’
8 SWEENEY, R. L. 1994, p.209-211. Sweeney offers a more in depth discussion of these
technologies and other influences Wright may have had when designing his textile block system.
9 HINES, T. S. 1996. The Blessing and the Curse: The Achievement of Lloyd Wright. In:
WEINTRAUB, A. 1998. Lloyd Wright: The Architecture of Frank Lloyd Wright, Jr. New York: Harry
N. Abrams, p.19.
44
The Henry Bollman House exterior and view of the block columns. Figure 2.4:
10
suggests circular channels either cast or drilled through the blocks . . . .”
11
Hines
asserts that the senior Wright was inspired to adopt this system and employed his son
to oversee construction on the Storer House. The fundamental difference, however,
was that “the knit-block system was the chief structural element of the Storer house,
[whereas] in the Bollman house it was used mainly to accent a more conventionally
constructed building of wood-framed stucco.”
12
While LW did take on the role of
supervising his father’s block houses while in construction, he went on with his own
exploration of concrete block and especially perfected the use of it as a decorative
element. Hines lauded this development in Lloyd Wright’s work when he wrote:
After the Taggart and Bollman houses, LW’s Deco-Expressionist penchant
was best expressed in four other Los Angeles-area houses of the mid-1920s,
10 WEINTRAUB, A. 1998. Lloyd Wright: The Architecture of Frank Lloyd Wright, Jr. New York:
Harry N. Abrams, p.55; 57.
11 SWEENEY, R. L. 1994, p.205.
12 HINES, T. S. 1996, p.21.
45
which collectively represented the pinnacle of his life’s work: the Sowden
house (1926), the Samuel-Novarro house (1928), the Derby house (1926),
and his own studio-house (1927). These buildings epitomized his talent for
merging his own brand of Expressionism, akin to contemporary European
developments, with his and his father’s interest in Southwest Indian cultures as
expressed in modern materials, particularly reinforced concrete.”
Lloyd Wright’s Sowden, Derby, and Samuel-Novarro houses. Figure 2.5:
13
Also in LW’s work with block construction, he experimented with different block
thicknesses and cement contents, and he recommended the use of a waterproof
membrane to seal the blocks.
14
He even went as far as sending his father a proposal
for a revised block wall system in 1931 (Figure 2.6). While his father’s response noted
some merits in this system, he did not wholeheartedly support his son, denouncing
the use of a mortar bed between the blocks.
Clearly, Frank Lloyd Wright was not the first architect to explore concrete
construction in the early 1900s and his system may have even drawn on the
innovations of other systems available. Where other systems had emphasized the
13 WEINTRAUB, A. 1998, p.80; 90; 116.
14 SWEENEY, R. L. 1994, p.205-206.
46
Lloyd Wright’s design for an improved concrete block wall. Figure 2.6:
15

economic benefit of rapid construction with pre-cast elements, which could be made
to look like nearly any style, both Wright and his son instead made use of concrete in
an artistic way, shaping the block with patterns that celebrated the building’s method
of construction. In this way, their system departed from the others, drawing upon
their affinity for cultural ornament and design seen in Mayan and Japanese buildings.
15 SWEENEY, R. L. 1994, p.206.
47
2.2 Wright’s period of transition in Southern California
Frank Lloyd Wright worked for approximately six years in Japan on the Imperial
Hotel, traveling back and forth to Chicago between trips. Lacking work in Chicago,
Wright came to Southern California on the recommendation of his son, who had
been working at that time in San Diego for Irving Gill.
16
Wright decided to establish
an office in Los Angeles in 1920 and proceeded to secure projects that would allow
him to explore a new direction in his architectural career, influenced by his time in
Japan and his fondness for Pre-Columbian architecture, primarily that of the Mayan
civilization. Though he would never write or speak about the connections and sources
of his Southern California work, the imagery clearly alludes to Mayan influences.
Frank Lloyd Wright (FLW) began a project for a series of buildings on Olive Hill
in 1916 for oil heiress Aline Barnsdall. The program included her residence (later
to be known as Hollyhock House), a performance theater, a movie theater, and a
housing development with studios and shops.
17
While FLW was engrossed in his
work on Olive Hill, Lloyd Wright was completing a residence for Henry Bollman
in Hollywood.
18
Hollyhock House was to become a major stepping stone to the
concrete block houses in its massing. The canted walls of the roof have unmistakable
connections to the designs of Mayan architecture. Wright began experimenting with
16 HINES, T. S. 1996, p.14.
17 ALOFSIN, A. 1993. Frank Lloyd Wright: The Lost Years, 1910-1922: A study of influence. Chicago:
University of Chicago Press, p.233.
18 HINES, T. S. 1996, p.19.
48
the expression of a stone construction through the use of cast concrete decorative
elements that are added as ornaments to the house. This was not his first work in
concrete, however. Unity Temple, completed in 1905, was built of cast-in-place
concrete walls.
Exterior of Hollyhock House with canted walls and cast ornaments. Figure 2.7:
19
Another project that Wright had worked on for Barnsdall was the Community
Playhouse, which was to be a private, experimental school for her daughter and other
children who would pay tuition.
20
The plan of the Little Dipper, as it would come
to be commonly referred to, is “the intersection of an irregular square and circle; it
is symmetrical on two axes, though it incorporates a nonconforming wing to disturb
19 McCARTER, R. 1997. Frank Lloyd Wright. London: Phaidon Press Limited, p.127.
20 For a comprehensive look at all of Wright’s projects for Aline Barnsdall, see SMITH, K. 1992.
Frank Lloyd Wright, Hollyhock House and Olive Hill: Buildings and Projects for Aline Barnsdall. New
York: Rizzoli.
49
the balance.”
21
This building would have been Wright’s first use of the textile block
system as shown in his patent drawing of 1923. In the Little Dipper drawings,
Wright even incorporates offset block courses that are not seen in any of the other
block houses that he built. This playhouse began construction in late 1923,
22

The Little Dipper, northwest elevation and block offset details. Figure 2.8:
23
but after only 226 blocks had been laid, the city’s building inspector required changes
to the design, which Barnsdall would not agree to pay for, leaving unfinished Wright’s
first attempt at using the textile block construction system.
24
After the completion of the Barnsdall residence and two other small houses on
Barnsdall’s site (Residences A and B), father and son began working together on new
commissions. One of their projects, which was also to use the textile block system,
was a ranch design for the Doheny family, described by Thomas Hines as “. . . a
21 SWEENEY, R. L. 1994, p.45.
22 SMITH, K. 1992, p.186.
23 SWEENEY, R. L. 1994, p.48.
24 SMITH, K. 1992, p.187.
50
luxuriant proposal for the development of land . . . in the then rural mountains above
Beverly Hills.” Hines goes on to describe the project that was to influence his, and
likely his son’s, later work:
In several of the most appealing drawings of their lives, the Wrights designed
concrete-block houses of varying types and sizes, integrated into the steep,
craggy landscape with subtly positioned walls, roads, and bridges. . . . the
ideas from the Doheny project survived as built fragments in Hollywood
and Pasadena in the form of the smaller but no less exotic Mayan-inspired,
concrete-block Millard, Storer, Freeman, and Ennis houses.
25
The Doheny Ranch project would have incorporated a series of buildings that Wright
designed to be monumental in scale. This could have been one of Wright’s greatest
achievements if it had been constructed. Instead, as Hines noted, the elements of
Doheny Ranch were incorporated into his later concrete block homes.
Rendering and partial plan of Doheny Ranch, House C. Figure 2.9:
26
25 HINES, T. S. 1996, p.17-18.
26 SWEENEY, R. L. 1994, p.16.
51
Perspective drawing of Doheny Ranch project. Figure 2.10:
27
After the lost commissions for the Little Dipper and Doheny Ranch, Lloyd Wright
and his father decided to pursue their own practices. While LW would be called on
to supervise his father’s block house projects, both FLW and his son continued to
experiment with designs for other concrete block homes in their individual practices.
Lloyd Wright would stay in Los Angeles, but his father returned to the Midwest.
FLW would use an improved version of his block system in several houses. He would
also later complete several buildings on the campus of Florida Southern University
using concrete block designs along with a block design based on palm tree branches
for the Arizona Biltmore Hotel in Phoenix, completed in early 1929.
A guest cottage at the Arizona Biltmore Hotel. Figure 2.11:
28
27 SWEENEY, R. L. 1994, p.12.
28 CHEEK, L. W. 2006. Frank Lloyd Wright in Arizona. T ucson: Rio Nuevo, p.12.
52
Timeline of Wright’s Southern California cast concrete projects starting Figure 2.12:
with Hollyhock House
29
, followed by the Millard House
30
, the Playhouse
31
for Aline
Barnsdall, then the Storer House
32
, Freeman House
33
, and finally the Ennis House.
34
29 WEINTRAUB, A. AND HESS, A. 2005. Frank Lloyd Wright: The Houses. New York: Rizzoli,
p.187.
30 McCARTER, R. 1997, p.160.
31 SMITH, K. 1992, p.188.
32 WEINTRAUB, A. AND HESS, A. 2005, p.191.
33 WRIGHT, E. L. 2008. Frank Lloyd Wright and the Freemans. Lecture. University of Southern
California. 20 November 2008.
34 WEINTRAUB, A. AND HESS, A. 2005, p.207.
1924-1925
Samuel & Harriet
Freeman Residence
Hollywood, CA
16” textile block wall
1923-1924
John Storer Residence
Hollywood, CA
16” textile block wall
1919-1921
Aline Barnsdall Residence
“Hollyhock House”
Hollywood, CA
Stucco with cast stone
ornamentation
1923-1924
Alice Millard Residence
“La Miniatura”
Pasadena, CA
15.5” concrete block wall
1924-1925
Charles Ennis Residence
“Ennis-Brown House”
Los Angeles, CA
16” textile block wall
1923
Olive Hill Community
Playhouse (incomplete)
“The Little Dipper”
Hollywood, CA
16” textile block wall
1920 1926
53
2.3 Freeman House architectural history
The Freeman House holds a rich, multifaceted history that includes the extramarital
affairs of its owners, renovations by noted architects, and major retrofitting for the
structure in order to save the house from collapse. This section provides an overview
of the most important points in the history of the house, including the Freemans
themselves, and highlights important factors in the design and construction to
establish the significance of the Freeman House.
2.3.1 The family and their home
Samuel Freeman was born in New York City in 1889. He moved to California with
his parents and two sisters in 1906. His name was originally Samuel Friedman but
he had his name legally changed around 1918 when he moved to Los Angeles. Soon
after Sam had arrived in Los Angeles, Harriet Press, the youngest of five children,
also came to Southern California to settle in Los Angeles with her sister Leah. Their
brother Abe was already in Southern California. Both Harriet and her sister had
an interest in teaching. Upon Harriet’s arrival to Los Angeles, however, she took
up a career in acting and dancing, which she would carry on for most of her life.
Meanwhile, Sam was the owner of a jewelry store in downtown Los Angeles. Harriet
met Sam through Abe’s friends and they were married on April 18, 1921.
35

35 CHUSID, J. M. 1989. Historic Structure Report: Samuel and Harriet Freeman House, Hollywood,
California, Frank Lloyd Wright, 1924. Los Angeles: School of Architecture, University of Southern
California, p.16-18. Chusid provides an even more detailed family history of the Freemans in his
manuscript for Modernist Threads.
54
Harriet and Samuel Freeman. Figure 2.13:
36
The couple met Frank Lloyd Wright soon after through Harriet’s sister Leah, who
was teaching at Aline Barnsdall’s nursery school on Olive Hill. At the time that
the newlyweds met Wright, he was in the process of finishing the construction of
the residence for theatre producer and oil heiress Aline Barnsdall, not far from the
eventual site of the Freeman House. The couple was impressed with the work that
Wright had done for Mrs. Barnsdall, but they did not want a home quite as decadent
as the Barnsdall Residence. They instead asked for something more modest since they
had no children.
Wright set out to design what he proclaimed to be a prototype low-cost solution for
housing in the arid Southern California climate. While this claim may have been true
in part for the climate and clients Wright served in Southern California, it must be
36 WRIGHT, E. L. 2008.
55
noted that Wright utilized “ . . . these same forms, stylistic antecedents, and materials
. . . in Japan, the American Midwest, . . . and Arizona.”
37
The home he designed is
one of the smallest of the three textile block homes at approximately 2,000 square
feet and sits on a site of approximately 70 feet by 75 feet. Wright made the most of
this site by capitalizing on the benefits of a steep hill with a 25 to 30 percent grade
down from Glencoe Way and overlooking Hollywood Boulevard to the south.
38
It is
commonly believed that Wright was the one who selected the site for the Freemans.
The original budget for the project was under $9,100, however, the final cost was
around $23,000. This increase in cost was likely the result of changes that had to be
made in the design on-site due to Wright’s absence from the project. His son, Lloyd
Wright, oversaw much of the work on all three textile block homes in his father’s
absence, likely resulting in confusion between architect and contractor. Wright had
moved back to Taliesin, making himself fairly inaccessible to the contractors.
39

Samuel Freeman lived in the home for 56 years and served as handyman and caretaker
until his passing in 1981. Over the years, he had constructed terraces and walls in
the yard for flower beds and tried to improve the site drainage.
40
Harriet lived in
the house for a total of 61 years before she decided to donate the it to the School of
Architecture at the University of Southern California.
37 CHUSID, J. M. 1989, p.38. Plans of the house are included in Appendix A.
38 AGUAR, C. E. AND AGUAR, B. 2002. Wrightscapes: Frank Lloyd Wright’s Landscape Designs.
New York: McGraw-Hill, p.196.
39 CHUSID, J. M. 1989, p.12.
40 CHUSID, J. M. 1989, p.159.
56
2.3.2 The design and construction
Three generations of Wright family architects contributed to the Freeman Figure 2.14:
House (left to right: Frank Lloyd Wright
41
, Lloyd Wright
42
, and Eric Lloyd Wright).
43
At Hollyhock House, Wright had used cast concrete decorative elements to adorn
the exterior of the residence. Likely, he was unsatisfied with the results and wanted
to take this idea further, leading to a period of experimentation, which he has been
known to do throughout his career. His experiment in this instance was the concrete
block home and the first instance of such an exploration in Los Angeles came at the
Millard House, also commonly referred to as La Miniatura. The Millard House only
uses mortar and expanded metal mesh to hold the blocks together rather than the
41 WRIGHT, E. L. 2008.
42 WRIGHT, E. L. 2008.
43 GEARY, M. 2004. Eric Lloyd Wright. Anne T. Kent California Room, Marin County Free Library.
Available at http://www.co.marin.ca.us/depts/lb/main/crm/oralhistories/ericlloydwright.html.
[Accessed 6 February 2009].
57
steel reinforcing that the subsequent textile block homes used. The following is a
passage from Wright that describes his intentions behind using concrete block:
We would take that despised outcast of the building industry – the concrete
block – out from underfoot or from the gutter – find hitherto unsuspected
soul in it – make it live as a thing of beauty – textured like trees. Yes, the
building would be made of the ‘blocks’ as a kind of tree itself standing among
other trees in its own native land. All we would have to do is to educate the
concrete block, refine it and knit it together with steel in the joints and so
construct the joints that they could be poured full of concrete after they were
set up and a steel strand laid in them. The walls would thus become thin but
solid reinforced slabs and yield to any desire imaginable. And common labor
could do it all. We would make the walls double of course, one wall facing
inside and the other wall facing outside, thus getting continual hollow spaces
between, so the house would be cool in summer, warm in winter and dry
always.
44
Sketch of the Freeman House by Frank Lloyd Wright. Figure 2.15:
45
Just after he had designed the Storer Residence, the first of the textile block homes,
Wright designed the Freeman’s home between 1923 and the first part of 1924.
Construction began in 1924 and was completed in March of 1925 when the couple
44 KAUFMANN, E. AND RAEBURN, B. eds. 1965. Frank Lloyd Wright: Writings and Buildings.
New York: Meridian Books, p.215-16.
45 WRIGHT, E. L. 2008.
58
received the Certificate of Occupancy and moved into their new home.
46
During the
construction of the Freeman’s home, Wright was working on the third of the three
textile block homes, the Charles Ennis House, just to the north of Aline Barnsdall’s
Hollyhock House.
The Freeman’s home signified a major turning point for Wright as it signaled what
was to come later in his career in projects like the famous Fallingwater residence.
Jeffrey Chusid calls the Freeman House Wright’s first “. . . ‘Usonian House,’ a product
of a changing cultural ethos of the American Home, and Wright’s vision for it . . .” as
it “. . . would imply a different society than had gone before.”
47
The textile pattern on
the blocks gives a character to the building that is reminiscent of patterns that Wright
may have drawn from Mayan architecture. Wright said about his textile blocks that,
“A building for the first time may be lightly fabricated, complete, of mono-material
– literally woven into a pattern or design as was the oriental rug.”
48
Also common in
Wright’s work was the use of patterns from nature, a trend which the Freeman blocks
seem to follow. Although it is unclear as to what exactly is depicted on the blocks,
since Wright never directly revealed this, it is speculated that the pattern represents
“. . . the house itself, nestled among the Eucalyptus trees.”
49
It is clear that some of
the designs of block carving used at the Imperial Hotel in Japan made their way into
46 CHUSID, J. M. 1989, p.89.
47 CHUSID, J. M. 1989, p.37.
48 WRIGHT, F . L. 1927. In the Cause of Architecture IV: Fabrication and Imagination. The
Architectural Record. reprint, 1975. New York: Architectural Record Books and McGraw-Hill, p.146.
49 CHUSID, J. M. 1989, p.42.
59
the design of the Freeman House textile block. In Figure 2.16, a relationship can be
seen between the parallelograms in the Imperial Hotel carving and similar shapes that
occur in the Freeman House block pattern.
Drawings of a stone carving design for the Imperial Hotel Figure 2.16:
50
and the
design of the Freeman House textile block.
51
Lloyd Wright was actively helping his father with supervision on the construction
sites starting with Hollyhock House in 1919.
52
In 1924 and 1925, he was left
alone to oversee the completion of the textile block homes. Likely due to poor
communication from his father, the jobs fell behind schedule and many of the
problems had to be dealt with on-site by LW himself. Lloyd Wright’s son, Eric
Lloyd Wright, also became an architect and would later help with the renovation and
seismic retrofitting of the Freeman house.
53

50 ALOFSIN, A. 1993, p.299.
51 ALOFSIN, A. 1993, p.300.
52 CHUSID, J. M. 1989, p.31.
53 CHUSID, J. M. 1989, p.31.
60
The Freemans moved into their new home without much furniture. The furniture
that Wright did complete was not to Harriet’s liking. Since she seemed to run the
house, Harriet called upon other architects to help her fix the home and the problems
that Wright left her and Sam with. The other architects that the Freemans called
upon and the alterations they made to the house are summarized in the next section.
2.3.3 Architectural modifications
Soon after the home’s completion, the Freemans were ready to make some changes.
In 1928, they commissioned a former apprentice of Frank Lloyd Wright, Rudolph
Schindler, to make revisions to the home and design new furniture. Schindler worked
for Wright in his Chicago office and was dispatched to Los Angeles to take over for
The Freeman living room decorated with furniture by Schindler. Figure 2.17:
54
54 ZIMMERMAN, S. AND DUNHAM, J. 1994. Details of Frank Lloyd Wright: The California
Work, 1909 – 1974. San Francisco: Chronicle Books, p.58.
61
LW on the Hollyhock House construction. Therefore, he must have had extensive
knowledge of the textile block system at the time that Wright was developing it.
While he may have clearly understood the system, he likely did not feel inclined to
repeat it during his time as “. . . family architect for the Freemans from 1928 until
1953,” when he passed away.
55
Schindler completed a number of major remodeling
projects including the replacement of Wright’s furniture, which the Freemans
considered Spartan and uncomfortable at best.
56
Much of Schindler’s furniture was
removed from the home during the seismic retrofitting work in 2000 to 2001 and
is currently in storage. The majority of the furniture that Wright designed, with the
exception of a lamp and table, has either been disposed of or lost.
Robert Clark, another of Frank Lloyd Wright’s Taliesin apprentices, was hired in 1958
to remodel the kitchen. He added a taupe-colored formica to the new cabinets and
shelves and possibly remodeled the second bathroom downstairs, however, there is no
record of Clark doing work in this bathroom.
57
Also during the seismic retrofitting
work in 2000 and 2001, the kitchen cabinetwork of previous architects was removed
and put in storage for protection.
Years later, around 1974, the Freemans asked their acquaintance John Lautner,
who was another apprentice of Wright’s at Taliesin, to change the wooden sashes
of the two-story living room windows to aluminum. The work was completed by
55 CHUSID, J. M. 1989, p.35.
56 CHUSID, J. M. 1989, p.93.
57 CHUSID, J. M. 1989, p.96.
62
a contractor named Wally Newiadamsky, who was working on Lautner’s Silvertop
Residence around the same time. Also at this time, Newiadamsky worked on some
of the other doors and windows around the house, splicing in wood or adding metal
brackets where necessary.
58
Aluminum mullions added by John Lautner. Figure 2.18:
59
The next modifications were implemented by James Reneau around 1979. Reneau
was responsible for designing and installing an elevator for Harriet’s use in her later
years when she was wheelchair-bound. The elevator was located outside on the
west side of the house, near a tool storage area, and was attached through existing
windows, enlarged to allow the addition of a door by removing some of the blocks.
60

The elevator has since been removed from the house and the entries to it are sealed.
58 CHUSID, J. M. 1989, p.96.
59 Author photo.
60 CHUSID, J. M. 1989, p.97.
63
Eric Lloyd Wright, the grandson of Wright, was called in by Harriet during the 1980s
to complete minor repairs. Harriet, however, was not one to spend a great deal of
money. One instance that Eric recalls fondly was when Harriet asked him to repair
the doors leading out to the living room balcony. Instead of replacing them, she
insisted on repairing them and when the bill came to $120 instead of the $80 he had
estimated, she would only pay him the $80.
61

