University of Southern California Dissertations and Theses

Vibration reduction using prestress in wood floor framing

(USC Thesis Other)

Vibration reduction using prestress in wood floor framing

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content VIBRATION REDUCTION USING PRESTRESS IN WOOD FLOOR FRAMING by Ruben Ayrapetyan 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 May 1992 Copyright 1992 Ruben Ayrapetyan UMI Number: EP41427 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. UMI' Dissertation Publishing UMI EP41427 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 U N IV E R S IT Y O F S O U T H E R N C A L IF O R N IA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 9 0 0 0 7 T h is thesis, w ritte n by R ube/ ;/ u n d e r the d ire ctio n o f h Thesis C om m ittee, and a pp ro ve d by a ll its mem bers, has been p re sented to and accepted by the D e an o f T he G ra dua te S chool, in p a r tia l fu lfillm e n t o f the requirem ents f o r the degree o f _ S c /(EAJCc Dean T)at, May 18. 1992 THEStS COMMITTEE Chairman ii ACKNOWLEDGEMENTS I would like to express my deepest gratitude and appreciation to my academic advisors Prof. Goetz G. Schierle, Prof. Dimitry Vergun, and Prof. James Ambrose. My special thanks to Prof. Goetz G. Schierle for his constant and irreplaceable support, and guidance throughout the whole research, and to Prof. Dimitry Vergun for the knowledge and the experience that he had contributed in shaping my future as an engineer. My sincere thanks to the USC School of Architecture, and to Lee Bolin and Associates Builders Inc., for their financial and physical support of the project. Most of all, I’d like to express my love and gratefulness to my family who had supported me every step of the way, making it possible for me to reach long desired goals. iii TABLE OF CONTENTS Introduction 1 Chapter 1 - Structural Floor Systems 3 Chapter 2 - Wood Construction 8 Chapter 3 - Competitive Systems IV Chapter 4 - Vibration Reduction using 21 Prestress in Wood Floor Framing Chapter 5 - Testing of the Static Simulation 26 Model Chapter 6 - Testing of Full Size Mock-up without Plywood 41 Chapter 7 - Testing of Full Size Mock-up 65 with Plywood Applied Chapter 8 - Perception Test on Human Response 73 to Vibration and Noise of the Floor System Conclusions 85 iv ABSTRACT The intention of this research is to demonstrate that introduction of simple, prestress mechanism into wood floor framing may reduce the vibration of the floor system. The problem of vibrating floors continues to be a great concern for the Architects and Civil Engineers of the United States and some other countries, where wood structures are still prevailing in residential and light commercial construction. Although structurally safe, vibrating floors create a strong psychological discomfort for the occupants. Technology for prestressed wood floor systems, developed at the University of Southern California, proved to be very effective to reduce vibration by up to 50%. It is an inexpensive way to reduce vibration, noise and deflection of wood framed floors. Key words: Prestress; Vibration Reduction; Deflection Reduction. INTRODUCTION 1 The construction industry has become the largest single production business of the American economy for the past several years (Clough, R.H. 1986). The magnitude and the importance of construction business is well illustrated by some figures recently reported by Associated General Contractors of America. The total annual volume of construction in this country at the present time is approximately $300 billion dollars. The annual expenditure for construction normally accounts for between 7 and 12 percent of the dollar volume of our gross national product, off of which 70 percent is privately financed, and 30 percent is paid by various public agencies. Considering the magnitude and the importance of the construction industry, the need for continuous research and technology development becomes immediately apparent and inevitable. Some of the findings and discoveries, resulting out of research, cause significant impact on the technical and financial aspects of construction business. The field of construction is widely diversified, and, although portions of it are very intermixed, it can be commonly divided into four main categories such as industrial, engineering, building and residential construction. All four categories are briefly discussed below. 1. Industrial construction includes the projects that are associated with the manufacture or production of a commercial product or service (Clough, R.H. 1986). This category includes some of the largest projects built and it accounts for 5 to 10 percent of the annual volume of new construction. 2. Engineering construction covers structures whose design is concerned more with functional considerations than aesthetics, and it accounts for approximately 20-25 percent of the new construction market. 3. Building construction includes buildings in the commonly understood sense, other than housing, that are erected, for institutional, educational, light industrial, commercial, social, religious, governmental and recreational purposes. This category accounts for 35-40 percent of the annual total of new construction. 4. Residential, or housing construction includes the building of single-family homes; condominiums; multiunit townhouses; low-rise, garden-type apartments, etc. This category of construction is dominated by small firms and normally accounts for about 30-35 percent of new construction during a typical year. 3 The proposed, prestressed floor system is targeted for the third and fourth construction categories, which combined account for over a 70 percent of the total new construction. Considering the size of the market for which this technology was developed, the importance and usefulness of the product becomes quite obvious. This product is designed primarily for