2.4 Previous studies of the Freeman House
Many consultants, groups, and individuals have put years of effort into studying the
Freeman House, documenting its history, analyzing its textile blocks, and determining
its structural stability. The studies undertaken encompass an enormous breath of
topics and almost always explore in-depth a specific area of study. Unfortunately,
many of these studies have been forgotten over the years as interest in the restoration
and rehabilitation of the Freeman House has waned. A majority of the proposals
have never been implemented and the information has been stored away in boxes.
The reasons for this are various including a lack of funding or alternate priorities
at an administrative level. In this section, these studies have been compiled as an
informative exhibition of the wide array of proposals and research that has been
completed over the past 23 years.
61 WRIGHT, E. L. 2008.
64
2.4.1 Wank Adams Slavin Associates proposal and report
In 1991, the architectural, engineering, and preservation firm of Wank Adams Slavin
Associates (WASA), at the request of Jeffrey Chusid, prepared a document titled
T est Program Proposal and Initial Site Visit Report: Samuel Freeman Residence for the
USC School of Architecture with their recommendations for the conservation and
restoration of the textile blocks at the Freeman House.
62
WASA appears to have been
well-suited for this type of work as they had previously consulted on the conservation
of other Frank Lloyd Wright concrete buildings including Unity Temple, Fallingwater,
and the Guggenheim Museum.
The visual observation of the exterior completed by the WASA team revealed that less
than one-third of the blocks had cracking or surface erosion. The interior was noted
to have little damage with the exception of efflorescence, a condition which was not
seen on the exterior. The damage to the exterior was attributed to “settlement, seismic
and other structural movements, organic growth and [inadequate] architectural
detailing.”
63
WASA also noted structural deficiencies at the stair tower and the
terrace retaining wall along with fracturing of the textile block system at the parapets.
During their site visit, a single block was taken from the house and misted with water.
It was noted in the report that “it took 35 seconds for water to reach the inner face
62 WANK ADAMS SLAVIN ASSOCIATES. 1991. T esting Program Proposal and Initial Site
Visit Report: Samuel Freeman Residence prepared for The University of Southern California, School of
Architecture, Samuel Freeman Residence. Completed 30 October 1991. New York: unpublished.
63 WANK ADAMS SLAVIN ASSOCIATES. 1991, p.4.
65
of the channel . . .” and “. . . 2 minutes for the water to reach the back side of the
original block.”
64
A larger scale test was then performed on the north wall of the
house and garage by spraying these walls with a garden hose. They noted the drying
patterns and photographed these patterns in steps. Three preliminary conclusions
were formed at that time: water first enters through cracks in the block, water next
reaches the rear face of the wall through the joints, and finally coffered blocks were
wetted through to the back side much faster than the solid blocks. In the same way,
the centers of the coffered blocks dried first, whereas the thicker areas took longer to
dry out.
To address all the issues found at the Freeman House, a four-phase program for
restoration was recommended: Phase 1, the initial site visit; Phase 2, the design
of a field and laboratory testing program; Phase 3, execution of the test program;
and Phase 4, a final report on recommendations. One of the recommendations
in the report that appears to have been carried out was a “block-by-block survey
. . . of cracking, surface erosion and discoloration of the exterior.”
65
This survey
was completed by Jeffrey Chusid between 1991 and 1992.
66
These drawings are
discussed further in the next section. The testing program that they recommended
included probes, borescope analysis, water testing, magnetometer scanning, infrared
thermography, and multiple-exposure X-ray imaging. Also, as a means of analyzing
64 WANK ADAMS SLAVIN ASSOCIATES. 1991, p.6.
65 WANK ADAMS SLAVIN ASSOCIATES. 1991, p.5.
66 See Appendix A for Chusid’s survey of block damage on the Freeman House exterior.
66
the block and grout composition they advised conducting petrographic microscopy
of thin sections, carbonation depth analysis, percent acid solubles retesting, and X-ray
diffraction of paste.
In regards to the treatment of the exterior blocks, WASA recommended an evaluation
of consolidants and water-repellent treatments with the assistance of the Getty. After
this assessment was completed, WASA was to carry out a program of prototype
conservation techniques on a limited area at the house to evaluate the best treatments
for cleaning, consolidation/water-repellent treatment, grout injection, steel treatment/
supplementing, block repair/replacement, and architectural detailing changes. Of
interest to this thesis would have been their block repair and replacement analysis,
which the report claims would have evaluated the “. . . impact of modification
of cement:sand ratio and of aggregate size distribution on block appearance and
behavior.”
67
Overall, the report recommended restoring the existing blocks “. . . to a satisfactory
level . . . .”
68
rather than replacing blocks. The intention behind this recommendation
seems to be linked to their philosophy that no matter how ordinary the material may
seem, it is original to the historic building and should be preserved when feasible.
However, the report also noted that the mixture of the block material “. . . does not
contain enough cement to coat and bond together the aggregate particles without
67 WANK ADAMS SLAVIN ASSOCIATES. 1991, p.14.
68 WANK ADAMS SLAVIN ASSOCIATES. 1991, p.4.
67
leaving a substantial volume of interconnected voids.”
69
This lean mix is attributed
to the hairline cracking of blocks and the structural instability of the house. Also,
WASA stated that the on-site curing process was likely not sufficient and the blocks
might have been better served by a steam curing process.
2.4.2 Jeffrey Chusid deterioration survey drawings
Between 1991 and 1992, following the directions outlined in the Wank Adams Slavin
Associates report, Jeffrey Chusid, who was at that time the Director of the Freeman
House, undertook a project to document the deterioration of the exterior facades of
the house. Using previously created drawings of the house drawn in AutoCAD, also
done by Chusid with the help of student Adam Smith, Chusid created a system of
depicting block damage conditions and identified the blocks by their type, whether
they had the pattern-face, perforated pattern-face, or flat-face. These drawings of the
deterioration visible in the early 1990s are included in Appendix A.
2.4.3 Nabih Youssef structural report
In 1995, the structural engineering firm of Nabih Youssef & Associates was asked
by the School of Architecture to study the seismic stability of the Freeman House
following the Northridge earthquake. The report sought to evaluate damage to the
structural systems for gravity and lateral forces, assess the vulnerability of the structure
in future earthquakes, and recommend potential repair options that would restore
69 WANK ADAMS SLAVIN ASSOCIATES. 1991, p.10.
68
structural integrity and ensure life safety while reducing damage to the historic
structure. A thorough description of the structural system was provided along
with visual surveys of damage, a geotechnical investigation, and a review of original
construction documents and details. The firm found that the existing structure was
severely lacking in lateral resistance elements and the foundations were inadequately
shallow to support the structure. Also, the earthquake and its aftershocks had
significantly undermined the existing integrity of the structure, necessitating repairs
to the foundations, foundation walls, south facade frames, garage columns, textile
blocks, stair tower, and diaphragms. In the Conclusions section of the report, the
following repairs were specifically recommended:
A. Underpin the entire structure to suitable bearing material.
B. Plumb existing foundation walls and strengthen with new reinforced
concrete retaining walls.
C. Provide new reinforced concrete frames at the south facade, and repair
damaged floor beams.
D. Repair the columns at the north and south sides of the garage door.
E. Repair and reinforce textile block walls.
F . Repair anchorage of stair tower to main house and repair stair tread support.
G. Strengthen diaphragm to shearwall connections and provide new drag
and chord members to provide a continuous load path from every part of the
structure to the foundations.
70
70 NABIH YOUSSEF & ASSOCIATES. 1995. Freeman House: Historic Building Seismic Study.
For the Freeman House, USC School of Architecture, Completed 16 August 1995. Los Angeles:
unpublished, p.20.
69
During the seismic retrofitting work undertaken by the School of Architecture
between 2000 and 2001, items A, B, C, and F appear to have been completed. It is
unclear as to whether the other recommendations have been addressed appropriately.
It can be stated with certainty that item E, the repair and reinforcement of the textile
block walls, has not been completed as there can still be seen many areas where
textile blocks remain damaged and in need of replacement or repair. It is likely that
item G was accomplished with the reconstruction of the south porch and balcony
in reinforced concrete, however, new analysis of the structure would be necessary to
determine if this is the case and what other structural enhancements may be necessary
to ensure the stability and survival of the house in the next seismic event.
2.4.4 Smith-Emery/American Petrographic Services block analysis
On November 13, 2001, Smith-Emery Laboratories submitted a report to the USC
School of Architecture detailing the composition of three samples of textile block
and one sample of grout that they had received from the School. The petrographic
analysis was completed by American Petrographic Services, Inc. (APS) of St. Paul,
Minnesota. The testing by APS looked at unit weight, volumetric proportions,
and petrography in accordance with ASTM:C1324 and APS Standard Operating
Procedure 00 LAB 016. They found that two out of three cube-shaped samples
of Freeman House concrete block used a mixture proportion of one part Portland
cement to two parts sand and gravel. The third block sample used a proportion of
one part Portland cement to one and a half parts sand and gravel. Additionally, they
70
Image of Sample 2 at 15x magnification. Figure 2.19:
71
Image of poorly hydrated belite clusters in Sample 3. Figure 2.20:
72
71 SMITH-EMERY LABORATORIES. 2001. Petrographic Analysis of Submitted CMU portions. For
the Freeman House, USC School of Architecture, Completed 13 November 2001. unpublished, n.p.
72 SMITH-EMERY LABORATORIES. 2001, n.p.
71
found that the samples exhibited soft and porous paste with carbonation up to 1/2”.
The aggregate included granite, quartz and feldspar sand up to 1/4” in diameter as
well as many mica particles and fine gravel up to 3/8” in diameter.
73
2.4.5 Sumit Brahmbhatt’s thesis
In 2006, Sumit Brahmbhatt studied thermal comfort in the Freeman House. He
monitored the interior climate data using a previously installed sensor system and
compared this data to bioclimatic comfort charts. Also, he used the analytical
software programs HEED and eQUEST to show how alternate glazing and insulation
options could improve the performance of this house and the other block houses.
Sensors and output data from Sumit Brahmbhatt’s thesis. Figure 2.21:
74
His study found that the thermal mass of the walls is not sufficient to moderate the
temperatures inside the house, as a typical concrete wall would. He recommended
73 SMITH-EMERY LABORATORIES. 2001, n.p.
74 BRAHMBHATT, S. A. 2006. The thermal energy performance study of the Freeman House. Thesis
(M.Bu.S.). University of Southern California, p.64; 49.
72
using modern glazing in all the windows to improve thermal performance and adding
insulation between the two wythes of the walls to increase the R-value.
2.4.6 Alice Ormsbee’s thesis
Karsten pipe testing on the wall of the Freeman House. Figure 2.22:
75
Also completed in 2006, Alice Ormsbee tested the textile block walls of the Freeman
house and found that the walls are so porous that water passes through them faster
than can be measured using the Karsten pipe test. Water has been known to run
across the floor inside the house while the Freemans still lived there, meaning that
the wall on the street side took in water like a sponge, emptying on the downhill
side. She further studied how water penetrates into the home and determined that
the majority of the water coming into the house passes through the blocks, rather
than leaking through the roof, which was originally thought to be the major source
of leakage. She suggested that more investigation into block coatings be done to
75 ORMSBEE, A. 2006. Water in the Freeman House: a study of the block in the textile block
construction assembly as the primary route for water infiltration. Thesis (M.Bu.S.). University of
Southern California, p.61.
73
determine if one could be used to seal the blocks but prevent negative impacts to their
composition because of the salts embedded within.
2.4.8 Peyton Hall’s USC class update of Chusid’s drawings
In May of 2007, Peyton Hall’s students at USC, in the course Architecture 551:
Conservation Materials and Methods, completed observation studies of the facades
of the Freeman House in order to make updates to Jeffrey Chusid’s 1992 drawings
of block deterioration. The students each surveyed a facade using photographic
documentation and recorded their findings in writing, block-by-block using a
condition matrix. Then, each student overlaid their condition assessment updates on
Chusid’s elevation drawings. In some cases, entire walls of block were missing where
they had been replaced by cast-in-place concrete walls. Damage requiring immediate
attention was also clearly noted.
Photos included in a student’s survey of block conditions. Figure 2.23:
76
76 Photos obtained from a CD included with the report of student Sharon L. Smith, located in the
USC Freeman House Archives. Photos dated 12 April 2007.
74
The reports also included a brief history of the house and what the student considered
to be defining characteristics of the house that should be preserved. They provided
recommendations about which treatment should be followed in keeping with the
Secretary of the Interior’s Standards along with what type of repair should be carried
out for each damage condition that was found. Generally, the students recommended
the treatment of preservation as defined by the Secretary of the Interior.
75
CHAPTER 3: Con CRETE blo Ck And m Ason Ry Cons ERv ATion
In the United States, conventions relating to the preservation and conservation of
historic buildings and sites have been established by the Secretary of the Interior in
the Standards for the Treatment of Historic Properties. Due to the City of Los Angeles
designation of the Freeman House as a Historic-Cultural Monument and because
of the fact that money from FEMA was used in the seismic retrofitting work, any
future work done at the house must now comply with the Secretary of the Interior’s
standards. Worldwide, the United Nations International Council on Monuments and
Sites is responsible for establishing such regulations and ensuring that World Heritage
Sites are correctly maintained. ICOMOS regulations were evaluated as a comparison
to the American standards. Agreed-upon standards are imperative when dealing with
historic properties because they allow restoration work to be consistently applied.
This chapter provides the relevant background information on currently accepted
conservation techniques for concrete block and masonry. Included also is an overview
of the types of deterioration found at the Freeman House and a discussion of what
repair techniques are available as well as which methods are appropriate for use on
Freeman House textile blocks specifically.
3.1 s ecretary of the interior’s standards
The Secretary of the Interior has established standards for the treatment of historic
properties which guide historic preservationists in appropriately handling historic
sites and buildings. There are four methodologies available in these guidelines. These
76
are preservation, rehabilitation, restoration, and reconstruction.
1
Complying with
Harriet’s wishes to use the house for the work of students and teachers, the most
appropriate treatment for the house would be rehabilitation. This treatment gives the
ability to use the house for a new purpose and allows the opportunity for additions
and alterations, where necessary and compatible with the existing structure, as long
as they do not detract from or destroy character-defining features. In all cases, the
standards given by the Secretary of the Interior stress the importance of doing work
that causes the least amount of intervention using gentlest means possible. Where
materials and features cannot be preserved, they should be repaired. When repair
is not feasible, replacement is acceptable using the same material or a compatible
substitute material. The standards for rehabilitation are given by the Secretary of the
Interior as follows:
1. A property will be used as it was historically or be given a new use that
requires minimal change to its distinctive materials, features, spaces, and
spatial relationships.
2. The historic character of a property will be retained and preserved. The
removal of distinctive materials or alteration of features, spaces, and spatial
relationships that characterize a property will be avoided.
3. Each property will be recognized as a physical record of its time, place,
1 For a full description of the available treatments given by the The Secretary of the Interior’s Standards
for the Treatment of Historic Properties, see WEEKS, K. D. AND GRIMMER, A. E. 1995. The
Secretary of the Interior’s Standards for the Treatment of Historic Properties: with Guidelines for Preserving,
Rehabilitating, Restoring & Reconstructing Historic Buildings. Washington, D.C.: U.S. Department of
the Interior, National Park Service. A useful chart comparing all the treatments is available online at
ohp.parks.ca.gov/pages/1054/files/standards%20chart1.pdf. More resources are also available online
at http://www.dsa.dgs.ca.gov/SHBSB/default.htm and http://ohp.parks.ca.gov/?page_id=1078.
77
and use. Changes that create a false sense of historical development, such as
adding conjectural features or elements from other historic properties, will not
be undertaken.
4. Changes to a property that have acquired historic significance in their own
right will be retained and preserved.
5. Distinctive materials, features, finishes, and construction techniques or
examples of craftsmanship that characterize a property will be preserved.
6. Deteriorated historic features will be repaired rather than replaced. Where
the severity of deterioration requires replacement of a distinctive feature,
the new feature will match the old in design, color, texture, and, where
possible, materials. Replacement of missing features will be substantiated by
documentary and physical evidence.
7. Chemical or physical treatments, if appropriate, will be undertaken
using the gentlest means possible. T reatments that cause damage to historic
materials will not be used.
8. Archeological resources will be protected and preserved in place. If such
resources must be disturbed, mitigation measures will be undertaken.
9. New additions, exterior alterations, or related new construction will not
destroy historic materials, features, and spatial relationships that characterize
the property. The new work shall be differentiated from the old and will be
compatible with the historic materials, features, size, scale and proportion, and
massing to protect the integrity of the property and its environment.
10. New additions and adjacent or related new construction will be
undertaken in such a manner that, if removed in the future, the essential
form and integrity of the historic property and its environment would be
unimpaired.
2
2 WEEKS, K. D. AND GRIMMER, A. E. 1995. The Secretary of the Interior’s Standards for
the Treatment of Historic Properties: with Guidelines for Preserving, Rehabilitating, Restoring &
Reconstructing Historic Buildings. Washington, D.C.: U.S. Department of the Interior, National Park
Service, p.62. Also available online at http://www.nps.gov/history/hps/tps/standguide/rehab/rehab_
index.htm.
78
3.2 international standards
The following standards were issued by The International Council on Monuments
and Sites (ICOMOS), which was set up in 1965 through the Charter of Venice.
Their mission is to advance the doctrine and techniques of preservation.
3
In a
meeting held in Athens in 1931, they passed the Athens Charter for the Restoration
of Historic Monuments, giving guidelines about every aspect of historic monuments
and their sites. Last updated January 12, 1996, the most applicable portions for the
topics covered in this thesis are quoted below. The full text of the Athens Charter can
be found in Appendix C.
. . . Proposed Restoration projects are to be subjected to knowledgeable
criticism to prevent mistakes which will cause loss of character and historical
values to the structures. ... Modern techniques and materials may be used in
restoration work.
. . . the Conference noted that there predominates in the different countries
represented a general tendency to abandon restorations in toto and to avoid
the attendant dangers by initiating a system of regular and permanent
maintenance calculated to ensure the preservation of the buildings.
When, as the result of decay or destruction, restoration appears to be
indispensable, it recommends that the historic and artistic work of the past
should be respected, without excluding the style of any given period.
The Conference recommends that the occupation of buildings, which ensures
the continuity of their life, should be maintained but that they should be used
for a purpose which respects their historic or artistic character.
. . .
3 UNESCO WORLD HERITAGE CENTRE. 2008. World Heritage Information Kit. Paris:
UNESCO World Heritage Centre. PDF document. Available at http://whc.unesco.org [Accessed 4
February 2009].
79
The experts . . . approved the judicious use of all the resources at the disposal
of modern technique and more especially of reinforced concrete.
They specified that this work of consolidation should whenever possible be
concealed in order that the aspect and character of the restored monument
may be preserved.
. . .
The Conference noted that . . . monuments throughout the world were being
threatened to an ever-increasing degree by atmospheric agents.
. . .
The Conference recommends:
That, in each country, the architects and curators of monuments should
collaborate with specialists in the physical, chemical, and natural sciences with
a view to determining the methods to be adopted in specific cases;
. . .
With regard to . . . monuments, the experts unanimously agreed that,
before any consolidation or partial restoration is undertaken, a thorough
analysis should be made of the defects and the nature of the decay of these
monuments. They recognised that each case needed to be treated individually.
. . .
2 August 1994; modified 12 January 1996
4
The standards set forth by ICOMOS are generally closely aligned with those set up
by the Secretary of the Interior. Any major differences are largely administrative and
4 INTERNATIONAL COUNCIL ON MONUMENTS AND SITES. 1996. Athens Charter for the
Restoration of Historic Monuments. Athens: ICOMOS. Available at http://www.icomos.org/athens_
charter.html. [Accessed 4 February 2009].
80
procedural in terms of documentation of sites and educational outreach. By having
standards in place worldwide, there can be a sense of agreement about how well a
structure has been preserved, the level of authenticity, and the important features
that imbue a structure with historical value. However, the limiting factor in the case
of the Freeman House is its existing designation and conditions of previous funding
which have predetermined the standards which will be followed. The importance
of following these standards like these in working with historic buildings cannot be
stressed enough. Without a common set of guidelines to work from, the definition of
what is appropriate would largely be left to individual property owners and architects
and could create a false sense of what is in fact historic versus new construction.
3.3 b lock deterioration found at the Freeman House
As discussed in Chapter 2, there is significant block deterioration found throughout
the exterior walls of the Freeman House as well as some limited areas of deterioration
on the interior walls. This section catalogues the types of deterioration found at
the house and gives the probable causes for the damage found. Included also is a
prognosis for what further damage could occur if these areas are not treated in a
timely manner. In order to monitor worsening damage sites Regularly scheduled
surveying of individual block condition should be completed. This allows the areas in
greatest danger to be treated before areas that are determined to be stable.
81
3.3.1 Chipping
Chipping on the exterior roof parapet Figure 3.1:
5
and on the east wall.
6
In many areas, chipping has occurred as a result of extreme cases of cracking, block-
on-block movement in earthquakes, or through accidental impact to the wall. Some
of this block-on-block contact chipping may have occurred during the Northridge
earthquake. It is difficult to determine where this is in fact the case because
insufficient photo documentation is available for the facades before and after 1994.
The best evidence available is the survey of block conditions completed by Jeffrey
Chusid in 1991 and 1992.
There are additional places at the Freeman House where blocks were intentionally
chipped away to allow the passage of mechanical, electrical, and plumbing services
that were part of later additions to the home. These intrusions to the historic block
present a major problem as they are not properly sealed and easily allow water and
debris to enter the wall cavity along with wildlife that might nest there. Despite the
5 Author photo.
6 Courtesy of the USC Freeman House Archives, CD3. Photo dated 3 October 2000.
82
marginal protection of the walls against water intrusion, holes in the block render
the task of sealing the house from impossible. Therefore, these areas are the first
priority for repair and retrofitting. If not repaired, these areas could allow further
deterioration of the block by permitting water to penetrate the interior core of the
block units, leading to spalling and possibly the delamination of a block face.
3.3.2 Cracking
Cracking found on an exterior retaining wall. Figure 3.2:
7
Many blocks exhibit cracking, ranging from minor to very severe. These cracks
have allowed water to penetrate not only to the interior of the home but also to the
steel reinforcement between the blocks. This causes the steel to rust and expand,
further cracking the blocks and promoting dislocation. Cracking has also occurred
in the grouting between blocks, which contributes to the rusting of the steel as well.
The total effect of this becomes a structural system that is significantly weaker than
Wright had designed it to be. The solution to this problem is not readily apparent.
7 Courtesy of the USC Freeman House Archives, CD3. Photos dated 19 January 2001 and
3 October 2000.
83
Any sealant will have a difficult time sealing large cracks. Some type of compatible
material patch would likely be the solution to cracking problems, followed by a
thorough sealant application.
3.3.3 Efflorescence
Efflorescence found in the living room and hallway. Figure 3.3:
8
Efflorescence has several possible causes, either inherent to the block mixture or
caused by outside forces. At the Freeman House, efflorescence has occurred as a
result of salt crystals embedded in the block being loosened and rising to the surface
of the porous block when water moves through the block. When the salt reaches
the surface, it crystallizes and appears on the surface as a white cosmetic blemish.
While the issue of efflorescence is largely cosmetic, it is a sure sign that there is water
infiltration through a masonry wall and can lead to deterioration of the material if
not treated. As noted in the description of Alice Ormsbee’s thesis work in Chapter 2,
efflorescence in the Freeman House blocks was found primarily on interior walls.
9
8 Courtesy of the USC Freeman House Archives, CD3. Photos dated 3 October 2000.
9 ORMSBEE, A. 2006. Water in the Freeman House: a study of the block in the textile block construction
assembly as the primary route for water infiltration. Thesis (M.Bu.S.). University of Southern
California.
84
3.3.4 Erosion
Erosion of block faces. Figure 3.4:
10
Erosion of block patterns and corners has occurred on nearly all of the exterior blocks
in varying degrees due to the friable nature of the concrete mixture. Many of the
pattern blocks have lost their crisp edges as can be seen in Figure 3.4. This decay is
likely to have resulted from the material that was used to make the blocks. Some
of the blocks have deteriorated faster than others, which could be attributed to the
different crews working on the blocks at various times during the construction,
varying grades of material being used in the mixtures, and a lack of cohesion between
the grains of sand or aggregate through insufficient hydration of the cement or
insufficient amounts of cement in the mixture. A solution to this issue may involve
consolidation of the block surfaces to halt the erosion where it is minor or repointing
of the pattern where damage is moderate. Replacement of the block face as a veneer
is an option where erosion is more severe, however, any water that gets behind this
repair could force the substrate to further erode and detach the repair work. If repairs
are not made, erosion will be allowed to continue and further destroy historic blocks.
10 Author photos.
85
3.3.5 Spalling
Block spalling on a retaining wall near the home’s entry. Figure 3.5:
11
Spalling at the Freeman house is quite prevalent. It is a condition where the top layers
of the block break off or peel away. This is a common phenomenon in fabricated
masonry materials, such as cement products and terra-cotta. Spalling is usually
attributed to weathering and the pressure of the salts that have been held in under
the surface of the block.
12
This type of deterioration at the Freeman House may also
be linked to the material composition of each block and the inconsistent mixture
used for the blocks. The inadequate adhesion of aggregate particles by insufficiently
hydrated cement could also be a contributing factor to the prevalence of spalling in
the exterior blocks. Blocks that include more soil material, added as a colorant, may
be more prone to crumbling than those that utilized a majority of sand and cement.
Likewise the cleanliness of the materials used in the mixture could have contributed
to the ability of ingredients to adhere to each other and form a solid block unit.
11 Courtesy of the USC Freeman House Archives, CD3. Photo dated 9 November 2000.
12 GRIMMER, A. E., 1984. A Glossary of Historic Masonry Deterioration Problems and
Preservation Treatments. Washington, D.C.: Department of the Interior, National Park Service, p.20.
86
A consolidant is likely the best form of treatment for this type of deterioration.
Without treatment, spalling could continue in exterior blocks, allowing layers of the
blocks to fall away.
3.3.6 Biological growth damage
Ivy growth on the walls and damage resulting from biological growth. Figure 3.6:
13
Previously, ivy was allowed and encouraged to grow on the walls of the house by
Harriet Freeman. While these vines were beautiful to Harriet and thought to be
innocuous, it was later found during the transfer of the house that the ivy had actually
deteriorated the block underneath and left stains on the face of the porous block.
The ivy plants trapped moisture against the block faces and their roots grew into the
cracks between blocks, worsening the problem of water infiltration into the structure.
Chusid did note that while destructive, the benefit of the ivy was that it created a
buffer for the walls and kept the inside of the house cooler during the warm summer
months. This effect was not fully realized until the plant growth had been removed
13 WRIGHT, E. L. 2008. Frank Lloyd Wright and the Freemans. Lecture. University of Southern
California. 20 November 2008.
87
from the home’s exterior.
14
Today, while all the ivy has been removed from the house,
much of the damage that resulted from this plant growth remains. Proper cleaning
techniques should be utilized to ensure that no further damage occurs in the blocks.
Cracks that occurred as a result of the roots growing into the blocks are continuing
to allow water to penetrate through the wall. Also, the staining left by the ivy growth
detracts from the visual character of the house and should be cleaned in a gentle
manner.
3.4 b lock conservation and repair techniques
Due to the extensive block deterioration found at the Freeman House, it will be
necessary to develop a plan for treating the blocks, both interior and exterior, so that
future damage and weathering is limited. However, not all of the deterioration found
will require immediate attention and repair. As discussed in Chapter 1, some experts
have argued that restoring the house to a “like-new” condition may not be the most
appropriate treatment for a home that can be viewed as a demonstration of Wright’s
experimentation with a new system of concrete, including the faults that appeared in
this system. It is, perhaps, appropriate to allow some non-structural wear to evidence
the problems that have been encountered with the textile block system. Some areas,
however, are in danger of further decay if treatment is not undertaken very soon. The
following sections describe common treatments for masonry that might be considered
14 Eric Lloyd Wright discussed these problems during his 2008 lecture at USC. Also, Jeffrey Chusid
has described the problems with the ivy in Modernist Threads on p.193-194.
88
for use at the Freeman House along with their ramifications for the future longevity
and integrity of the historic fabric of the home.
3.4.1 Waterproof coating
A waterproof coating can be applied to the block surfaces to seal them from water and
vapor infiltration. Coatings are either clear or opaque. In the case of the Freeman
House, a clear, matte finish coating would be ideal to preserve the coloration and
visual character of the blocks. However, waterproof coatings have the potential to
present problems to historic structures when water does get inside or there is water
extant in the substrate. Once a waterproof coating is applied, this water has no place
to escape, creating damage to the coating and risking further damage to the substrate
block. An additional concern with coatings is that any coating may have the effect
of changing the coloration and appearance of the block (eg. matte to shiny). Any
coating option would need to be extensively tested on sample blocks to ensure that
no harm is done to the original blocks while gaining durability and to maintain visual
consistency with the original finish.
3.4.2 Water repellent coating
The advantage of a water repellent coating is that it may allow the block to breathe
better than a waterproof coating by allowing water vapor to enter and leave. The
same problems still apply with the possible change in coloration to the blocks,
especially due to improper application. The downfall of a water repellent coating is
89
when water vapor condenses inside the wall, it cannot escape out through the coating
in liquid form. Additionally, waterproof and water repellent coatings can prevent
dissolved salts within the wall from evaporating. This is especially a concern at the
Freeman House given the prevalence of efflorescence seen on the blocks. It would be
necessary to try to remove these salts from the blocks by water washing.
3.4.3 Water washing
Water washing is a very gentle technique of washing exterior walls of a historic
masonry structure and can use various methods including a “prolonged spraying
using a fine mist, high or low pressure washing, steam, water in combination with
detergents and water in combination with chemicals.”
15
Damage is possible using
high pressure and it is likely that such a washing would cause damage to the Freeman
House blocks. This treatment can be effective in removing efflorescence, however,
the results may be temporary as deeply embedded salts can rise to the surface after
washing. Washing can occur over an extended period of time and is generally done
by spraying the water through small holes in a pipe or from a hose nozzle. Any
technique used would need to be tested first to ensure that it does not damage the
block surface.
15 GRIMMER, A. E., 1984, p.46.
90
3.4.4 Caulking
A clear silicone or color-matched caulking could be applied between the blocks to seal
the large gaps where water may be entering. This treatment may degrade the historic
appearance of the home if the caulk has a glossy finish that is incompatible with the
existing block. Also, a caulking product has the potential to damage the existing
historic block material by pulling off block material as it peels away during material
decay or if the caulk is removed for repairs.
3.4.5 Chemical cleaning
There are several areas at the Freeman House that may require some extent of
chemical cleaning to remove biological growth. As mentioned above, water washing
may also use chemicals as part of that treatment. Chemical cleaners are generally
either acidic, having a low pH, or alkaline, having a high pH. For the Freeman
House, an acidic cleaner would likely be acceptable, but it should be tested to verify
that no damage to the blocks would occur during the chemical cleaning. The acidic
cleaner would be rinsed from the surface with either a thorough water washing or
with a neutralizer. Great care must be exercised to ensure that the chemicals used are
strong enough to adequately remove the substances that threaten further damage to
the block composition yet gentle enough to not damage the blocks even further.
91
3.4.6 Poulticing
This technique can be used to reverse the effects of plant material infiltration and
efflorescence. This technique uses a cloth that draws out the salts from the block
rather than redepositing the stain or pushing it deeper into the masonry. While time
intensive, this process may be necessary in some areas. Further evaluation would need
to be done to determine the effectiveness of this treatment at the Freeman House.
3.4.7 Composite patching / plastic repair
Patching is a technique that uses a cementitious material to repair small areas of decay
and can be quite successful when the substrate material is properly prepared. This
repair would be a viable option in areas with minor chipping, cracking, erosion, and
spalling as long as the patching material is a like or compatible substitute material.
Possibly, a similar mix could be made and applied to the block face, however, the
attachment of this repair mix would be questionable as the blocks themselves have
trouble holding together.
3.4.8 Consolidation
Consolidation is a technique that attempts to strengthen masonry by joining the
separating pieces of a masonry unit that are separating as a result of spalling. The
treatment of the surface is usually done with barium hydroxide or a chemically-
92
curable monomer like methyl methacrylate and n-butyl methacrylate or a clear
silicone polymer.
16
The downside of this process can be that the substance used to
treat the surface does not reach deep enough into the block to adequately consolidate
the particles, leaving the block to continue its decay. Despite these difficulties, this is
likely a treatment option that would be used for the Freeman House.
3.4.9 Mechanical repair
Mechanical repair is used to reattach pieces of masonry that have come loose from the
building. It can also be used in areas where chipping has occurred. This technique
uses drilled pins with grouting to create a mechanical bond between the two detached
pieces of masonry. The difficulty with application of this technique at the Freeman
House is that many of the missing pieces may be lost and the blocks are likely too
fragile to drill into.
3.4.10 Paint and stucco
These treatment techniques could be used, however, they would severely alter the
historic look of the home and remove the characteristic color and texture of the
blocks. Some interior areas in the bathrooms and the apartment below the garage
have received an off-white paint. While this coating may have helped to preserve
the block on the interior and prevent surface damage, the paint does not maintain
consistency with the material texture in the remainder of the home and goes against
16 GRIMMER, A. E., 1984, p.52.
93
Wright’s intent to celebrate the nature of the material.
3.5 Conservation options summary
In complying with Harriet Freeman’s requests for the use of her former home, it will
likely be necessary to apply the treatment of rehabilitation given by the Secretary of
the Interior in order to bring this house back into normal use and provide the School
of Architecture with a valuable asset as a teaching tool and event space. In light of the
various degrees of damage and diverse deterioration conditions including chipping,
cracking, efflorescence, erosion, spalling, and biological growth cracking and staining,
a multi-faceted approach must be undertaken to repair the home’s exterior. Ideally,
the least invasive and gentlest treatments should be considered first. One portion of
this approach will include full replacement of blocks. While all of the repair methods
are important to consider in the full rehabilitation of the house, this study is generally
concerned only with the block replacement portion of the repair spectrum.
3.5.1 Block replacement principle
There has been no definitive statement made about when a block should be replaced,
what mixture should be used for replacement blocks, and how to replace blocks
within the existing walls. It is likely that blocks exhibiting significant damage that
could undermine the structural stability of the home will require replacement. A
clear statement about prerequisites that must be met for replacing blocks should be
established so that the maximum amount of historic material can be preserved intact
94
with the house. As was done during the seismic retrofitting work in 2000 and 2001,
blocks removed from the house shall be properly protected from damage with a foam
wrap, documented, and stored in a safe, weatherproof location. As will be discussed
in Chapter 9, further work will need to be done in order to determine what criteria
will determine which blocks need to be replaced and how newly manufactured blocks
will be placed into the existing wall without damaging adjacent blocks.
95
CHAPTER 4: TEXTILE BLOCK RESEARCH METHODS
4.1 Methodology
In François de Larrard’s book Concrete Mixture Proportioning: A scientific approach, he
describes very succinctly the process of optimizing a concrete mixture when he writes,
“Before designing a material it is important to clarify the nature of the problem with
regard to the specific application that is aimed for. Most often the goal is to find a
combination of constituents that will give a concrete whose properties comply with
certain specifications. This is the basis of what is sometimes called a ‘performance-
based’ approach.”
1
The fundamental goal of replacing severely damaged blocks at the
Freeman House with better performing reproductions was the driving force in this
thesis. These replacement blocks would be made using modern-day mixture materials
and the construction method developed by Frank Lloyd Wright in 1923. The newly
made blocks would then be compared to the existing blocks to see if they did in fact
perform better than the ones that were originally made in 1924.
Therefore, the scope of work was designed around a three step process that
included the making of new blocks, testing these blocks to compare their physical
characteristics, and deriving conclusions about their performative qualities from the
testing. The blocks would be evaluated for their ability to resist water penetration,
maintain consistency with the existing coloration and texture, have appropriate
1 LARRARD, F . D. 1999. Concrete Mixture Proportioning: A scientific approach. London; New York:
E & FN Spon; Routledge, p.251.
96
slump, and demonstrate a level of durability beyond the existing blocks. While the
first three criteria are fundamental to a successful rehabilitation of the house, it must
be noted that the assessment of durability is fundamentally at odds with accepted
historic preservation practices. This study looks at durability in the hypothetical
sense of what Frank Lloyd Wright might have done differently if he were designing
the textile block system in the present. If he could see how the concrete blocks have
deteriorated, he might have decided to make fundamental changes in the way the
mixture was designed.
4.2 Block manufacturing
The process of making a Freeman House textile block. Figure 4.1:
2
The first step of this thesis research was to manufacture new textile blocks as samples
for my testing efforts. Both flat-face and pattern-face blocks were to be made in order
to determine if the mix being used was suitable to creating all the types of blocks
that are found on the house or if different mixes might be necessary for patterned
versus flat-faced blocks. It was initially assumed that 20 blocks could be made using
a technique perfected approximately five years ago by students led by Assistant to
the Dean Dana Smith at the USC School of Architecture. The process involved
2 Author photos.
97
compacting the blocks using a hydraulic hammer press and then allowing the blocks
to air cure outdoors, protected from rain under a covered structure and misted with
water periodically during curing as Wright had originally specified. Then these blocks
would be put through the series of tests described in the previous section.
Instead, the process was in fact more difficult to produce blocks that were
satisfactorily formed. It was found through research that the block mix that was
specified by Frank Lloyd Wright did not match the measured composition of the
blocks on the house found thought testing at Smith-Emery Laboratories.
3
This
measured composition of two parts sand to one part cement was what students were
using in previous attempts at block making and was inti ally the formula used for the
blocks made in this research effort. At this point, a more extensive approach to block
making was undertaken including the variation of proportions of mixture ingredients
to arrive at a mixture that was dry enough to be removed immediately from the mold,
yet had a certain amount of workability that would allow it to me properly formed.
This extensive undertaking involved the use of alternate aggregates and cements in
combination with various additives. Also, the proportions of the mix (water, sand,
cement, and additives) was varied to test other possible mixes that would prove more
water resistant and less prone to the deterioration that is seen in the existing blocks.
Other techniques were also attempted such as covering blocks with plastic wrap to
3 SMITH-EMERY LABORATORIES. 2001. Petrographic Analysis of Submitted CMU portions. For
the Freeman House, USC School of Architecture, Completed 13 November 2001. unpublished.
98
hold in moisture during the curing process or applying waterproof coatings after the
blocks had cured to initial strength.
Peter Purens making a Freeman House block using the original mold. Figure 4.2:
4
4 SMITH, K. Summer 2005. The L.A. Textile Block Houses. Frank Lloyd Wright Quarterly. 16 (3),
p.7.
99
Finally, because of difficulty encountered with forming patterned blocks, two different
casting methods were attempted: face-up and face-down. It is unclear in all the
sources available exactly how the patterned blocks were originally formed. Jeffrey
Chusid gave a conjectural interpretation in his Historic Structure Report and has
suggested that the patterned blocks may have been formed face-down.
5
Peter Purens,
a mason who worked on the restoration of the Millard House, had worked out a
system of layering material in the block mold and tamping it by hand, which was
photographed by Kathryn Smith in 1986 as seen in Figure 4.2.
6

Based on these photos, it was decided that both hand compaction and mechanical
compaction would be attempted to determine if there was a significant difference
in block formability for pattern-face blocks and whether the compaction methods
changed the compressive strength of the blocks. Even more importantly, it was
necessary to determine whether the water resistance of a block unit could be increased
with hydraulic compaction. The major downside encountered with face-down casting
was the difficulty of removing the block from the mold before it had cured to initial
strength, since the mold is much better suited for face-up casting. Any blocks that
were cast face-down would have to be left overnight to cure before removal, negating
the ideal of a simple, on-site block manufacturing method that Wright hoped to
implement with this system.
5 This suggestion was made to the author by Jeffrey Chusid through e-mail communication on 12
September 2008.
6 See SMITH, K. Summer 2005. The L.A. Textile Block Houses. Frank Lloyd Wright Quarterly. 16
(3), p.7 for a full description of the blocks and the manufacturing process.
100
It was apparent that this step was to be far more iterative than originally imagined.
Each round of block making built on the next and lessons learned from testing steps
done between block batches informed the next round of blocks that were made.
While the duration of this step extended far beyond original plans, it provided
valuable discoveries about the process of block making and how these blocks might
be improved. Changes to the block mixture were based in large part on the results
obtained in the Karsten pipe testing completed between block batches.. Also at this
time, test samples were cast for later compressive strength testing of the material being
used in the newly manufactured blocks.
4.3 Block testing
4.3.1 Slump test
The slump test is a measure of the wetness of a concrete mixture, which determines
the ability for the concrete to flow during placement.
7
Blocks were evaluated for
slump characteristics using an informal test based on a standardized slump testing. A
plastic, tapered cup, similar to a slump cone, was filled with the mix material, covered
with a trowel, and then inverted without compressing the material. The cup was
slid off the trowel onto a table at which point the cup was removed and the slump of
the material was recorded. A standard slump test designed by the American Society
for Testing and Materials (ASTM) uses a metal slump cone (ASTM C 143) that is
7 AMERICAN CONCRETE INSTITUTE. 1974. ACI Manual of Concrete Practice, Part 1:
Materials and Properties of Concrete, Construction Practices and Inspection, Pavements and Slabs.
Detroit: ACI, p.211-2
101
filled from the top.
8
When the cone is lifted off, the slump is measured in relation
to the original height of the cone. It was decided that this standardized test could be
satisfactorily replicated with a plastic cup for the Freeman House block mixture due
to an initial expectation that the slump would essentially be zero since the mixture
uses very little water. Other laboratory tests for no-slump concrete described in
the American Concrete Institute’s Manual of Concrete Practice, which use the Vebe
apparatus, the compacting factor apparatus, and the Thaulow drop table, were not
attempted in this thesis. Future research may benefit from these tests in being able to
more fully understand the properties of the Freeman House blocks.
4.3.2 Karsten pipe testing
Diagram of the Karsten pipe test on horizontal and vertical surfaces. Figure 4.3:
9
The absorptive characteristics of the completed blocks were tested through the Karsten
pipe test. This test is a standard for measuring the amount of water that penetrates a
surface in a given period of time under low pressure. The Karsten pipe, a graduated
8 AMERICAN CONCRETE INSTITUTE. 1974, p.211-54.
9 GUARD INDUSTRIE. 2000. ProtectGuard: Oil and Water Repellent: Water absorption test under
low pressure. Montreuil Cedex, France: Guardindustrie.com. Available at http://www.guardindustry.
com/gb/produits/1-ProtectGuard/ [Accessed 22 April 2008].
102
tube measured in milliliters, was adhered to the block surface using a ring of putty to
form a watertight seal between the pipe and the test surface. Water was then poured
into the tube and filled to the top line, which is labeled “0.” Once the pipe was full,
the timer was started. Measurements were taken at two minutes, five minutes, and
ten minutes. This test provided a clear benchmark by which all the block mixtures
were compared. As stated in the previous section, block manufacturing mixtures
and methods were adjusted based on Karsten pipe test results obtained between
manufacturing of the block batches.
4.3.3 Compressive strength of block material
Compressive strengths of select mixtures were tested through standardized cylindrical
sample testing. T wo samples of six inches in diameter by twelve inches in height were
initially poured and allowed to cure for a minimum of 30 days before testing. The
actual age at testing for these two samples was 42 days. One sample was made using
hand compaction and the second was made with mechanical compaction. Testing
was carried out through the assistance of Smith-Emery Laboratories, located in Los
Angeles, California. The cylindrical samples of mixture material were evaluated in
a test machine designed to measure the total load in pounds required to cause the
cylinder to fail. Based on this value and the cross sectional area of the sample, the
test engineer could calculate the pounds per square inch (psi) that the material can
withstand, or its compressive strength.
103
Compressive strength test stand at Smith-Emery Laboratories. Figure 4.4:
10
It was suggested by the test engineer during the first round of testing that smaller
samples be used for testing due to the small size of the sand aggregate relative to the
volume of the test cylinder. Therefore, later in the block manufacturing process, seven
additional samples of four inches in diameter by eight inches in height were cast and
allowed to cure for a minimum of 30 days. Testing was again carried out through
the assistance of Smith-Emery Laboratories. The test results were photographed and
videos were recorded documenting the sample preparation and testing process.
10 Author photo.
104
4.3.4 Color and texture comparison
Finally, after all the block manufacturing had been completed, color and texture
comparisons were carried out to assess the appropriateness of the new blocks to the
historic character of the existing house. The color of the blocks manufactured were
compared to the color of blocks that were removed during renovation work at the
Freeman House. These original, but detached, blocks were chosen as benchmarks
instead of blocks currently on the house because of the ease of controlling the lighting
under which both new and original blocks would be photographed.
The Munsell ColorChecker® card. Figure 4.5:
11
First, the camera was white balanced under the interior lighting conditions using a
Munsell brand ColorChecker® card. This card provides 24 squares of standardized
colors that are manufactured in conformance with procedures set forth by the
National Institute of Standards and Technologies (NIST).
12
Photographs of newly
11 Author photo.
12 According to the product packaging, “Munsell Color Services Laboratory is an internationally
recognized laboratory holding accreditations from ISO 9001 and ISO 2000.”
105
manufactured blocks and original blocks were taken under interior, controlled
lighting, both ambient and directional, with the Munsell ColorChecker® card in the
frame. Then, utilizing Adobe Photoshop, the photographs were color corrected based
on the standard colors in Munsell ColorChecker® card. Next, again using Photoshop,
pixel histograms were used to compare the photographs based on the amount and
luminosity of red, green and blue pixels found in the block face. The pixel histograms
were overlaid on their respective photos and these were placed side-by-side to
determine if the color of the new block is on par with that of existing blocks. The
Example of histogram sampling in Adobe Photoshop. Figure 4.6:
13
13 Author diagram.
106
coloration was assumed to be acceptable if it was within the range of block colors
found on the house and if multiple blocks from the same mixture could give multiple
variations in the color.
Based on a visual analysis, new blocks were also compared to the original blocks in
terms of their surface texture. The primary motivation here was to develop a block
that would have a similar feel to the existing blocks on the house, which have a
rough, sandy texture, very dissimilar from a traditional smooth concrete finish. This
analysis, which was ongoing during the block manufacturing phase, became essential
to determining the final composition of the replacement block recommended. In
keeping with the friable texture of the Freeman House exterior, new blocks would
have to embody the same character, at least on their exterior face.
4.4 Data collection and presentation
Based on all the testing results and the discoveries made during block manufacturing,
recommendations were then derived as to the most appropriate block mixture to use
for replacement blocks and what the implications of this choice would have for the
house. Pictures of the block making process and a video of this process were recorded
as a record of the process for making the Freeman House blocks, since little record of
the actual process used exists beyond Kathryn Smith’s photos of Peter Purens making
a block. It was hoped that by recording this process, it would eliminate future
questions or misunderstandings about the process of block manufacturing used in this
research. Photos were also taken of each of the blocks manufactured to record the
107
physical characteristics and the success or failure of each mixture recipe. The photos
and video are included in Appendix D. Photos and descriptions of all the necessary
tools used in the block-making process are provided in Chapter 5.
Recipes for the mix designs along with any special instructions were recorded for each
block manufactured during the course of this thesis research. Tables of each batch
were compiled with the proportions of each ingredient and any special treatment that
the block received. These tables are found in Chapter 5. These block recipes and test
results were further organized for analysis in a comprehensive table that can be found
in Chapter 7.
Photos of the Karsten pipe testing and videos of the water draining through the
pipe were recorded. A log of all these photos and videos can be found in Appendix
E. The results of the Karsten pipe tests were compiled in a table located in Chapter
6 and analyzed to arrive at a conclusion determining what block is the most water
resistant and could function as a replacement block at the Freeman House. All of
the pixel histogram comparisons can be found in Chapter 6. Photos and videos of
the compression testing were taken during both compression tests at Smith-Emery
Laboratories. These are included in Appendix F .
108
CHAPTER 5: TEXTILE BLOCK MANUFACTURING
In the process of this thesis research, a series of different mixture proportions,
additives and forming methods were attempted in order to determine which mixture
of ingredients is easiest to form, which will hold together when the aluminum mold
is removed, which is most consistent with the color and texture of the original blocks,
and which will best reduce or prevent water penetration through the exterior face of
the block. This chapter details the mixture ingredients and tools used to manufacture
replacement block samples along with the challenges encountered throughout the
process. The final textile block mixtures offer possible solutions to the criteria for
analysis established in Chapter 4.
5.1 Textile block composition
In Frank Lloyd Wright’s original construction specifications for textile blocks of the
Freeman House, he instructed his contractor to use a mixture of one part Portland
cement to four parts sand. During an analysis of block composition provided to the
USC School of Architecture on October 15, 2001 by American Petrographic Services,
Inc., it was revealed that two out of three cube-shaped samples of Freeman House
concrete block had a volumetric proportion of one part Portland cement to two
parts sand and gravel. The third block sample had a proportion of one part Portland
cement to one and a half parts sand and gravel.
1
These mixture proportions are
1 SMITH-EMERY LABORATORIES. 2001. Petrographic Analysis of Submitted CMU portions. For
the Freeman House, USC School of Architecture, Completed 13 November 2001. unpublished, n.p..
109
significantly different than those that Frank Lloyd Wright had specified in his original
construction documents. Likely, construction crews found the cement content
to be too low and changed the mix during construction in order to successfully
manufacture the textile blocks. As was attempted in Block 32 of this study, the
mixture of one part Portland cement to four parts sand was excellent in its ability
to be formed, however the low cement content was insufficient to bond the sand
particles together, causing the block to fall apart on the curing rack. In Figure 5.1,
Block 32, which failed within hours after its casting. Figure 5.1:
2
it is clear that Block 32 had a well-formed pattern. Consequently, the low cement
content, while allowing the mixture to move in a more plastic manner, prevented the
mixture from developing the necessary paste between aggregate particles, causing the
upper portions of the channel sides to collapse within hours of the block being cast.
2 All figures in this chapter are by the author unless otherwise noted.
110
The variation in the original block mixtures found by American Petrographic
Services, Inc. could be attributed to the “hand of the artisan,” as Frank Lloyd Wright
might have described it. The other block houses also varied in their composition,
even including soil in the mixture to give a buff color to the mixture. This proved
especially catastrophic at the Ennis House where a significant block replacement effort
was undertaken to restore the home’s facade in 2008. The mix used for new blocks,
as discussed in Chapter 1, utilized several different materials including white cement,
birdeye gravel, Gillibrand washed concrete mix
3
, Unamin #8, bunker sand, water, and
two coloring dyes.
In a previous effort by USC students to make replacement blocks for the Freeman
House in 2004, the ratio of materials followed the testing results from American
Petrographic Services of one part cement to two parts sand with very little water.
During mixing, the water was added gradually and the mixture was evaluated by
squeezing it in the mixer’s hand. Once the mix began to clump with slight fracturing,
it was ready to be cast in the formwork.
4
This low water content makes it possible
to remove the form from the block almost immediately after compacting it so that
the formwork can be reused multiple times in a single day. Typically, much more
3 This is a coarse sand with all the fines washed out. It comes from Gillibrand’s Simi Valley pit.
Information obtained from an e-mail conversation with Jeff Keenan, President of Moonlight Molds,
Inc. on 23 February 2009.
4 This qualitative description was provided to the author by Dana Smith, former Assistant to
the Dean for Special Projects & Programs at the USC School of Architecture, during a phone
conversation. Smith and his student workers produced reproductions of the Freeman House blocks
and tiles in 2004. An article highlighting their efforts appeared in the Los Angeles Times. See HART,
H. 2004. When the answers just aren’t concrete. Los Angeles Times, 26 September, p.E.42.
111
water would be used in a conventional concrete mixture, however, the higher waster
content would cause the block mixture to cure more slowly. Therefore, blocks would
need to be left in the formwork much longer, preventing the immediate reuse of the
formwork. Wright’s specification for this dry-pack mixture was tied to his aspirations
for a system that was able to be rapidly constructed without a great deal of formwork
and could be done by the average homeowner. This idea may have been driven by a
desire for economic savings or by Wright’s interest in the merging of hand-craft with
the machine.
In this study, the mixture ratio of one part cement to two parts sand was deemed
to be the base mixture to which changes were made including the alteration of
the mixture proportions, substitution of mixture ingredients and the addition of
admixtures. Specifically, the base ingredients were Colton Portland Cement and
Quikrete Washed Plaster Sand. Approximately 15% of the total base mixture was
water. The newly cast blocks made throughout this study were assessed in terms
of their workability inside the formwork, how soon the block could be removed
from the mold after casting, and whether the block held the shape of the mold after
the formwork was removed. The possibility of making a block that reaches initial
strength more quickly to allow moving the block more easily was also a consideration.
Alternative ingredients and additives that were tried in the mixtures made during this
research included Quikrete Premium Play Sand, Quikrete Medium Sand, Quikrete
Hydraulic Water-Stop Cement, Quikrete Concrete Acrylic Fortifier, Sika Acrylic
Fortifier & Bonding additive, Quikrete Concrete Mix Accelerator, and Concrete
112
Pharmacy Flow Control. Also, some blocks received two coats of Sika Natural Look
Clear Sealer after curing.
5.2 Tools and formwork used in manufacturing replacement blocks
The majority of the tools used to make textile blocks are standard for concrete work,
or can be easily made. Some of the formwork, however, is rare and expensive to
reproduce. In keeping with the methods used by workmen on the Freeman House
construction site, the mixture ingredients for blocks made in this study were mixed in
batches both by hand and using a concrete mixing machine (Figure 5.2). A metal
Concrete mixing machine. Figure 5.2:
mesh screen attached to a wooden box was placed over the form box and the mixture
material was pushed through this mesh to break up clumps in the mixture (Figure
5.3). Then a leveling stick can was used to level off the mixture material to the top of
113
Wooden sifting box with metal mesh screen. Figure 5.3:
the form box. The original form box and pattern-face mold for the Freeman House
along with a reproduction form box and flat-face mold were used as the formwork for
this manufacturing step of the thesis. The pieces of the formwork necessary to make a
block include the form box (16” x 16” inside dimensions with half-round aluminum
bars around the inside to form the channels on the block edges), the coffered
aluminum back support (or the wooden back support when making concrete tiles),
which fits inside the form box, a coffered steel backing plate (or a flat steel backing
plate when making concrete tiles), which rests on top of the back support, the flat-
face mold (16” x16” square), and the pattern-face mold (16” x16” square).
The original Freeman House form box open (left) and closed (right). Figure 5.4:
20” 16”
Fixed channel molds Double lock mechanism
114
The reproduction form box open (left) and closed (right). Figure 5.5:
Coffered aluminum back support. Figure 5.6:
Coffered steel backing plate placed on top of aluminum back support. Figure 5.7:
Removable channel molds
Double lock mechanism
Notches allow
steel backing
plate to be lifted
off for transfer
to curing rack
115
Form box with coffered back support and backing plate in place. Figure 5.8:
Original Freeman House pattern-face mold. Figure 5.9:
Reproduction flat-face mold (impression surface shown at left; top Figure 5.10:
surface with handles shown at right).
116
All of the formwork was sprayed extensively with WD-40 to lubricate the surfaces
and discourage the concrete mixture from sticking to the forms.
5
Once the material
was placed into the form box and the pattern-face or flat-face mold applied on top,
the box was vibrated for approximately 20 seconds to help the material settle and
promote better compression. To compress the block in the hydraulic hammer press,
a pressure distribution plate was used so that one point in the center did not receive
all the pressure and ruin the formwork. When compressing by hand, a wooden block
was placed between the mallet and the aluminum formwork to protect the formwork
from damage.
The vibration device and speed control located below the hydraulic Figure 5.11:
hammer press.
5 This technique was suggested to the author by Dana Smith, Michael Curiel, and Michael
Villaseñor, who all have experience with making the textile blocks in these aluminum forms. It is
the belief of the author that other form release materials may work as well or better than WD-40 as a
form release. This topic is discussed further in Chapter 9.
117
Pressure distribution plate used for machine compaction. Figure 5.12:
Hydraulic hammer press and compressor used for machine compaction Figure 5.13:
of the replacement textile blocks.
Extreme care was exercised to protect the formwork from scratches, thus it was
cleaned thoroughly after every use. As discussed in Chapter 2, these are the last
remaining forms of any of Wright’s textile block homes. While reproductions
have been created, the reproduction pattern-face mold could not be used for
manufacturing because of its increased thickness due to concerns about the warping
seen in the original pattern-face mold.
118
5.3 Method for manufacturing textile blocks and tiles
As mentioned previously, the exact process followed in making the original blocks
of the Freeman House in 1924 is not clearly known. The replacement blocks made
in 2004 used a process in which blocks are formed face-up so that they could be
transferred to a drying rack on the steel backing plate. This process was similar to that
demonstrated by mason Peter Purens, who was photographed making a block in this
manner by Kathryn Smith in her 2005 article,
6
however Purens used only manual
compaction with a mallet and wooden block. Jeffrey Chusid has suggested the
possibility that pattern-face blocks were formed face-down in order to create a well-
defined impression.
7
This technique was explored later in the block manufacturing
process after difficulty was encountered with casting a pattern-face block in the face-
up position.
The following is a description of the process for block making in the face-up
orientation which was the primary method used in this thesis. This is a compilation
of recommendations given by previous block makers and photos of blocks being
made along with the experience that came with manufacturing blocks for this thesis.
8

Figure 5.14 shows all of the primary steps in the block manufacturing process. A
video of this process can be found in Appendix D.
6 SMITH, K. Summer 2005. The L.A. Textile Block Houses. Frank Lloyd Wright Quarterly. 16 (3),
p.7.
7 Information obtained through e-mail communication with Jeffrey Chusid on 12 September 2008.
8 The use of rubber gloves, eye protection, and a respirator is highly recommended during mixing.
119
The block manufacturing process. Figure 5.14:
120
5.3.1 Block manufacturing steps
1. The first and most important step is to prepare the work area. Ensure that the
concrete mixer is clean of excessive concrete build up and empty it of any concrete
left over from a previous batch. Set up the forms and clean all surfaces including
grooves and hinges of leftover mixture using a soft-bristle brush and WD-40. This
will ensure that a clear impression is produced and that the form box will close
properly. Also, clean the mesh screen of the wooden sifting box. Clear off space
on the curing rack to accommodate the new blocks. Fill spray bottles with water
to mist the blocks after they are transferred to the curing rack.
2. Add sand and concrete in the proper proportions into the concrete mixer. T urn
on the mixer to combine the ingredients. Begin to add the appropriate amount of
water slowly, ensuring equal distribution throughout the entire mixture. During
the mixing, the side of the mixer should be scraped with a trowel to release
mixture material that clumps on the sides.
3. The next step is to prepare the formwork to accept the mixture material. If more
than one person is making blocks, this step can be done concurrently with Step 2.
To make a full size block with the side channels, place the coffered aluminum back
support on the table. On top of this, place the coffered steel backing plate. Spray
the plate with WD-40.
121
4. Close the form box around the backing plate and support plate and lock the box
on its side. Spray inside the form box with WD-40, ensuring that all surfaces are
coated. T ransfer the mix from the concrete mixer to a wheelbarrow and roll it to
the table. Sift the mix into the mold using the wooden sifting box. Once the box
is half full, compact the material into the underside of the channel mold using a
wood block and mallet. Then continue sifting the material into the form box until
it is filled to the top. Level the top with a leveling stick.
5. Spray the flat or pattern-face mold with WD-40 thoroughly and place it inside the
edges of the form box. Lightly tap on the mold using a wood block and hammer
to seat it in the mixture material.
6. Slide the form box under the hydraulic hammer press. T urn on the vibration
knob for 20 seconds to allow the material and mold to begin compacting. Add
the pressure distribution weight to the top of the assembly and align the form box
under the hydraulic piston. Depress the lever and compress the mold to 15,000
psi. The top surface of the flat or pattern-face mold should be level with the
top of the form box sides. Release the lever and retract the piston. Remove the
pressure distribution weight and slide the box out from under the hydraulic press.
7. Release the lock on the side of the form box and pull away one side at a time,
being sure that the other sides do not pull away. Once the sides of the form box
are cleared from the block, lift the top flat or pattern-face mold from the block by
pulling straight up using the handles. Pulling at an angle will destroy the block.
122
8. Lift the block underneath the steel backing plate where there are notches in the
aluminum back support. Move the block to the curing rack. Be careful not to
jolt the block as it is set down on the rack as any sudden movements will almost
inevitably cause damage to the block.
9. Mist the block with water as it cures approximately every 10 to 15 minutes for at
least 2 hours. The material will be cured to initial strength in approximately 24 -
48 hours.
5.4 Mix Designs
One of the primary goals of this thesis research was to develop a mixture that could
be used in replacement blocks for the Freeman House that would not only mimic
the texture and color of the existing blocks at the house, but would prevent water
infiltration to the interior. In this process, many additives have been tried along with
different ratios of mixture ingredients and water content. The following ten tables
give detailed information for every batch of test samples and blocks produced in
this research effort. The notes indicate special treatments given in some instances
or problems that occurred in specific blocks. In Chapter 7, an evaluation of all the
different mixtures, additives, and sealants will be presented using the discoveries made
during the block manufacturing phase and data obtained during testing.
Without a doubt, the materials available to today’s builders are far more varied than
those available in 1924. Therefore, the concrete block mixture ingredients used here
123
may not be historically accurate, however their use can be justified for this research
in the spirit of utilizing the house as a laboratory for research. Nonetheless, great
consideration has been given to creating blocks that are similar and compatible to the
appearance of the existing blocks. Generally, blocks that are cast with a low water
content mixture produce an appearance in keeping with the texture of the Freeman
House blocks. If the blocks were formed with a high water content, they would
have a more smooth surface once cured than the dry-pack mixture. Techniques that
would not produce a similar appearance to the original blocks were avoided since they
would significantly alter the character and appearance of the facade. In order for any
conservation effort to be successful at the Freeman House, it must be able to blend
with the existing textile blocks.
Block 31 cast with a wet mix exhibits a smooth surface in contrast to the Figure 5.15:
texture of the existing Freeman House blocks.
124
Notes
Compressed
from 13” to full
compaction then
filled again and
compacted until
full compaction
reached at 12”.
Compressed from
13” to 12”.
Compressed
from 12” to 8”,
more added and
recompressed.
Same as #S1
Compressed from
12” to 8”
Water added to
top of tube when
poured.
Additional
Ingredients
Water Content
11 cups
Mixed as a
batch with #W
10 cups
Mixed as a
batch with #S1
Mixed as a
batch with #S1
Mixed as a
batch with #S1
Aggregate
Content
40 cups washed
plaster sand
Mixed as a batch
with #W
40 cups washed
plaster sand
Mixed as a batch
with #S1
Mixed as a batch
with #S1
Mixed as a batch
with #S1
Cement
Content
20 cups Portland
Cement
Mixed as a batch
with #W
20 cups Portland
Cement
Mixed as a batch
with #S1
Mixed as a batch
with #S1
Mixed as a batch
with #S1
Type
6”x12”
Test
Sample
6”x12”
Test
Sample
4” x 8”
Test
Sample
4” x 8”
Test
Sample
4” x 8”
Test
Sample
4” x 8”
Test
Sample
Mix Number /
Compression
S1
Machine
S2
Manual
S3
Machine
S4
Machine
S5
Manual
S6
Manual
Date
October 30, 2008 November 19, 2008
Table 5.1: Test samples cast on October 30, 2008 and November 19, 2008
125
Notes
Very wet mix;
not quite
enough for a tile
Very wet mix
Additional
Ingredients
1 1/3 cups
concrete acrylic
fortifier
Natural clear-look
sealer applied after
curing
1 1/3 cups
concrete acrylic
fortifier
1 1/3 cups
concrete acrylic
fortifier
Natural clear-look
sealer applied after
curing
Water Content
4 cups
8 cups
6 cups
1.5 cups
3 cups
6 cups
Aggregate
Content
22 cups washed
plaster sand
32 cups washed
plaster sand
26 cups washed
plaster sand
26 cups washed
plaster sand
52 cups washed
plaster sand
26 cups washed
plaster sand
Cement
Content
10 cups Portland
Cement
12 cups hydraulic
water-stop cement
12 cups Portland
cement
12 cups Portland
cement
24 cups Portland
cement
12 cups Portland
cement
Type
Flat-face
tile
Flat-face
tile
Flat-face
tile
Flat-face
tile
Flat-face
block
Flat-face
tile
Mix Number /
Compression
1
Machine
2
Machine
3
Machine
4
Machine
5
Machine
6
Machine
Date/
T emp.
December 3, 2008 - Outdoor Temperature = 68° F
Table 5.2: Blocks cast on December 3, 2008
126
Notes
One batch used
for all day’s blocks;
water doesn’t seem
to scale accurately
Cast upside down;
last block of day-
mix too dry
Cast second of
day; covered with
plastic wrap on
shelf
Cast third of day
Additional
Ingredients
4 cups concrete
acrylic fortifier
Mixed as a batch
with #7
Mixed as a batch
with #7
Mixed as a batch
with #7
Water Content
28 cups
Mixed as a
batch with #7
Mixed as a
batch with #7
Mixed as a
batch with #7
Aggregate
Content
160 cups washed
plaster sand
Mixed as a batch
with #7
Mixed as a batch
with #7
Mixed as a batch
with #7
Cement
Content
80 cups Portland
Cement
Mixed as a batch
with #7
Mixed as a batch
with #7
Mixed as a batch
with #7
Type
Flat-face
block
Flat-face
block
Flat-face
block
Flat-face
block
Mix Number /
Compression
7
Manual
8
Manual
9
Manual
10
Machine
Date/
T emp.
January 6, 2009 - Outdoor Temperature = 64°F
Table 5.3: Blocks cast on January 6, 2009
127
Notes
Mixed with #15;
covered with
plastic wrap on
shelf
Mixed with #13
Mixed with #12
Left to cure in
form box because
mix was too wet
Mixed with #11
Cast upside down
Additional
Ingredients
1 1/2 cups
concrete acrylic
fortifier
Natural clear-look
sealer applied after
curing
1 1/2 cups
concrete acrylic
fortifier; Natural
clear-look sealer
1 1/2 cups
concrete acrylic
fortifier
Water Content
8 cups
10 cups
10 cups
12 cups
8 cups
8 cups
Aggregate
Content
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
Cement
Content
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
Type
Flat-face
block
Flat-face
block
Flat-face
block
Flat-face
block
Flat-face
block
Flat-face
block
Mix Number /
Compression
11
Manual
12
Manual
13
Machine
14
Manual
15
Manual
16
Machine
Date/
T emp.
January 7, 2009 - Outdoor Temperature = 70°F
Table 5.4: Blocks cast on January 7, 2009
128
Notes
Mixed with #18
Mixed with #17
Mixed with #20;
Covered with
plastic wrap on
shelf
Mixed with #19;
Covered with
plastic wrap on
shelf
Same batch used
for #24 and #25
Covered with
plastic wrap on
shelf
Additional
Ingredients
Natural clear-look
sealer applied after
curing
Natural clear-look
sealer applied after
curing
Water Content
10 cups
10 cups
10 cups
10 cups
10 cups
10 cups
10 cups
Aggregate
Content
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
Cement
Content
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
Type
Flat-face
block
Flat-face
block
Flat-face
block
Flat-face
block
4” x 8”
Test
Sample
4” x 8”
Test
Sample
4” x 8”
Test
Sample
Mix Number /
Compression
17
Manual
18
Machine
19
Manual
20
Machine
23
Manual
24
Manual
25
Machine
Date/
T emp.
January 8, 2009 - Outdoor Temperature = 67°F
Table 5.5: Blocks cast on January 8, 2009
129
Notes
Cast last of
day upside
down; removed
from mold on
1/14/2009
Cast last of day
upside down
Too wet to cast
blocks; added
sand and remixed
Additional
Ingredients
Water Content
Mixed as a
batch with #29
Mixed as a
batch with #29
10 cups
10 cups
10 cups
24 cups
Aggregate
Content
Mixed as a batch
with #29
Mixed as a batch
with #29
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups washed
plaster sand
80 cups washed
plaster sand;
remixed with 12
cups extra
Cement
Content
Mixed as a batch
with #29
Mixed as a batch
with #29
20 cups Portland
Cement
20 cups Portland
Cement
20 cups Portland
Cement
40 cups Portland
Cement
Type
Pattern
block
Flat-face
block
Pattern
block
Pattern
block
Pattern
block
Pattern
block
Mix Number /
Compression
21
Machine
22
Manual
26
Machine
27
Machine
28
Machine
29
Machine
Date/
T emp.
January 9, 2009 - Outdoor Temperature = 56°F
Table 5.6: Blocks cast on January 9, 2009
130
Notes
Mixed with #31;
play sand is a bit
more wet than
washed plaster
sand – higher
moisture content
Cast last of
day upside
down; removed
from form on
1/21/2009
Additional
Ingredients
Water Content
10 cups
10 cups
Aggregate
Content
40 cups play sand
40 cups play sand
Cement
Content
20 cups Portland
Cement
20 cups Portland
Cement
Type
Pattern
block
Pattern
block
Mix Number /
Compression
30
Machine
31
Manual
Date/
T emp.
January 16, 2009 - Outdoor
Temperature = 83°F
Table 5.7: Blocks cast on January 16, 2009
131
Notes
T ried to flip
block - not very
successful; very
fine grain and dry;
fell apart when
mold was removed
Fine grain
mixture; block
hardened in 30
minutes
Best pattern to
this point
Decent pattern
but cracked easily
Consistency of
peanut butter;
left in form until
2/16/2009
Additional
Ingredients
1 box concrete
accelerator
4 cups concrete
acrylic fortifier
7 cups Sika
Acrylic fortifier &
bonding
Water Content
8 cups
10 cups
6 cups
none
16 cups
Aggregate
Content
40 cups washed
plaster sand
40 cups washed
plaster sand
40 cups medium
Quikrete sand
40 cups play sand
30 cups washed
plaster sand
Cement
Content
10 cups Portland
Cement
15 cups Portland
Cement
20 cups Portland
Cement
15 cups Portland
Cement
30 cups Portland
Cement
Type
Pattern
block
Flat-face
block
Pattern
block
Pattern
block
Block
“blank”
Mix Number /
Compression
32
Machine
33
Machine
34
Machine
35
Machine
36a
Machine
Date/
T emp.
February 12, 2009 - Outdoor Temperature = 62°F
Table 5.8: Blocks cast on February 12, 2009
132
Notes
Topping for
block #36a
Additional
Ingredients
1 box concrete
accelerator
1 box concrete
accelerator
Water Content
10 cups
10 cups
Aggregate
Content
40 cups washed
plaster sand
40 cups washed
plaster sand
Cement
Content
20 cups Portland
Cement
20 cups Portland
Cement
Type
Pattern
topping
Pattern
block
Mix Number /
Compression
36b
Machine
37
Machine
Date/
T emp.
February 19, 2009 - Outdoor
Temperature = 83°F
Table 5.9: Blocks cast on February 19, 2009
133
Notes
Cast face down
with just enough
mix to create the
pattern; cure for
30 minute before
adding #38b; mix
seemed more dry
than earlier mixes;
sprayed with water
in form box.
Mixed while
#38a was curing;
poured wet on
back of #38a;
allowed to set 20
hours before form
was removed;
some adhesion of
mix to form was
noticed.
Additional
Ingredients
2 boxes concrete
accelerator
2 boxes concrete
accelerator
Water Content
10 cups
18 cups
Aggregate
Content
40 cups washed
plaster sand
40 cups washed
plaster sand
Cement
Content
20 cups Portland
Cement
20 cups Portland
Cement
Type
Pattern
top face
Block
backing
Mix Number /
Compression
38a
Manual
38b
Machine
Date/
T emp.
February 28, 2009 - Outdoor Temperature = 83°F
Table 5.10: Blocks cast on February 28, 2009
134
Notes
Cast face down
with just enough
mix to create the
pattern; cure for
30 minute before
adding #38b; mix
seemed more dry
than earlier mixes;
sprayed with water
in form box.
Mixed while
#38a was curing;
poured wet on
back of #38a;
allowed to set 20
hours before form
was removed;
some adhesion of
mix to form was
noticed.
Additional
Ingredients
Mixed as a batch
with #38a
Mixed as a batch
with #38b
Water Content
Mixed as a
batch with
#38a
Mixed as a
batch with
#38b
Aggregate
Content
Mixed as a batch
with #38a
Mixed as a batch
with #38b
Cement
Content
Mixed as a batch
with #38a
Mixed as a batch
with #38b
Type
Flat top
face
Block
backing
Mix Number /
Compression
39a
Manual
39b
Machine
Date/
T emp.
February 28, 2009 - Outdoor Temperature = 83°F
Table 5.10 (continued): Blocks cast on February 28, 2009
135
Notes
Sand mixed with
diluted Flow
Control and
left overnight;
cement added
next day with 6
cups water and
6 cups sand; mix
had insufficient
adhesion
Not enough
material put in
form to create the
pattern. Material
still did not seem
to flow that well.
Additional
Ingredients
2 bags Concrete
Pharmacy Flow
Control
1 bag Concrete
Pharmacy Flow
Control
Water Content
10 cups
10 cups
Aggregate
Content
40 cups washed
plaster sand
40 cups washed
plaster sand
Cement
Content
20 cups Portland
Cement
20 cups Portland
Cement
Type
Pattern
block
Pattern
block
Mix Number /
Compression
40
Machine
41
Manual
Date/
T emp.
March 1, 2009 - Outdoor Temperature = 71°F
Table 5.11: Blocks cast on March 1, 2009
136
5.5 Challenges encountered during manufacturing
The manufacturing process during this research took a great deal of effort and saw
many problems with mixing ingredients, forming blocks, and transferring those
blocks to the curing rack. Undoubtedly, Frank Lloyd Wright’s contractor must have
encountered some of these same issues in 1924. However, if there were any problems
in 1924, they were overcome. Batch by batch, problems were solved and the
invention of a replacement block solution came closer to being realized.
5.5.1 First attempts
What seems on the surface to be a straightforward process turns out to be more
difficult than previously imagined or described by others. After mixing some test
batches in a five gallon bucket, it was found that mixing larger batches in the mixing
machine did not exactly translate in terms of the water content. It was also seen early
on how important it is to get the water content just right. Too little water produces a
Checking the consistency of the mixture. Figure 5.16:
137
mix that is dry and will fall apart once the block is removed from the formwork.
Too much water caused the mix to take longer to set up and therefore it cannot be
removed from the forms immediately. Another important finding was that the dry
mixture strained the mixing machine’s motor if a large batch was being prepared.
After making three flat-face tiles, it began to be clear what the correct method is and
how to compact the tile and remove it from the mold without damaging it. These
first three were deficient in several ways including a broken corner, a stress fracture
down the middle from lifting the metal backing plate in the center (incidentally, the
plate is not thick enough to resist bending under the weight of the concrete mixture),
and a large gouge caused by a stray finger while transferring the freshly formed tile
from the wooden block to the curing rack. The tendency of the backing plate to
stick to the wooden block was remedied to some extent by applying WD-40 to the
underside of the plate before it was placed inside the form box.
The first attempts at making Freeman House tiles showing a broken Figure 5.17:
corner (left), a stress fracture (center), and a gouge in the edge (right).
By the time the fourth tile was attempted, sufficient experience had been gained so
that a successful tile could be produced and transferred to the curing rack without
damage. After a few more successful tiles, a patterned tile was attempted, but
138
A successfully formed tile on top of the wooden support block. Figure 5.18:
the pattern did not compact evenly. The deepest areas of the pattern were fully
compressed, however, the fine detail on the upper surface of the pattern did not show
at all because the pattern could not be pushed deep enough into the mold. It was at
this point that the struggle with forming a patterned block began.
A patterned block was also attempted with the assumption that a full-size block
would give more volumetric area for the material to compress and thus produce a
clear pattern. Instead, similar problems were experienced with the impression of the
Unsuccessful pattern-face block with cracks and broken corners. Figure 5.19:
139
Pattern-face block with lower channels uncompacted. Figure 5.20:
pattern and the underside of the channel was not get evenly compacted. It appeared
that the material was not moving in a plastic manner and the pressure from the
hydraulic press was not adequately reaching the underside of the channel molds.
Consequently, a step was added to the process so that the concrete mixture could be
packed by hand into the channel underside, ensuring that mixture material filled all
parts of the form box.
Misting tiles with water on the curing rack. Figure 5.21:
140
According to Wright’s specifications, the blocks would be sprayed with water while
they were curing for ten days. However, due to time constraints and the lack of a
large work crew, this was a requirement was not accurately adhered to during the
block manufacturing process. Blocks were sprayed every fifteen minutes while curing
on the day they were cast and were again sprayed during the following day when
possible. It was unclear in the early stages whether this would have any negative effect
on the durability of the blocks.
5.5.2 Improvements in the process through experience
After experiencing problems with forming the pattern-face blocks, it was decided
that a lower water content was necessary for a more workable mixture that could
better move inside the form box under the weight of the patterned mold. Using such
a mixture, the pattern was able to be transferred flawlessly and was far superior to
previous attempts. However, when the form box was opened, the block fell apart like
a pile of sand. After several more attempts, a mixture of 40 cups of sand to 20 cups of
Block 10 perfectly formed with flat-face mold and 40:20:10 mix. Figure 5.22:
141
cement with 10 cups of water emerged as the optimal combination for forming a flat-
face block. Despite this success, the proper mix or technique for making a pattern-
face block was still uncertain. In casting Block 8, the flat-face mold was used to form
the block face-down to see if face-down casting had any viability for possibly creating
a pattern-face block. This option seemed to be viable until the form box was removed
and the sides of the block above the flat-face mold sheared off. The mixture was still
too dry to create a sufficient paste between the aggregate particles.
Block 8 in the form box (left) and after removal (right). Figure 5.23:
More attempts at the pattern-face block were made, even trying to add material
incrementally and recompress the block each time material was added until the
pattern was well-formed in all areas. This seemed to have some merit, however, the
addition of material created an uneven top layer that sloped from one side to the
other. It was clear at this point that using the 40:20:10 mixture that worked will for
the flat-face blocks would not be so successful for the pattern-face blocks. Instead, a
different mix would need to be used or the face-down casting technique would need
to be perfected so that it could produce blocks as fast as the face-up method.
142
An improvement in compacting the underside of the channel was achieved by using
a wood block and mallet to better compact the material. It appears from the photos
shown in Figure 4.2 that this is what Peter Purens did as well. This greatly improved
the formation of the lower half of the channels.
At this point in the process, more alternatives were tried and commercially available
products were added into the mixture in hopes of decreasing water penetration
through the block. Three different products were tried: Quikrete Concrete Acrylic
Fortifier, Quikrete Hydraulic Water-Stop Cement, and Sika Natural Look Clear
Sealer. The acrylic fortifier was added to the mix as a water reducing additive.
The additives tried included acrylic fortifier (left), water-stop cement Figure 5.24:
(center), and clear sealer (right).
Initially, it seemed that blocks were better formed using this product and they did
not fall apart as often when they were transferred to the curing rack. The water-stop
cement was substituted for Portland cement. The downside of this product was that
it set up quickly and this made it difficult to fill the form box and compress the block
143
fast enough before the mixture reached an initial curing point. Also, the cement
became very thick with just a small amount of water, so more water had to be added
to allow proper mixing of the ingredients. The sealer was tried on several blocks after
they had cured for several days. Also, it was thought that holding in the moisture
would help to wet cure the blocks so that daily misting with water would not be as
great a priority. In order to accomplish this, a clear plastic food wrap was draped over
the block once it was set on the curing rack. The first problem with this approach was
the fragility of the blocks immediately after casting. It was nearly impossible to spread
the plastic wrap over the blocks without it clinging to itself or without damaging the
block. For Block 9, the plastic wrap was successfully applied and moisture could be
seen on the underside of the wrap for several days after casting.
Block 9 with plastic wrap to trap moisture during curing process. Figure 5.25:
144
Even with these improvements, there were still mishaps. An attempt to increase the
water in the mixture in order to promote better formation of the block turned out
to unsuccessful in Block 14. The addition of too much water to the mixture caused
it to ooze out water when compressed, creating an undesirable block texture and
appearance as can be seen in Figure 5.22. The face of the block cured to a much
softer gray color and a very different texture than the blocks using the 40:20:10
mixture. Unlike the rest of the blocks, the sand grains on the surface of Block 14 did
not rub off when the surface was rubbed. Since this did not match the existing blocks
on the Freeman House, it was concluded that too much extra water in the mixture
was not the answer.
Block 14 cured to a more gray finish than other blocks. Figure 5.26:
5.5.3 Refining the method
In addition to the Quikrete Concrete Acrylic Fortifier, Quikrete Hydraulic Water-
Stop Cement, and Sika Natural Look Clear Sealer, other additives were tried in
successive block making attempts. These included Sika Acrylic Fortifier & Bonding,
145
Quikrete Concrete Mix Accelerator, and Concrete Pharmacy Flow Control. Block
35 used the Sika Acrylic Fortifier & Bonding, but very little difference was found
between this and the Quikrete Concrete Acrylic Fortifier. There was still some benefit
to the use of an acrylic product noted.
The Quikrete Concrete Mix Accelerator did show more promise, however, when it
was used in Block 33, a flat-face block. With the addition of one box of accelerator
to the mixture, the block set up within 30 minutes of curing. This appeared to be the
solution to face-down casting for the pattern-face blocks. If the blocks could be set
to an initial curing strength in 30 minutes, then the block could be removed from the
mold to allow the reuse of the formwork.
Face-up and face-down casting methods that were attempted. Figure 5.27:
Another product that was intended to help the pattern-face casting in the face-up
position was the Concrete Pharmacy Flow Control. This is a water reducing additive
that the manufacturer claims will increase compressive strength and increase fluidity.
An increase in fluidity inside the mold was exactly what was necessary to form the
pattern-face block in a face-up orientation. However, despite two attempts with
this product, a successful pattern was not achieved. Likely, there would need to be a
Back support and backing plate Pattern-face mold
Form box
146
higher water content in the mixture for this product to be effective. The product did
not seem to fully dissolve in the water nor did it mix well with the sand and cement.
Between block batches, Karsten pipe testing was being carried out also. Initial
findings in this testing showed that the high water content mixtures performed better
in the Karsten pipe test than any of the other blocks, even those that were sealed.
Based on this finding and the success with using a concrete mix accelerator, the
development of a hybrid concrete block emerged. The first attempt at a hybrid block
was for Block 36, which was literally two separate block pieces that were joined. A
wet concrete mixture (Block 36a in Table 5.8) was cast in the form box and flattened
on top, but with a large open space between the top of the mixture and the top of
the form box. This was to allow a topping layer to be cast later. After leaving this
wet casting to cure for several days, a dry-pack mixture (Block 36b in Table 5.9)
was prepared and added to the top of the smooth cast block backing. This attempt
appeared to be somewhat successful, with the exception that the indentations in the
pattern reached deeper in to the form box than the first casting would allow, so the
final thickness of the block was not accurate. The major disappointment with this
block was the separation of the two layers when the block was being moved after
several days on the curing rack. The topping layer slid off the smooth casting of Block
36a. This option would be sure to fail if applied to the Freeman House.
Based on the failure of Block 36, the hybrid block idea was pursued further, but
with one important change: the two mixtures would be adhered while they were
147
still uncured. The block would be cast in the face-down orientation. The thinking
behind this concept was that the face of the block could exhibit the color and texture
of the existing Freeman House blocks, while the interior half of the block would help
to prevent water transmission through the block. First, the dry mixture (Block 38a
The hybrid block process used in Blocks 38 and 39. Figure 5.28:
in Table 5.10) was prepared with the mix accelerator added and this mixture was
compacted by hand over the pattern-face mold. Then a higher water content mixture
(Block 38b in Table 5.10), also with accelerator, was made and poured on top of the
uncured first layer. The backing plate and support plate were then placed on top
and the block was compressed with the hydraulic hammer press. This technique was
also tried for a flat-face mold (Block 39 a & b in Table 5.10) so that the block could
easily be tested with the Karsten pipe. Once cured, these blocks were a success. The
two layers of mixture material did not detach from each other and the block looked
consistent with the original Freeman House blocks on the exterior face.
Block 38 successfully cured. Figure 5.29:
STEP 1 STEP 2
148
5.5.4 Considerations
One of the major bottlenecks in this process that touts speed of production and ease
of manufacture is the formwork itself. With only one form box, the total production
ability of a group of workers is severely limited. When a large batch is mixed, it needs
to be used quickly, and this cannot be accomplished with only one or two sets of
formwork.. At least one additional form would be ideal so that one block could be
prepped in the mold while another is being compacted.
An improvement for future attempts at block manufacturing might be to set up
a misting system, similar to those found in the produce section at a grocery store,
which would turn on at a specified interval and mist the blocks on the curing rack
throughout the day for ten days. While this would fulfill the specifications that
Wright called for in the construction documents, it is still unclear whether this would
provide any significant improvement in the blocks.
The greatest difficulty experienced in making the pattern-face blocks in a face-up
orientation was the reluctance of the mixture material to flow into all the crevices of
the pattern to produce a well-defined block. Instead, the deepest parts of the pattern
were super-compressed, while the rest of the block pattern was under-compressed and
poorly defined. Michael Villaseñor, a student at USC who previously made blocks
under the supervision of Dana Smith, suggested that the material was not sifted
finely enough. Despite efforts to ensure sifting of the material before it went into the
mold, the ability to form a successful pattern-face block in the face-up orientation
149
was still out of reach. The blocks made during this process used a variety of materials
including concrete acrylic fortifiers, play sand, medium grade sand, and concrete
accelerator. Despite all these products meant to improve the mixture, the problem of
incomplete impression of the pattern was still encountered when the block was cast
in a face-up position. The only success in producing a consistently clear pattern was
when the block was cast in a face-down position. This, however, makes it difficult to
make more than one block per day, as the block had to cure to initial strength before
it was able to be removed. The only other option would be to flip the filled formwork
onto its backside without ruing the block. This is quite difficult considering the
weight of the form box when it is full and the design of the formwork assembly,
which is geared towards a face-up casting.
5.6 Final recommendations on the manufacturing process
Throughout the entire process of block manufacturing, it was realized that the
procedure for making blocks seems to be fatally flawed, at least in the case of the
pattern-face blocks. With a mix that is congealing when squeezed in the hand, it
would follow with some logic that the mix will have difficulty moving in a plastic
manner as a more typical concrete mix would do. The only solution found in this
research effort was to cast blocks face-down or to meticulously scrape out mix material
from the face-up pattern and recompress the block until the pattern was properly
formed in all areas. Both these methods seem unlikely to have been used by Wright’s
workmen considering their time-intensive nature. The face-down casting method
150
would have only been viable in 1924 if there were a way to flip the block onto its
backside after compressing it face down, allowing the block to be successfully removed
from the form box and moved to a drying rack.
Modern technologies could be used to create replacement blocks on a massive
scale. Such a technology was used in the reconstruction efforts at the Ennis House.
Moonlight Molds was contracted to manufacture the replacements and used a
fiberglass mold that was formed from a perfectly cast block.
1
The same method
could be used to make replacement blocks for the Freeman House. Another possible
method for creating formwork could be to vacuum form the molds from a medical
grade plaster casting of a patterned block. This medical grade casting has been
utilized already in efforts to create reproduction pattern molds for the Freeman House
and may be a viable solution to the problem of too few molds. The solution of using
concrete accelerator seems to have promise. More mixtures would have to be tested
to come to the right mixture that is optimized for appearance, strength, and speed of
curing, allowing multiple blocks to be cast face-down in a single day.
1 Information obtained during a phone conversation with Jeff Keenan, President of Moonlight
Molds, Incorporated. 17 November 2008.
151
CHAPTER 6: TESTING OF CONCRETE TEXTILE BLOCKS
6.1 Testing choices
As discussed in Chapter 4, the tests chosen for this research were paired with the
block manufacturing process to evaluate the physical characteristics of replacement
block options for the Freeman House textile blocks. The primary objective was to
find a way to keep water from entering the block through its exterior face, but this
had to be accomplished in a way that would not compromise compatibility with the
texture and color characteristic to the original Freeman House blocks.
The slump test was chosen for its ability to measure the consistency of a concrete
mixture. This standard test, used in most concrete construction, measures the amount
a mixture slumps when a metal cone filled with the concrete mixture and lifted off.
For this study, the test was simplified using a tapered, plastic cup. The Karsten pipe
test was selected as a straightforward method to determine the porosity of the block
material and its resistance to water penetration. This is an important factor for the
Freeman House, since the existing block wall readily allows water to penetrate to the
interior. Compression testing was completed using test cylinders that were cast from
the same mixtures used during the manufacturing of new blocks discussed in Chapter
5. These cylinders were tested in an off-site laboratory that is highly experienced in
materials testing. The goal of the compression test was to establish the long term
durability of a replacement block, especially in response to seismic forces. Finally,
a color comparison and texture analysis was done to evaluate the compatibility of
152
the newly manufactured blocks with the existing blocks from the house. This test
attempted to establish whether there was color and texture consistency between
original Freeman House blocks and newly manufactured blocks.
6.2 Slump testing
As described in Chapter 4, the slump test is a standardized concrete test method
which measures the distance that a concrete mixture slumps when an ASTM C 143
metal test cone is filled and then lifted off from the test mix. It was decided that the
Preparing the informal slump test. Figure 6.1:
1
test should be included for the Freeman House mix material in a simplified method
to determine the placeability of the mixture and possibly draw some conclusions
about what the slump of the material means for the future lifespan of the block. As
was encountered in the block manufacturing phase of this study, the larger the batch
1 All figures in this chapter are by the author unless otherwise noted.
153
size during mixing, the greater it will slump. François de Larrard notes that, “There
is often an increase in slump when a mixture designed in a laboratory is produced in
a concrete plant.”
2
In the early batches that were mixed in small quantities by hand,
more water was necessary to achieve the desired consistency. Once the concrete mixer
was used for making enough mix to form two or three blocks, a proportional amount
of water as was used in the smaller batches produced a mix that was far too wet and
had to be left in the formwork overnight. Once it was realized that the water ratio did
not scale linearly with an increase in batch size, the correct slump was again achieved.
As expected before the test, the slump of the material was found to be essentially zero
when the mixture met the requirement set out in Frank Lloyd Wright’s specifications
that it would hold its shape when squeezed in the worker’s hand.
Completed slump test showing virtually zero slump of the material. Figure 6.2:
This result shows that the material is what the concrete industry refers to as a no-
slump mix, which is desirable for precast concrete that requires immediate form
2 LARRARD, F . D. 1999. Concrete Mixture Proportioning: A scientific approach. London; New York:
E & FN Spon; Routledge, p.252.
154
removal. However, this no-slump mix is also a contributing factor in the difficulty
experienced when attempting to form the pattern-face blocks. The mixture has a
tendency to compress in a downward direction but resists moving up into the cavities
of the pattern-face mold. In Figure 6.2, it is clear that the mixture did not slump
and actually retained the form of the cup, including the indentation rings around
the top and bottom. Another important consideration that Larrard addresses in his
book is the role that time and temperature play in affecting material slump. From the
time that water touches the cement, slump begins to decrease. This must be taken
into account when designing the mixture so that there is sufficient workability in the
time between when the mixture is prepared and the time when the mixture will be
formed. Also, higher heat in summer can cause a greater decrease in slump as the
moisture content evaporates from the mixture to some extent. Therefore, the design
of a successful block mixture must take into account three important and interrelated
factors: batch size, amount of time in which a batch will be used, and the climatic
conditions during batch preparation.
6.3 Karsten pipe testing
The blocks made in this study were evaluated using the Karsten pipe test to determine
how fast they allow water to transmit from one side to the other. Blocks made to
replicate the existing mix of two parts sand to one part cement along with blocks that
use different concretes, additives and sealants were evaluated for their abilities to resist
water penetration. Also, a standard 8” x 8” x 16” concrete masonry unit (CMU) was
155
tested to as a comparison between the industry standard for concrete block masonry
and the innovative block design developed by Frank Lloyd Wright. The Karsten pipe
and the clay putty used to attach it to the block surface are shown in Figure 6.3. It
was essential when attaching the apparatus that there was a firm seal around the
base of the pipe. Otherwise, water would leak out when the pipe was filled. This
happened several times during testing. Also important to obtaining accurate results
was to ensure that air bubbles were not created in the pipe base when pouring the
water into it. Water had to be added slowly to prevent water bubbles from forming.
The first attempt at testing Block 1 was flawed because of a bubble trapped in the
base of the Karsten pipe and this prevented the water from penetrating into the block.
When the bubble was removed, the water was able to flow into the block.
Clay putty and Karsten pipe used for this testing. Figure 6.3:
Blocks being tested were checked at two minutes, five minutes, and ten minutes
to see how much water had been drained from the Karsten pipe. Any problems or
156
inconsistencies for each test were noted. Generally, the pattern-face blocks made in
the manufacturing step were not tested with the Karsten pipe due to the difficulty in
attaching the pipe to the patterned surface. Some pattern-face blocks were tested in
the small square indentation of the pattern, but great difficulty was experienced in
trying to adhere the putty to the block successfully in this small area. Also, the blocks
that were made with perforations in their pattern were not tested with the Karsten
pipe. The results of the Karsten pipe testing recorded in this study are provided below
in Table 6.1.
Block
Number
Loss at 2
minutes
Loss at 5
minutes
Loss at 10
minutes
Notes
1 1/2 mL 3/4 mL 1 mL First try at this test gave zero loss due
to a bubble in the pipe
2 1/2 mL 1 mL 1 1/2 mL Wet mix
3 near 0 loss 1/4 mL 1/2 mL Wet mix
4 - - - Absorbed water too fast to take a
measurement
5 - - - Absorbed water too fast to take a
measurement
6 - - - Absorbed water too fast to take a
measurement
7 - - - Absorbed water too fast to take a
measurement
8 - - - Absorbed water too fast to take a
measurement
9 - - - Absorbed water too fast to take a
measurement
10 - - - Absorbed water too fast to take a
measurement
11 - - - Absorbed water too fast to take a
measurement
Table 6.1: Karsten pipe test results
157
Block
Number
Loss at 2
minutes
Loss at 5
minutes
Loss at 10
minutes
Notes
12 - - - Absorbed water too fast to take a
measurement
13 - - - Absorbed water too fast to take a
measurement
14 0 0 1/4 mL Block difficult to adhere to; no
immediate absorption like others
15 - - - Tested in 2 sites; even worse than
untreated block; absorbed water
much too fast to take a measurement
16 - - - Absorbed water too fast to take a
measurement
17 - - - Absorbed water too fast to take a
measurement
18 - - - Absorbed water too fast to take a
measurement; absorbed slower than
most at approx. 1/2 mL per second.
19 - - - Absorbed water too fast to take a
measurement
20 - - - Absorbed water too fast to take a
measurement
21 - - - Absorbed water too fast to take a
measurement; absorbed slower than
most at approx. 1/2 mL per second;
tested in the small recessed square of
the pattern.
22 Absorbed water too fast to take a
measurement; absorbed slower than
most at approx. 1/2 mL per second;
tested in the small recessed square of
the pattern.
30 - - - Absorbed at about 1/10 mL per
second.
31 1/2 mL 1 mL 2 1/4 mL Wet mix; no evidence of leaking
through back side
Table 6.1 (continued): Karsten pipe test results
158
Block
Number
Loss at 2
minutes
Loss at 5
minutes
Loss at 10
minutes
Notes
36 <1/4 mL <1/2 mL <3/4 mL
38 1/4 mL 1/2 mL 3/4 mL
39 ~0 mL 1/4 mL 1/2 mL
8” x
8”x 16”
CMU
4 mL 5 1/2 mL N/A Absorbed water quickly but has
a cavity where water can drain
out without entering the interior
envelope.
It is readily apparent by examining Table 6.1 that most of the blocks absorbed the
water too quickly to allow measurement. At the time of testing this was surprising,
but upon further consideration, it was realized that the blocks that performed well
used a wet mixture. Because this testing was ongoing throughout the manufacturing
process, it was possible to adjust the manufacturing technique to develop the hybrid
block, capitalizing on the water-resisting characteristic of the wet mixture.
6.4 Compression testing of cylinders
While there is not minimum requirement for compressive strength of concrete in
general, it seems like an important consideration at the Freeman House for blocks
that were originally made as structural wall elements. If Wright were building his
textile block designs today, he might have specified a higher compressive strength
to support the weight of the blocks above and resist compression from thermal
expansion of surrounding blocks and grout..
Table 6.1 (continued): Karsten pipe test results
159
In order to carry out testing for the first two samples, two 6” inside diameter by 13”
tall PVC tubes were obtained, and filled with a sample mixture. The standard size for
this test is either a 6” diameter x 12” tall cylinder or a 4” diameter x 8” tall cylinder.
In sample tube S1, the mixture was poured in to the top at 13” and then compressed
with the hydraulic hammer press using a plug cut from a wood block. Then more
material was added and recompressed until the cylinder was filled to a height of
12”. For S2, the mix was placed inside the tube up to the top at 13”. Then it was
compressed by hand to 12”. Both test sample tubes were sprayed with water at the
top to moisten the mixture, however, there was likely little beneficial effect achieved in
this step since the water would not likely reach the bottom of the test sample. These
two tubes were left to cure for 42 before they were tested. Sample S1, with a test psi
of 1,145, performed astoundingly well compared to S2.
Placing the concrete cylinder sample in the testing apparatus. Figure 6.4:
160
The next seven samples were cast again in PVC pipes, but this time the inside
diameter used was four inches and the height was reduced to eight inches. This
alteration to the sample size had been suggested during the first round of testing by
the testing engineer at Smith-Emery Laboratories, who had helped to test the first two
samples. The reasoning he gave for this change was that the size of the aggregate, in
this case sand, was too small in comparison to the cross sectional area of the sample.
This was likely a wise choice also in consideration of the block dimensions. Nowhere
on the block is there a 6” x 12” section that can be found. The block is less than 4”
thick on the edges so a 4” diameter test sample was ideal for evaluating the textile
blocks.
Test samples that were cast in PVC tubes. Figure 6.5:
In order to test the samples, the concrete cylinders were capped with a sulfur cement
material as seen in Figure 6. This capping step ensures that the ends are level and can
compensate for a cylinder that is slightly shorter than the 8” or 12” height. Testing
was carried out on a compression testing stand and the measurements of load on the
test cylinder were read by the testing engineer on a scale attached to the machine.
161
Test cylinders made ready with a sulfur cement cap on both ends. Figure 6.6:
Cylinder S4 was the highest performing of the second set, with cylinder S3 close
behind. Cylinder S6 was cast dry but then water was added to the top of the PVC
tube. In testing, the sample crumbled at the bottom, showing that the water did
not permeate all the way through the cylinder. Cylinders 23, 24 and 25 were all
very close in test psi values, despite being compacted by two different methods. The
complete test results are presented in Table 6.2. Videos of the testing process can be
found in Appendix F .
Date of Test Cylinder
number
Age at
test
Load Test PSI Compaction
Method
December 12, 2008 S1 42 days 32,400 1,145 Machine
December 12, 2008 S2 42 days 3,700 130 Manual
February 13, 2009 S3 86 days 36,450 2900 Machine
February 13, 2009 S4 86 days 41,050 3,270 Machine
February 13, 2009 S5 86 days 8,600 680 Manual
February 13, 2009 S6 86 days 5,750 460 Manual
February 13, 2009 23 36 days 21,900 1,740 Manual
February 13, 2009 24 36 days 19,850 1,580 Manual
February 13, 2009 25 36 days 22,050 1,750 Machine
Table 6.2: Compression test results.
162
Test cylinder S1 before (left), during (center), and after testing (right). Figure 6.7:
Test Figure 6.8: cylinder S2 during (left) and after testing (right).
163
Test cylinder S3 after testing. Figure 6.9:
Test cylinder S4 during (left) and after testing (right). Figure 6.10:
164
Test cylinder 23 before (left) and after testing (right). Figure 6.11:
Test cylinder 24 before (left) and after testing (right). Figure 6.12:
165
6.5 Color and texture comparison
The premise of this test was to ensure that newly manufactured replacement blocks
would be consistent with the existing historic character of the blocks at the Freeman
House. This comparison would be straightforward if all the blocks looked the same,
however, it is made more difficult by the reality that the original textile blocks were
made at different times, by different crews who used different grades of ingredients.
This results in color and texture variations all over the house. Therefore, not even all
the existing blocks match each other in color and mix consistency.
The benefit of having blocks that are so diversified in color and texture is that the
task of matching was made more simple. As long as the replacement block fit within
a range of colors that were consistent with the existing Freeman blocks, then it was
considered to be acceptable. Essentially, the key during block manufacturing is to
ensure that the color of the blocks being produced varies. Any new blocks might
match a certain batch of the existing blocks, but may be noticeably different from
other existing blocks in the wall. This is naturally expected and this was what Wright
had intended for the facade to a certain degree.
If the replacement mix were perfected so that all the blocks made looked exactly
the same, then it could be said that these would be readily distinguishable from the
existing blocks because they would stand out as significantly different. This might
also create a spotted effect on the facade when viewed from a distance. The goal
of this test was not to definitively answer this question, but to provide a sampling
166
of the possible color and texture comparisons that occur. Utilizing the Munsell
ColorChecker® card placed in the photographs of newly manufactured and original
Freeman House blocks, pixel histograms of the block faces confirm that, in most
cases, blocks being manufactured today are consistent with the color of blocks made
85 years ago. Below are a series of photos that compare existing blocks and newly
manufactured blocks.
Original pattern-face Freeman House block (no number). Figure 6.13:
Original pattern-face Freeman House block (370 B-8). Figure 6.14:
167
Original flat-face Freeman House block (55 B-6). Figure 6.15:
Original flat-face Freeman House block (no number). Figure 6.16:
168
Original Figure 6.17: block compared to new block #2.
Original block compared to new block #14. Figure 6.18:
Original block compared to new block #16. Figure 6.19:
169
Original block compared to new block #17. Figure 6.20:
Original block compared to new block #21. Figure 6.21:
Original block compared to new block #31. Figure 6.22:
170
Original block compared to new block #36, parts a and b. Figure 6.23:
Original block compared to new block #38. Figure 6.24:
Through a visual survey which compared newly manufactured blocks and original
Freeman House blocks, a trend emerged in terms of their surface textures. This
step was an essential measuring stick by which replacement block mixtures were
either accepted or denied for their success or failure with maintaining a consistency
with the texture of the original Freeman House blocks. The goal was to replicate a
rough, sandy texture, found on the original blocks, which is quite different from a
traditional smooth concrete finish. For instance, Block 31, while having a pattern
171
that is well-formed, is too smooth to be considered as a match with the historic fabric
of the Freeman House. Additionally, as seen in Figure 6.22, the color of Block 31
appears inconsistent with the range of colors found in the Freeman House samples.
This analysis, which was ongoing during the block manufacturing phase, helped
to determine the design of the mixture for hybrid blocks 38 and 39. These blocks
embraced the same friable character on their exterior, but they capitalized on the
benefits of a high water content in the concrete mixture to create a block that satisfied
multiple criteria simultaneously.
172
CHAPTER 7: ANALYSIS AND EVALUATION OF TESTING RESULTS
7.1 Block formation and testing analysis
The test results presented in Chapters 5 and 6 are evaluated here for their
interrelationships and connections that can be drawn between the manufacturing
process and the individual tests. A discussion of the implications of such connections
is given in this chapter. The results of the testing have dispelled previous theories and
brought about new questions that will need to be addressed in future research efforts.
In Table 7.1 below, each of the blocks manufactured in this study are classified by
their ingredients and manufacturing process along with the criteria by which the
block samples were evaluated during testing. This chart is a straightforward method
to readily understand the characteristics that elicit a high-performing block versus
those that produce a porous or weak block. All the additives and sealants that
were tried seemed to have no effect on the ability of a block sample to resist water
penetration during the Karsten pipe test. An examination of trends in this chart
shows that there is a direct relationship between compaction method and strength.
The machine compressed blocks have a higher compressive strength than those that
are manually compacted. Because not every block could be tested for compressive
strength and water penetration, hypothetical results inferred from trends observed in
the other blocks of similar mixtures have been denoted in the chart with a (*)instead
of a (•).
173
Table 7.1: Block properties matrix
Pattern Ͳface
Flat Ͳface
Test Cylinder
Machine
Manual
Wet pour
Dry pack
Quikrete Washed Plaster Sand
Quikrete Play Sand
Quikrete Medium Sand
Colton Portland Quikrete Hydraulic Water ͲStop
Quikrete Conc.Acrylic Fortifier
Sika Acrylic Fortifier & Bonding
Sika NaturalLook Clear Sealer
Cover in plastic wrap
Quikrete Conc.Mix Accelerator
Conc.Pharmacy Flow Control
High
Low
Loss too rapid to measure
2 minutes
5 minutes
10 minutes
S1 • • • • • • *
S2 • • • • • • *
S3 • • • • • • *
S4 • • • • • • *
S5 • • • • • • *
S6 • • • • • • *
1 • • • • • • * 0.5mL 0.75mL 1mL
2 • • • • • • * 0.5mL 1mL 1.5mL
3 • • • • • * ~0mL 0.25mL 0.5mL
4 • • • • • • * •
5 • • • • • • * •
6 • • • • • • * •
7 • • • • • • * •
8 • • • • • • * •
9 • • • • • • • * •
10 • • • • • • * •
11 • • • • • • • * •
Additives/Treatments Test P.S.I. Water Penetration
Block Type
Block Number
Compaction Method
Mix Consistency
Aggregate Cement
174
Table 7.1 (continued): Block properties matrix
Pattern Ͳface
Flat Ͳface
Test Cylinder
Machine
Manual
Wet pour
Dry pack
Quikrete Washed Plaster Sand
Quikrete Play Sand
Quikrete Medium Sand
Colton Portland Quikrete Hydraulic Water ͲStop
Quikrete Conc.Acrylic Fortifier
Sika Acrylic Fortifier & Bonding
Sika NaturalLook Clear Sealer
Cover inplastic wrap
Quikrete Conc.Mix Accelerator
Conc.Pharmacy Flow Control
High
Low
Loss too rapid to measure
2 minutes
5 minutes
10 minutes
Additives/Treatments Test P.S.I. Water Penetration
Block Type
Block Number
Compaction Method
Mix Consistency
Aggregate Cement
12 • • • • • * •
13 • • • • • • * •
14 • • • • • * 0mL 0mL 0.25mL
15 • • • • • • • * •
16 • • • • • • * •
17 • • • • • • * •
18 • • • • • • * •
19 • • • • • • * •
20 • • • • • • * •
21 • • • • • * •
22 • • • • • * •
23 • • • • • • *
24 • • • • • • • *
25 • • • • • • *
26 • • • • • * *
27 • • • • • * *
28 • • • • • * *
175
Table 7.1 (continued): Block properties matrix
Pattern Ͳface
Flat Ͳface
Test Cylinder
Machine
Manual
Wet pour
Dry pack
Quikrete Washed Plaster Sand
Quikrete Play Sand
Quikrete Medium Sand
Colton Portland Quikrete Hydraulic Water ͲStop
Quikrete Conc.Acrylic Fortifier
Sika Acrylic Fortifier & Bonding
Sika NaturalLook Clear Sealer
Cover inplastic wrap
Quikrete Conc.Mix Accelerator
Conc.Pharmacy Flow Control
High
Low
Loss too rapid to measure
2 minutes
5 minutes
10 minutes
Additives/Treatments Test P.S.I. Water Penetration
Block Type
Block Number
Compaction Method
Mix Consistency
Aggregate Cement
29 • • • • • * *
30 • • • • • * •
31 • • • • • * 0.5mL 1mL 2.25mL
32 • • • • • • *
33 • • • • • • • *
34 • • • • • • • *
35 • • • • • • • *
36a • • • • • • <0.25mL <0.5mL <0.75mL
36b • • • • • • • *
37 • • • • • • • *
38a • • • • • • • 0.25mL 0.5mL 0.75mL
38b • • • • • •
39a • • • • • • • ~0mL 0.25mL 0.5mL
39b • • • • • •
40 • • • • • • • *
41 • • • • • • • *
CMU • • • * 4mL 5.5mL n/a
176
The sections that follow discuss each of the findings during the course of this thesis
research in greater detail. Possible reasons for the block performance characteristics
seen in the testing phase are discussed as well. The four groups of results encompass
the topics of density and water content of the mix, the advantages encountered in
comparing coloration, the function of slump in the mix, and the importance of
compressive strength in terms of replacement blocks.
7.2 Density, hydration, and capillary action
7.2.1 Karsten pipe testing revelations
By examining the Table 7.1, it is evident that blocks 1, 2, 3, 14, 31, 36, 38, and 39
all performed well in the Karsten pipe test based on the characteristic of having a
high water content mixture that could be poured into the mold versus the dry pack
mixture that is characteristic of the blocks found at the Freeman House. It was
discovered during the Karsten pipe testing that water easily poured through most
of the blocks manufactured in this study. This was similar to the result that Alice
Ormsbee found in her Karsten pipe testing of the Freeman House exterior facade.
1

While this was not the intended outcome of the new block compositions, there were
valuable lessons learned in the process of testing the sample block mixtures with the
Karsten pipe.
1 ORMSBEE, A. 2006. Water in the Freeman House: a study of the block in the textile block construction
assembly as the primary route for water infiltration. Thesis (M.Bu.S.). University of Southern
California.
177
First and foremost, it can be said that, based on the results of testing the blocks
utilizing the dry pack mix, the compaction of the block material inside the formwork
had little to no bearing on its ability to resist water penetration. Previously, as
expressed by Jeffrey Chusid in the Historic Structure Report
2
, it was believed that the
manual compaction of the blocks was substantially responsible for their porosity. It
was thought that the manual compaction did not sufficiently eliminate gaps between
aggregate particles, thus a high level of water infiltration was permitted to pass to the
interior of the home. Based on the test results presented in Table 7.1, it is clear that
this is simply not true. In fact, it is possible that capillary action is allowing water to
pass though a highly compressed block even faster. The major revelation that can be
taken away from the Karsten pipe testing is that even 16,000 pounds per square inch
of compaction pressure still does not yield a block that is impervious or even resistant
to water penetration. Yet there were blocks that performed well in the initial Karsten
pipe testing. The difference for these blocks was in the water content of their mixture.
7.2.2 Dry versus wet
According to François de Larrard, the slump test, which measures the wetness of
a concrete mixture “. . . is related to the yield stress, which is a more fundamental
property. However, it reflects only indirectly a most crucial aspect of fresh concrete:
its placeability. Generally speaking, a higher slump means an easier placement but a
2 CHUSID, J. M. 1989. Historic Structure Report: Samuel and Harriet Freeman House, Hollywood,
California, Frank Lloyd Wright, 1924. Los Angeles: School of Architecture, University of Southern
California.
178
more expensive concrete for a given strength.”
3
Following from this statement, it is
worth noting that the blocks that seemed to be mistakes during the manufacturing
process, in that the mixture was too wet when it was cast in the form, actually
performed better in the Karsten pipe test. Blocks 1, 2, and 3, cast with a wet mix,
were even better than the blocks that had received a sealant after curing. The question
became what function the water in the mixture was performing to seal out water after
the block had cured.
The reason behind this phenomenon linked to the above finding about the
compaction pressure and its bearing on the porosity. While the particles may be
tightly packed together under 16,000 psi, the Portland cement in the mixture is
not sufficiently hydrated to create a paste between the particles of aggregate. Even
though the block will hold itself together marginally well during the transfer to the
curing rack and can cure successfully, it has tiny holes between the particles where the
cement did not bind the sand together. While there is some level of paste binding
that is occurring, there are still a sufficient number of pores created in the block that
allow water to penetrate into the block.
Similar to Ormsbee’s testing
4
, a low water content block that had been tested with the
Karsten pipe and when it was turned over to view the back side, it was seen that water
was draining through the block. The water from the Karsten pipe was also able to
3 LARRARD, F . D. 1999. Concrete Mixture Proportioning: A scientific approach. London; New York:
E & FN Spon; Routledge, p.251.
4 ORMSBEE, A. 2006.
179
move in a horizontal direction through the block, sometimes reaching the side of the
block, which if it was actually in the facade of the house would be coming in contact
with the grout between the blocks. If this grout was missing or cracked as it has been
found to be in many areas, the steel reinforcing would be exposed to water.
Only when the mixture is sufficiently wet are the particles able to gel together and the
cement is able to form a paste that binds the aggregate, in a way suspending the sand
particles in a paste to create a water repellent bond. The major problem with the wet
mixture for this research scope was that the blocks look different when made with
the high water content mixture. They are not consistent with the color nor with the
texture of the original blocks found at the Freeman House. The next question that
arose from these findings was how to successfully create a wet block but maintain the
character of the dry blocks.
7.2.3 The effect of Karsten pipe testing on mixture designs
The findings of the initial rounds of Karsten pipe testing, which were contrary to
the initial thinking about manufacturing replacement blocks, prompted the design
of a mixture that would incorporate the benefits of a wet mixture, yet maintain the
historical character and texture of the dry pack mix that was specified by Frank Lloyd
Wright. Mixtures 36, 38 and 39 all attempted to find the right balance between wet
and dry by using two separate mixtures to create one block.
180
The idea of combining the two mixtures, wet and dry, came first in the form of Block
36. The form was filled initially with a wet mixture until the mixture reached just
above the side channel molds and was then smoothed out flat for curing. It was left
covered by a flat steel plate in the form box for four days. After removal from the
form, which proved more difficult than expected because of the way the wet concrete
shrunk around the form, a batch of dry-pack concrete mix was prepared. This mix
was packed into the pattern face mold as on the table top and then the block backing
was lowered on top and compressed. A problem arose initially with actually adding
the topping slab to the block backing. The pattern-face mold has portions that
Block 36 after the addition of the pattern-face topping. Figure 7.1:
5
actually push deep into the form box, leaving approximately 1/4” to 3/8” of material
between the block surface and the plate on the underside. Also, the pattern mold
sides were not as tall as the recessed areas, causing material to slide off the edges and
5 All figures in this chapter are by the author unless otherwise noted.
181
ultimately leave a missing lower portion all around the pattern face as it sat on top of
the blank.
Once the patterned topping had cured and the pattern mold was removed, the block
was generally considered a success, however the recessed areas of the pattern showed
through as a different color and texture since they were cast using the wet mix. Also it
was noted that even though the wet mixture block was superior in its forming ability
and cohesion, it was still highly susceptible to cracking at the corners and along the
channel edges.
Side view of Block 36 showing a dry mixture topping added to a cured Figure 7.2:
wet mix block.
The major failure of this hybrid block was discovered after several weeks when
the block was being moved from the curing rack to be photographed. It was not
realized during the manufacturing of this block that a cured concrete block will not
successfully bind with a new batch of concrete. Thus, the topping that was added to
36a slid off once the block was moved from the curing rack. It was evident with this
182
failure that this technique would not be a viable solution for making replacement
blocks.
The solution to this problem was to create the block again with two mixtures, but to
combine them before one was able to cure, which would have to be done in the face-
down orientation. Figure 7.3 shows the method that was used to cast these blocks.
The first layer, the dry mixture, is in orange and the second layer, the wet mixture,
Diagram of the hybrid block casting method for Blocks 38 and 39. Figure 7.3:
is in blue. The dry pack mix was made using the two part sand to one part cement
ratio plus two boxes of concrete mix accelerator. This was placed onto the pattern or
flat-face mold as it sat face up in the form box. In the case of block 38, the patterned
top mold was used and for 39 the flat-faced mold was used. The intention behind
using the accelerator was to get the mix to set quickly so that the entire block could be
flipped over and transferred to the drying rack after the wet backing began to set. It
was found, however, that the accelerator did not set up as quickly as was expected or
promised by the manufacturer, thus requiring the blocks to sit face down in the form
boxes overnight before they could be removed successfully.
While these top faces were beginning to cure, a new mix was made using more water
(18 cups versus 10 in the topping mix) and again adding two boxes of concrete
STEP 1 STEP 2
183
accelerator. In the case of block 39, the result of leaving the block face-down in
the form was a much more gray look to the face since the water likely seeped to the
bottom of the form and hydrated the cement more completely. In the case of block
38, there was some adhesion of the mixture material to the patterned mold, but the
pattern was still clearly formed and without major defect.
Blocks 38 and 39 both used this hybrid manufacturing system with excellent results.
When both blocks were tested with the Karsten pipe, they performed on par with
the blocks that were cast with a wet mix in the beginning of the block manufacturing
phase. In both blocks, less than 1 mL was lost after 10 minutes. When tested prior
to its demise, block 36 also had similarly favorable test results. This was a major
improvement over all the other blocks that used a dry mix, despite any additives or
coatings that may have been tried. Even more encouraging was the appearance of the
exterior. Instead of having a smooth finish, the exterior was textured, like the blocks
found at the Freeman House.
Block 38 front, side, and back faces. Figure 7.4:
184
This evidence seems to establish the importance of having some element of the
block be cast with a wet mixture to resist water penetration through the block. The
downside of the coatings on the dry pack mixes was that even though a thorough
coating was made and even reapplied after drying in some cases, the coating was no
match for the lack of cohesion between the particles. This may perhaps be where a
consolidation treatment, as described in Chapter 3, would be appropriate. However,
it still may not overcome the lack of a paste between the aggregate sand particles.
7.3 Color comparison findings
The major outcome of the color comparisons is that the Freeman House has sufficient
color variation that can be found in its blocks that most blocks made in this study
would be viable candidates, in terms of their color, to be replacement blocks.
The variety in the Freeman House blocks was due to the fact that different crews
worked on the house over the course of its construction. Also, it has been found in
documents that Wright wrote after the construction of the textile block houses that
he intended to have variations in the color of the block. Therefore, a successfully
compatible replacement block would not be a single, consistent mixture. Instead,
there would need to be several mixtures or a mixture that ensures different colorations
to adequately replicate the variety of coloration and texture found throughout the
house.
This seems like it may create an exhausting task for the person who makes the
replacement blocks. In order to determine whether the blocks made in this study
185
would actually be visually compatible with the current Freeman House block
walls, a visual comparison was made between a sampling of blocks from the house
and a sample set of newly manufactured blocks made during this research. In the
qualitative visual comparison, it was found that almost none of the blocks exactly
matched the sample set of blocks from the Freeman House. However, some blocks
seemed to be within a range that could be considered as consistent with the existing
blocks. These were blocks 2, 12, 14, 16, 17, 21, 27, 38, and 39. It should be noted
here that even between the blocks made in this study using the same ingredients
and proportions, there were wide variations in coloration and texture. Thus it is not
surprising that the house exhibits such wide variations in color. It seems likely, then,
that the new blocks will have some consistency to the color variation found in the
original textile blocks, despite any color variations form one block to the next.
An original Freeman House block compared to Block 31. Figure 7.5:
186
As a more qualitative approach, the quantitative comparison of blocks using pixel
histograms was completed to visually capture and compare data that was obtained
in the block photographs. The pixel histograms show the number and brightness
of pixels in each of three ranges - red, green and blue. Based on the evaluation of
Blocks 2, 14, 16, 17, 21, 31, 36a, 36b, and 38, it can be said that most of the blocks
are within a conceivably acceptable range to be considered viable replacement blocks
solely on the basis of their color. The blocks that stood out as different were 31,
and 36a. The blocks all use the same ingredients as, or match as closely as possible,
the original blocks, so it is not surprising that the coloration was so consistently
acceptable. The important consideration for making replacement blocks in the future
will be to ensure that the blocks exhibit a similar range of colors. A mixture that
consistently produced the same color, if such a mixture could be found, might stand
out as foreign to the house more than a mixture that consistently produces variations
in color.
7.4 Slump and its effect on block making
In the slump testing of the mixture material produced during this study, it was found
that the sand and cement mixture did not slump at all. If the definition of slump is
understood as a measure of the wetness of a mixture, then it is clear that the mixture
used for these blocks was very dry. Any slump it did have was too small to measure.
This sheds some light on the problems experienced with actually manufacturing the
blocks with the patterned face. If the material has a low tendency to move in a plastic
187
manner, then the material cannot be expected to successfully move when under
pressure and fill the entirety of the pattern. Because of this, it was only when the
blocks were cast in a face-down position that the pattern would come out properly
formed. Every other attempt to press the pattern into the mixture failed by not being
able to push deep enough into a mixture that had a very low plasticity.
There is possibly some promise in an additive that would make the material flow more
readily. The Flow Control additive attempted in the mixtures for Blocks 40 and 41
did not seem to drastically improve the plasticity of the material. Compaction was
attempted in the face-up orientation and was still unsuccessful. It was decided that
face-up compaction would test the effect of the admixture better than face-down,
since it was already known at this point that a successful pattern could be made in the
face-down orientation without the addition of a plasticizer.
The reason for the failure of this admixture may be tied to the dry nature of the
mix. The additive is designed to work with the cement in wet form. If there is not
sufficient water, the additive may be losing a great deal of its effectiveness. Further
research is needed in order to determine if an additive exists for dry mixes that would
help its plasticity and make the mixture able to move more readily inside the forms.
7.5 Importance of compressive strength
After the first set of cylindrical samples were tested at Smith-Emery Laboratories
on December 12, 2008, it was apparent that a major difference existed between the
188
manual compaction and the machine compaction methods. While both methods
still allowed water to penetrate the block, the machine compression made the block
significantly stronger in compression. This result was expected to a point, but
the actual strength results far exceeded the level of compaction improvement that
machine compaction gave over manual compaction.
While it would seem to be beneficial to have a stronger block, there are concerns
raised in the case of the existing, historic blocks that call into question the use of
a stronger block. Typical preservation guidelines hold that any repair or patching
material should be weaker than the existing fabric of the building so that it can
be easily removed at a later time and, in an earthquake, the repair material would
fail first, preventing excessive damage to the historic fabric.
6
Another concern
with a more compact, stronger block is that the compact block would need more
material, making it heavier and putting more weight on the structure. This may
not be a concern at the Freeman House now that the seismic retrofitting has been
implemented to support the building, however, it could still present a problem with
how the walls would flex in an earthquake. Additionally, the machine compacted
block has the potential to expand and contract differently than its neighbors, which
could further the damage to the surrounding textile blocks.
7
Once again, damage to
6 Information obtained through e-mail communication on 31 October 2008 with Peyton Hall,
a historic architect with Historic Resources Group in Hollywood, California. Hall has extensive
preservation work experience, especially in Southern California.
7 These two points were advanced to the author by Beril Bicer-Simsir, Assistant Scientist and concrete
expert at The Getty Museum Science Department, during a phone conversation on 1 December
2008.
189
the historic fabric is to be avoided as a first priority. It would be more preferable to
have to change out weaker replacement blocks on a continual basis than to lose many
more original blocks due to damage from stronger replacements.
So the question remains as to what the best mode of manufacture would be. This
question is one of the most important to answer before replacement blocks are
manufactured and installed at the house because it encompasses so many other issues.
For instance, if it is decided that the blocks should use the same mixture as originally
specified by Wright and should be compacted just as the workers did in 1924, then
there is still a problem with water infiltration. This could possibly be solved by a
heavy sealant, but a sealant on the exterior could undesirably change the appearance
of the block, creating a glossy sheen or altering the coloration. A sealant that would
be applied to the back side of the block before installation would be ineffective in
preventing water infiltration into the home because it would not solve the problem
of all the existing blocks around it allowing water to penetrate the wall. Further
investigation of epoxy additives and non-marring sealants available commercially may
reveal a product that is ideal to this application. The key to the water infiltration issue
is that the whole house must be treated at one time. Otherwise, water will find a
route inside through the path of least resistance, whether it be an unsealed block or a
leak in the roof.
190
7.6 Overall evaluation
On an overall basis, the testing in this thesis was considered a success. A great deal of
new knowledge about the block composition was revealed and the characteristics of
wet mixtures versus dry mixtures were defined clearly for Freeman House replacement
blocks. Further research will undoubtedly be necessary before replacement blocks can
actually be installed at the Freeman House. The development of a prototype hybrid
replacement block in this thesis gives a definite direction that future researchers can
follow. The final replacement must consider all the issues discussed here in order to
be compatible with the existing blocks and improve the performance of the house.
191
CHAPTER 8: Con Clusions
This research effort set out to create a replacement block mixture for the Freeman
House that would outperform the existing blocks in terms of its resistance to water
and water vapor infiltration and its strength, yet maintain the textural character
and coloration of the existing blocks. In Chapter 1, it was hypothesized that, if a
replacement block is made to repair the existing walls, then it will need to utilize
modern-day mixture materials and techniques to create a new block that is both
compatible with the existing structure and superior in terms of its water resistance and
durability. With the development of a prototype replacement block using a hybrid
of wet and dry mixtures, there were important findings made in the course of the
research that will likely guide further research efforts.
8.1 Block manufacturing
In terms of the difficulty encountered with manufacturing the pattern-face block
in a face-up orientation, it might be concluded that the original construction crew
either had better suited materials or that their experience far exceeded that which can
be acquired in making a few batches of blocks. While these possibilities exist, they
seems unlikely. With some certainty, based on the petrographic analysis by American
Petrographic Services, it can be said that the right mixture was used for this study.
The only variable is the water content. With a range of water contents tried between
Blocks 1 and 41, it also seems unlikely that the water content was not right in any of
the mixtures. According to Jeffrey Chusid, it is possible that pattern-face blocks were
192
cast face-down. The two original pattern-face blocks used for the color comparisons
in this study were completely filled on the back side. The absence of the coffered
back suggests that these may have been cast face-down. The best solution may be to
develop a device that could flip the form box over after the casting is done face-down.
This way, more than one block per form box could be cast in a single day.
8.2 Cement hydration
The second major conclusion drawn from this research is that water permeability
through the block is a function of the hydration of the cement. The more wet the
mixture, the better the cement is able to form a paste between the particles of sand.
This was important in dispelling previous theories that the manual compaction of the
blocks was primarily responsible for their permeability. Following from this, it was
found that a hybrid block could be cast to obtain the benefits of a wet block while
maintaining the look and texturing of the original blocks. This was a major success
because it can now lead future research in the direction of a real solution to the
problem of water infiltration. It is likely that if Lloyd Wright had made this discovery,
he would have included it in his proposal of variations to the textile block system
that he sent to his father in 1931. The key advantage of this hybrid block is that
it does not utilize any kind of sealant or coating that would have to be continually
maintained. Rather, it capitalizes on the ability of fully hydrated concrete to resist
water penetration. Further exploration along this path is likely to yield even more
promising results in the search for a replacement block.
193
While the hybrid block did not exactly match the two sample existing blocks in terms
of its color, it can be said that the color was near to the existing and within what
would be considered an acceptable range if the blocks on the entirety of the house are
considered. The inconsistency of the coloration among all the blocks on the house
makes the task of comparing color all the more simple. The essential consideration
here is to decide what the best approach is for the Freeman House in terms of
designing a concrete mixture for replacement blocks. This is probably not a right or
wrong answer, but a compromise among those who have a say in the rehabilitation
of the house. It must be decided whether one mix is used for all the replacement
blocks based on some kind of average color or whether a handful of mixes are used
to blend in with a range of block colors found around the house. Research into the
pixel histograms associated with each block can guide this decision, but ultimately
this is a decision based on cost and time that will be made by those in control of the
house. It is the recommendation of this research that replacement block mixtures
be intentionally designed to vary the coloration so that new blocks do not stand out
from the existing blocks to the point that they detract from the overall appearance of
the house.
8.3 Making repairs to the house
The greatest concern for any repair or replacement work done at the Freeman House
is to do the repairs right the first time. It would not be prudent to make decisions
about repairs that create a benefit in the short term but compromise the character
194
of the historic fabric in the long term. For instance, a replacement block made too
strong that could end up damaging existing original blocks would be an unwise
choice for repairing the house. Likewise, a sealant applied over all the exterior blocks
to stop water penetration, which might change the color of the blocks or give the
blocks a glossy sheen that was not intended by the architect, would be detrimental to
the visual character of this historic home.
The most important conclusion that can be drawn from this research is that
rehabilitating the Freeman House cannot happen overnight and will involve extensive
testing of materials and methods of construction for both the individual blocks and
the textile block system as a whole. Finding an optimized replacement block solution
will only come by way of a thorough exploration of available admixtures and an
investigation of innovative methods for casting the blocks. Only when such research
is completed and its results are understood can a viable solution be implemented
to repair the exterior of the Freeman House and prevent water from entering the
structure.
195
CHAPTER 9: Fu Tu RE Wo Rk
This chapter provides descriptions of research work that would complement the
research presented in this thesis. The investigations listed are necessary next steps in
the determination of the best way to create and replace blocks for the Freeman House.
If a successful rehabilitation of the house is to be achieved, these questions will need
to be answered in order to move forward in confidence that the steps taken during
repair and restoration work will not further damage or put at risk the home’s historic
fabric.
9.1 Performance of a replacement block next to an existing block
Based on the Karsten pipe test findings in this study and the discussion of the results
in Chapter 7, it would follow that a block made using a wet and dry combination
mixture would need to be evaluated for its ability to exist next to an existing dry pack
block without damaging the existing block. The possibly has been suggested that
damage could occur through differential thermal expansion or during seismic activity.
These theories need to be tested to determine if casting blocks with a wet backing
would cause them to no longer be a viable option as a replacement block due to the
block’s thermal expansion rate and performance in an earthquake. A full-scale mock-
up of the block wall would likely be necessary to evaluate these questions and would
involve the use of a shake table for replicating seismic forces.
196
9.2 Testing of other additives or sealants
Following from the above investigation, it may be necessary to test other commercially
available admixtures or sealants that could be either included in the mixture or
applied after the curing of a block. It would be ideal to partner with an individual or
firm that is involved in the concrete construction field and has a thorough knowledge
of and experience with concrete admixtures. One of the limitations in this thesis was
the use of only retail consumer type products for concrete.
Commercially available products for the concrete industry cover a much wider range
of possibilities and an exploration of these products for use in producing replacement
Freeman House blocks may prove to be productive. Products available from Prosoco
such as the Sure Klean Weather Seal H40 are said to have deep penetration that
provides a long period of protection before reapplication would be necessary. It is
also supposed to strengthen and protect deteriorating surfaces in brick, natural stone,
terra cotta, historic concrete, stucco, and cast stone. Admixtures from Sika are said to
be suited to dry-pack mixtures and help to plasticize the material for use in casting.
Sikamix® AE-1 High performance Compaction Aid and Efflorescence Reducing
Admixture is one such product that might help to form the pattern-face blocks more
easily.
While this thesis research has explored the topic of creating replacement blocks, there
are a majority of blocks on the house that are in satisfactory condition that they do
not require replacement, but do require some kind of waterproof or water repellent
197
coating to resist water infiltration to the interior of the home. Several options exist
and have been recommended in various reports from the consultants listed in Chapter
2. New products undoubtedly exist and would need to be investigated for their
effectiveness and degree of change they impart on the block coloration and finish.
Many of these products have been known to darken the substrate and give it a glossy
sheen. These results would be undesirable for the Freeman house.
9.3 Innovative ways to form a block
As discussed in Chapter 1, the company which made reproduction blocks for the
Ennis house restoration project, Moonlight Molds, used a series of fiberglass molds
to cast the various blocks that were needed. While some may argue that blocks
should be made the way Frank Lloyd Wright intended it to be done, the availability
of modern forming materials and methods and the cost factor associated with such
a slow construction method make it attractive to explore other options for forming
replacement blocks at the Freeman House. Beyond fiberglass molds, there may
be other opportunities such as injection molding or 3D printing using sand and a
colorant to replicate the block color. Also, the use of alternative form releases should
be investigated. Likely, WD-40 is not the best form release available for making the
blocks and the formability of the mixture may be improved by the implementation of
an effective form release. Finally, there may be a way to develop a device that could
flip the form box when full to allow face-down compaction for multiple blocks in a
198
single day. Alternatively, the form box could be redesigned to capture the back plate
and top plate so that it can be turned over without separating.
9.4 Study of the viability of the current structure in an earthquake
Because the seismic retrofitting work completed in 2001 has largely taken the role of
bearing support away from textile block walls and into cast-in-place concrete beams,
columns and walls, it would be an interesting theoretical study to find what would
happen to the current structure in a major earthquake. How the walls perform may
affect further renovations and could require additional retrofitting if it is found that
blocks have the potential to forcefully detach from the walls. It is hypothesized that
the new poured-in-place walls and columns will make the house too rigid and could
cause the existing walls to vibrate violently in an earthquake.
9.5 Method for replacing a textile block in the center of the wall
While a great deal of effort has been put into the difficult task of developing a
replacement textile block, the most challenging step will be determining a method
for inserting these replacements into the existing block walls. The problem with the
textile block system as it currently stands is that blocks are not able to be replaced
because they are locked in on all four sides by the grouting and steel reinforcement
that fills their channels. Blocks that were removed from the house had to be cut away,
destroying the channel and potentially harming adjacent blocks. Even if a damaged
block could be cleanly removed, the replacement block would have no way to slide
199
into its place because the new block would have fully formed channels, unable to
wrap around the existing reinforcing on all four sides at once. Possibly, a block that
is cast without the back half of the channels could utilize a system with spring loaded
fasteners, similar to those used in drywall where there is no stud, to anchor the block
once it is pushed into place. An exploration of what systems may be available and
how to cast the attachment into the block will need to be considered. A method for
this connection detail will need to be developed if visual continuity of the facade is to
be restored.
9.6 Method for tiling the cast-in-place concrete walls
Currently, the cast-in-place concrete walls that were added during the seismic retrofit
work in 2000-2001 are left uncovered and detract visually from the rest of this
historic home. The original intent of the retrofit was to tile these concrete walls
with one and one half inch thick block tiles to recreate the original appearance of
the house, both on the interior and exterior. The walls were sized to allow for the
application of such tiling so as not to exceed the original dimensions of the walls.
While channels were embedded into the concrete walls to allow for some kind of
attachment, the system was never resolved and nothing has been done to resolve
an attachment detail. Research into how new tiles would be attached and how the
connection would be embedded into the tiles would help to restore visual continuity.
200
9.6 Method for insulating the house
If the house is to be rehabilitated for use by the School of Architecture as either an
event space or as a house for faculty and students, it would be important to find a way
to insulate the home. Sumit Brahmbhatt already showed in his thesis that the home
does not perform well in climatic temperature swings as a high-mass structure would.
The house could greatly benefit from the addition of insulation.
1
A method for
inserting insulation, however, is more difficult. Possibly, a material could be dropped
into the wall cavity from above. This could be problematic since an even distribution
would be difficult to achieve. Ideally, a sprayed-on foam on the back side of the
exterior blocks would seal the openings between the blocks and act as a continuous
insulation barrier. The problem with this is that it would require removing the
interior wythe of blocks to apply the spray-on foam. More research into available
insulation options would have to be done to find a solution.
9.8 Methodology for determining which blocks to replace
While several evaluations of the facade conditions have been documented (Jeffrey
Chusid; Peyton Hall’s class), there has been no definitive declaration of a methodology
for when a block needs to be replaced. Some cases are quite obvious such as a missing
block or a block with significant loss to its pattern. However, some blocks may
be structurally weak, but show no signs of this weakness in their exterior face. An
1 BRAHMBHATT, S. A. 2006. The thermal energy performance study of the Freeman House. Thesis
(M.Bu.S.). University of Southern California.
201
extensive evaluation of the facades by a structural engineering consultant would need
to be done to determine how each of the blocks are performing. Additionally, this
evaluation should include someone well-versed in concrete preservation. Following
this evaluation, a set of guidelines can be established to determine when a block
qualifies for replacement and when it should be left in service and repaired.
9.9 Future work summary
The potential impacts of the proposed research presented in this chapter will be vitally
important to the continued use of the Freeman House as a rehabilitated asset for the
USC School of Architecture. The textile blocks are the major identifying feature of
the home and comprise most of its wall surface, therefore, their appearance is of the
utmost importance. The ability to replace blocks without compromising the integrity
of the structure or risking adjacent blocks would be a significant step in preserving
the home for many generations to come. While some may disagree about what the
ideal replacement block would be, it is necessary to complete this research so that a
generally accepted replacement block can be defined and its use can be implemented.
The Freeman House is still at risk. Without full attention to all these issues, the visual
character and historic fabric that define this home will continually be in danger of
further deterioration.
202
BiBliography
AGUAR, C. E. AND AGUAR, B. 2002. Wrightscapes: Frank Lloyd Wright’s Landscape
Designs. New York: McGraw-Hill.
ALOFSIN, A. 1993. Frank Lloyd Wright: The Lost Years, 1910-1922: A study of
influence. Chicago: University of Chicago Press.
AMERICAN CONCRETE INSTITUTE. 1974. ACI Manual of Concrete Practice,
Part 1: Materials and Properties of Concrete, Construction Practices and Inspection,
Pavements and Slabs. Detroit: ACI.
BOSLEY, E. R. 1992. Gamble House: Greene and Greene. London: Phaidon Press
Limited.
BRAHMBHATT, S. A. 2006. The thermal energy performance study of the Freeman
House. Thesis (M.Bu.S.). University of Southern California.
CHEEK, L. W. 2006. Frank Lloyd Wright in Arizona. T ucson: Rio Nuevo.
CHUSID, J. M. 1989. Historic Structure Report: Samuel and Harriet Freeman House,
Hollywood, California, Frank Lloyd Wright, 1924. Los Angeles: School of Architecture,
University of Southern California.
CHUSID, J. M. 1990. Frank Lloyd Wright’s Textile Block System: The Freeman
House. In: 8th Annual Association of Collegiate Schools of Architecture T echnology
Conference, February 1990 Los Angeles. Carpenters/Contractors Cooperation
Committee of Southern California in conjunction with the School of Architecture at
the University of Southern California, 13-19.
CHUSID, J. M. 2004. Modernist Threads: The Life, Death and Reconstruction of Frank
Lloyd Wright’s Freeman House. Los Angeles: unpublished.
CUMMING, E. AND KAPLAN, W. 1991. The Arts and Crafts Movement. New York:
Thames and Hudson.
DREXLER, A. 1962. The Drawings of Frank Lloyd Wright. New York: Bramhall
House.
203
Ennis-Brown House. 1992. Video. Edited by Steve DANFORTH. Hollywood: The
Wrightian Association.
Frank Lloyd Wright’s Ennis-Brown House. 1990. Video journal. Edited by Steve
DANFORTH. Hollywood: The Wrightian Association.
GEARY, M. 2004. Eric Lloyd Wright. Anne T. Kent California Room, Marin County
Free Library. Available at http://www.co.marin.ca.us/depts/lb/main/crm/oralhistories/
ericlloydwright.html. [Accessed 6 February 2009].
GREENHALGH, P . ed. 1990. Modernism in Design. London: Reaktion Books
GRIMMER, A. E., 1984. A Glossary of Historic Masonry Deterioration Problems and
Preservation Treatments. Washington, D.C.: Department of the Interior, National Park
Service.
GUARD INDUSTRIE. 2000. ProtectGuard: Oil and Water Repellent: Water absorption
test under low pressure. Montreuil Cedex, France: Guardindustrie.com. Available at
http://www.guardindustry.com/gb/produits/1-ProtectGuard/ [Accessed 22 April
2008].
HART, H. 2004. When the answers just aren’t concrete. Los Angeles Times, 26
September, p.E.42.
HEINZ, T. A. 2005. Frank Lloyd Wright: Field Guide. Evanston, IL: Northwestern
University Press.
HESS,A. 2007. Frank Lloyd Wright: Mid-Century Modern. New York: Rizzoli.
HINES, T. S. 1996. The Blessing and the Curse: The Achievement of Lloyd Wright.
In: WEINTRAUB, A. 1998. Lloyd Wright: The Architecture of Frank Lloyd Wright, Jr.
New York: Harry N. Abrams.
HIRSCHL & ADLER GALLERIES. 1989. From Architecture to Object. New York:
Hirschl & Adler Galleries.
204
INTERNATIONAL COUNCIL ON MONUMENTS AND SITES. 1996. Athens
Charter for the Restoration of Historic Monuments. Athens: ICOMOS. Available at
http://www.icomos.org/athens_charter.html. [Accessed 4 February 2009].
JIROUSEK, C. 1995. Art, Design, and Visual Thinking: The Arts and Crafts Movement.
Available at http://char.txa.cornell.edu/art/decart/artcraft/artcraft.htm. [Accessed 27
January 2009].
KAUFMANN, E. AND RAEBURN, B. eds. 1965. Frank Lloyd Wright: Writings and
Buildings. New York: Meridian Books.
KELLY, B. 1951. The Prefabrication of Houses: A Study by the Albert Farwell Bemis
Foundation of the Prefabrication Industry in the United States. Cambridge and New
York: The Technology Press of The Massachusetts Institute of Technology and John
Wiley and Sons.
KREILICK, T. S. 2000. An Investigation of Electrochemical T echniques Designed to
Mitigate the Corrosion of Steel in Historic Structures: Frank Lloyd Wright’s Freeman
House, Hollywood, CA. Thesis (M.H.P). University of Pennsylvania.
LARRARD, F . D. 1999. Concrete Mixture Proportioning: A scientific approach.
London; New York: E & FN Spon; Routledge.
LA Block House Panel: K. Smith, et al; Hollyhock House: J. Steele (Video 2). 2005.
Video. In: 2005 Conference - FLLW and his Los Angeles Progeny 1917 - 1941, 19-23
October 2005 Los Angeles. Los Angeles: Frank Lloyd Wright Building Conservancy.
LA Concrete Block Houses: Issues and Challenges, et al (Video 4). 2005. Video. In: 2005
Conference - FLLW and his Los Angeles Progeny 1917 - 1941, 19-23 October 2005 Los
Angeles. Los Angeles: Frank Lloyd Wright Building Conservancy.
LASEAU, P . AND TICE, J. 1991. Frank Lloyd Wright: Between Principles and Form.
New York: John Wiley and Sons.
205
LENTZ, J. B. 1969. Aerial Isometric From Southeast: Sheet 2 of 7. Historic
American Building Survey No. CA-1989. U.S. Department of the Interior,
National Park Service. Washington, D.C.: Library of Congress. Available
at http://memory.loc.gov/cgi-bin/displayPhoto.pl?path=/pnp/habshaer/ca/
ca0200/ca0228/sheet&topImages=00002a.gif&topLinks=00002r.tif,00002a.
tif&title=&displayProfile=0. [Accessed 6 February 2009].
MARKS, J. R. 2006. The Freeman House : a case for the expansion of significance. Thesis
(M.H.P .). University of Southern California.
McCARTER, R. 1997. Frank Lloyd Wright. London: Phaidon Press Limited.
McCARTER, R. ed. 1991. Frank Lloyd Wright: A Primer on Architectural Principles.
New York: Princeton Architectural Press.
McCARTER, R. ed. 2005. On and By Frank Lloyd Wright: A Primer of Architectural
Principles. New York: Phaidon Press Limited.
MEEHAN, P . J. ed. 1984. The Master Architect: Conversations with Frank Lloyd
Wright. New York: John Wiley and Sons.
MOOR, A. 2002. Californian T extile Block. New York: Sterling.
NABIH YOUSSEF & ASSOCIATES. 1995. Freeman House: Historic Building Seismic
Study. For the Freeman House, USC School of Architecture, Completed August 16,
1995. Los Angeles: unpublished.
ORMSBEE, A. 2006. Water in the Freeman House: a study of the block in the textile
block construction assembly as the primary route for water infiltration. Thesis (M.Bu.S.).
University of Southern California.
RADFORD, W. A. 1909. Cement houses and how to build them ... perspective views
and floor plans of concrete block and cement plaster houses. Chicago: The Radford
Architectural Company.
ROSENBAUM, A. 1993. Usonia: Frank Lloyd Wright’s Design for America.
Washington, D.C.: The Preservation Press.
206
SCHOOL OF ARCHITECTURE. History of the Freeman House. University of
Southern California. Available at http://www.usc.edu/dept/architecture/slide/
Freeman/01.html [Accessed 27 August 2008].
SCHOOL OF ARCHITECTURE. Restoration Efforts. University of Southern
California. Available at http://www.usc.edu/dept/architecture/slide/Freeman/02.html
[Accessed 27 August 2008].
SCHOOL OF ARCHITECTURE. Tours. University of Southern California.
Available at http://www.usc.edu/dept/architecture/slide/Freeman/04.html [Accessed
27 August 2008].
SMITH-EMERY LABORATORIES. 2001. Petrographic Analysis of Submitted CMU
portions. For the Freeman House, USC School of Architecture, Completed November
13, 2001. unpublished.
SMITH, K. 1990. Chicago - Los Angeles: The Concrete Connection. In: 8th Annual
Association of Collegiate Schools of Architecture T echnology Conference, February 1990
Los Angeles. Carpenters/Contractors Cooperation Committee of Southern California
in conjunction with the School of Architecture at the University of Southern
California, 5-11.
SMITH, K. 1992. Frank Lloyd Wright, Hollyhock House and Olive Hill: Buildings and
Projects for Aline Barnsdall. New York: Rizzoli.
SMITH, K. Summer 2005. The L.A. Textile Block Houses. Frank Lloyd Wright
Quarterly. 16 (3).
SWEENEY, R. L. 1994. Wright in Hollywood: Visions of a new architecture.
Cambridge, MA and London: The MIT Press.
TREIBER, D. 1995. Frank Lloyd Wright. London: E & FN Spon.
TREIBER, D. 2008. Frank Lloyd Wright. 2nd ed. London: Springer.
UNESCO WORLD HERITAGE CENTRE. 2008. World Heritage Information Kit.
Paris: UNESCO World Heritage Centre. PDF document. Available at http://whc.
unesco.org. [Accessed 4 February 2009].
207
UNIVERSITY OF SOUTHERN CALIFORNIA. 2002. Interim Report; the Getty
Grant Program Preserve L.A. planning grant for The Harriet and Samuel Freeman House
Frank Lloyd Wright, Architect. Los Angeles: University of Southern California School
of Architecture.
WANK ADAMS SLAVIN ASSOCIATES. 1991. T esting Program Proposal and Initial
Site Visit Report: Samuel Freeman Residence prepared for The University of Southern
California, School of Architecture, Samuel Freeman Residence. Completed 30 October
1991. New York: unpublished.
WEEKS, K. D. AND GRIMMER, A. E. 1995. The Secretary of the Interior’s Standards
for the Treatment of Historic Properties: with Guidelines for Preserving, Rehabilitating,
Restoring & Reconstructing Historic Buildings. Washington, D.C.: U.S. Department of
the Interior, National Park Service.
WEINTRAUB, A. 1998. Lloyd Wright: The Architecture of Frank Lloyd Wright, Jr. New
York: Harry N. Abrams.
WEINTRAUB, A. AND HESS, A. 2005. Frank Lloyd Wright: The Houses. New York:
Rizzoli.
WRIGHT, E. L. 2008. Frank Lloyd Wright and the Freemans. Lecture. University of
Southern California. November 20, 2008.
WRIGHT, F . L. 1914. In the Cause of Architecture: Second Paper. The Architectural
Record. reprint, 1975. New York: Architectural Record Books and McGraw-Hill. 121-
129.
WRIGHT, F . L. 1927. In the Cause of Architecture IV: Fabrication and Imagination.
The Architectural Record. reprint, 1975. New York: Architectural Record Books and
McGraw-Hill. 145-151.
WRIGHT, F . L. 1928. In the Cause of Architecture VII: The Meaning of Materials−
Concrete. The Architectural Record. reprint, 1975. New York: Architectural Record
Books and McGraw-Hill. 205-211.
ZIMMERMAN, S. AND DUNHAM, J. 1994. Details of Frank Lloyd Wright: The
California Work, 1909 – 1974. San Francisco: Chronicle Books.
208
Appendix A: d r Awings of the freem An house
A.1 h istoric American Building s urvey (hABs) drawings
Drawings available from Historic American Building Survey. U.S. Department of the
Interior, National Park Service. Washington, D.C.:Library of Congress. Available at http://
hdl.loc.gov/loc.pnp/hhh.ca0228.
209
210
211
212
213
214
215
A.2 Jeffrey Chusid 1988 CAd translation of fL w original drawings
In 1988, Jeffrey Chusid was assisted by Adam Smith in recreating Frank Lloyd Wright’s
original drawings for the Freeman house in AutoCAD. At the time, Smith was a student
in the USC School of Architecture. The drawings depict a semi-circular patio that was
planned for the south end of the home but was never built. Drawings not to original scale.
216
217
A.3 Jeffrey Chusid 1991-1992 survey of f reeman h ouse exterior
218
219
220
221
222
223
224
225
226
227
228
Appendix b: n ote from H Arriet f reem An
229
Appendix c: TH e ATHen S c HARTe R FOR THe Re STORATiOn OF
HiSTORic MOnUMenTS
Adopted at the First international congress of Architects and Technicians of
Historic Monuments, Athens 1931
________________________________________
At the Congress in Athens the following seven main resolutions were made and called
“Carta del Restauro”:
International organizations for Restoration on operational and advisory levels are to
be established.
Proposed Restoration projects are to be subjected to knowledgeable criticism to 1.
prevent mistakes which will cause loss of character and historical values to the
structures.
Problems of preservation of historic sites are to be solved by legislation at national 2.
level for all countries.
Excavated sites which are not subject to immediate restoration should be reburied 3.
for protection.
Modern techniques and materials may be used in restoration work. 4.
Historical sites are to be given strict custodial protection. 5.
Attention should be given to the protection of areas surrounding historic sites. 6.
________________________________________
General conclusions of the Athens conference
i. -- dOcTRineS. GeneRAL pRincipLeS.
The Conference heard the statement of the general principles and doctrines relating to
the protection of monuments.
Whatever may be the variety of concrete cases, each of which are open to a different
solution, the Conference noted that there predominates in the different countries
represented a general tendency to abandon restorations in toto and to avoid the
attendant dangers by initiating a system of regular and permanent maintenance
calculated to ensure the preservation of the buildings.
230
When, as the result of decay or destruction, restoration appears to be indispensable,
it recommends that the historic and artistic work of the past should be respected,
without excluding the style of any given period.
The Conference recommends that the occupation of buildings, which ensures the
continuity of their life, should be maintained but that they should be used for a
purpose which respects their historic or artistic character.
ii. -- AdMiniSTRATiVe And LeGiSLATiVe MeASUReS ReGARdinG
HiSTORicAL MOnUMenTS
The Conference heard the statement of legislative measures devised to protect
monuments of artistic, historic or scientific interest and belonging to the different
countries.
It unanimously approved the general tendency which, in this connection, recognises a
certain right of the community in regard to private ownership.
It noted that the differences existing between these legislative measures were due to
the difficulty of reconciling public law with the rights of individuals.
Consequently, while approving the general tendency of these measures, the
Conference is of opinion that they should be in keeping with local circumstances
and with the trend of public opinion, so that the least possible opposition may
be encountered, due allowance being made for the sacrifices which the owners of
property may be called upon to make in the general interest.
It recommends that the public authorities in each country be empowered to take
conservatory measures in cases of emergency.
It earnestly hopes that the International Museums Office will publish a repertory and
a comparative table of the legislative measures in force in the different countries and
that this information will be kept up to date.
iii. -- AeSTHeTic enHAnceMenT OF AncienT MOnUMenTS.
The Conference recommends that, in the construction of buildings, the character
and external aspect of the cities in which they are to be erected should be respected,
especially in the neighbourhood of ancient monuments, where the surroundings
should be given special consideration. Even certain groupings and certain particularly
picturesque perspective treatment should be preserved.
A study should also be made of the ornamental vegetation most suited to certain
231
monuments or groups of monuments from the point of view of preserving their
ancient character. It specially recommends the suppression of all forms of publicity,
of the erection of unsightly telegraph poles and the exclusion of all noisy factories and
even of tall shafts in the neighbourhood of artistic and historic monuments.
iV. -- ReSTORATiOn OF MOnUMenTS.
The experts heard various communications concerning the use of modern materials
for the consolidation of ancient monuments. They approved the judicious use of all
the resources at the disposal of modern technique and more especially of reinforced
concrete.
They specified that this work of consolidation should whenever possible be concealed
in order that the aspect and character of the restored monument may be preserved.
They recommended their adoption more particularly in cases where their use makes
it possible to avoid the dangers of dismantling and reinstating the portions to be
preserved.
V. -- THe deTeRiORATiOn OF AncienT MOnUMenTS.
The Conference noted that, in the conditions of present day life, monuments
throughout the world were being threatened to an ever-increasing degree by
atmospheric agents.
Apart from the customary precautions and the methods successfully applied in the
preservation of monumental statuary in current practice, it was impossible, in view of
the complexity of cases and with the knowledge at present available, to formulate any
general rules.
The Conference recommends:
That, in each country, the architects and curators of monuments should 1.
collaborate with specialists in the physical, chemical, and natural sciences with a
view to determining the methods to be adopted in specific cases;
That the International Museums Office should keep itself informed of the work 2.
being done in each country in this field and that mention should be made thereof
in the publications of the Office.
With regard to the preservation of monumental sculpture, the Conference is of
opinion that the removal of works of art from the surroundings for which they were
designed is, in principle, to be discouraged. It recommends, by way of precaution, the
232
preservation of original models whenever these still exist or if this proves impossible,
the taking of casts.
Vi. -- THe TecHniQUe OF c OnSeRVATiOn.
The Conference is gratified to note that the principles and technical considerations set
forth in the different detailed communications are inspired by the same idea, namely:
In the case of ruins, scrupulous conservation is necessary, and steps should be taken
to reinstate any original fragments that may be recovered (anastylosis), whenever this
is possible; the new materials used for this purpose should in all cases be recognisable.
When the preservation of ruins brought to light in the course of excavations is found
to be impossible, the Conference recommends that they be buried, accurate records
being of course taken before filling-in operations are undertaken.
It should be unnecessary to mention that the technical work undertaken in
connection with the excavation and preservation of ancient monuments calls for close
collaboration between the archaeologist and the architect.
With regard to other monuments, the experts unanimously agreed that, before any
consolidation or partial restoration is undertaken, a thorough analysis should be made
of the defects and the nature of the decay of these monuments. They recognised that
each case needed to be treated individually.
Vii. -- THe c OnSeRVATiOn OF MOnUMenTS And inTeRnATiOnAL
c OLLABORATiOn.
a) T echnical and moral co-operation.
The Conference, convinced that the question of the conservation of the artistic and
archaeological property of mankind is one that interests the community of the States,
which are wardens of civilisation,
Hopes that the States, acting in the spirit of the Covenant of the League of Nations,
will collaborate with each other on an ever-increasing scale and in a more concrete
manner with a view to furthering the preservation of artistic and historic monuments;
Considers it highly desirable that qualified institutions and associations should,
without in any manner whatsoever prejudicing international public law, be given an
opportunity of manifesting their interest in the protection of works of art in which
civilisation has been expressed to the highest degree and which would seem to be
threatened with destruction;
233
Expresses the wish that requests to attain this end, submitted to the Intellectual Co-
operation Organisation of the League of Nations, be recommended to the earnest
attention of the States.
It will be for the International Committee on Intellectual Co-operation, after an
enquiry conducted by the International Museums Office and after having collected all
relevant information, more particularly from the National Committee on Intellectual
Co-operation concerned, to express an opinion on the expediency of the steps to be
taken and on the procedure to be followed in each individual case.
The members of the Conference, after having visited in the course of their
deliberations and during the study cruise which they were able to make on this
occasion, a number of excavation sites and ancient Greek monuments, unanimously
paid a tribute to the Greek Government, which, for many years past, has been itself
responsible for extensive works and, at the same time, has accepted the collaboration
of archaeologists and experts from every country.
The members of the Conference there saw an example of activity which can but
contribute to the realisation of the aims of intellectual co-operation, the need for
which manifested itself during their work.
b) The role of education in the respect of monuments.
The Conference, firmly convinced that the best guarantee in the matter of the
preservation of monuments and works of art derives from the respect and attachment
of the peoples themselves;
Considering that these feelings can very largely be promoted by appropriate action on
the part of public authorities;
Recommends that educators should urge children and young people to abstain from
disfiguring monuments of every description and that they should teach them to take a
greater and more general interest in the protection of these concrete testimonies of all
ages of civilisation.
c) Value of international documentation.
The Conference expresses the wish that:
Each country, or the institutions created or recognised competent for this purpose, 1.
publish an inventory of ancient monuments, with photographs and explanatory
notes;
234
Each country constitute official records which shall contain all documents relating 2.
to its historic monuments;
Each country deposit copies of its publications on artistic and historic monuments 3.
with the International Museums Office;
The Office devote a portion of its publications to articles on the general processes 4.
and methods employed in the preservation of historic monuments;
The Office study the best means of utilising the information so centralised. 5.
________________________________________
2 August 1994; modified 12 January 1996
Accessed 4 February 2009 at http://www.icomos.org/athens_charter.html
235
Appendix d: Block m Anuf Acturing photo Journ Al And
process video
d.1 o ctober 8, 2008 block manufacturing
236
237
d.2 n ovember 2, 2008 photos of cured blocks
238
239
240
d.3 d ecember 3, 2008 block manufacturing
241
242
243
244
d.4 c atalog of blocks manufactured
245
246
247
248
249
250
251
d.5 video of block manufacturing process
Click on image to start video.
Background song, Jerry’s Breakdown, used with permission from the artist, Kenneth Belcher.
252
Appendix e: K Arsten p ipe t esting p hotos And Videos
e.1 p hotos
253
254
255
256
257
258
259
e.2 Videos
Click on image to start video.
Sample Dry-pack Block Test
260
Sample Wet-mix Block Test
261
Appendix F: Compression Tes Ting Videos Click on image to start video.
Cylinder capping process
Testing of cylinder S3
262
Testing of cylinder S6
Testing of cylinder 24
263
Testing of cylinder 25

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Frank Lloyd Wright's textile block: the essential qualities, challenges and alternative methods

PDF

The textile block system: structural analysis and alternative seismic upgrading to IBC 2003

PDF

Water in the Freeman House: a study of the block in the textile block construction assembly as the primary route for water infiltration

Asset Metadata

Creator McAlister, Benjamin P. (author)

Core Title Conservation and reconstruction of textile blocks: an investigation of treatment and replacement options at the Frank Lloyd Wright Freeman House

School School of Architecture

Degree Master of Building Science

Degree Program Building Science

Publication Date 06/04/2009

Defense Date 04/27/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Building Science,Concrete construction,Frank Lloyd Wright,Freeman House,Historic Preservation,material testing,OAI-PMH Harvest,textile block

Place Name buildings: Freeman House (geographic subject)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Noble, Douglas (committee chair), Borden, Gail P. (committee member), Breisch, Kenneth (committee member)

Creator Email benjamin.mcalister@gmail.com,mcaliste@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2286

Unique identifier UC1422959

Identifier etd-McAlister-2913 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-250704 (legacy record id),usctheses-m2286 (legacy record id)

Legacy Identifier etd-McAlister-2913.pdf

Dmrecord 250704

Document Type Thesis

Rights McAlister, Benjamin P.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu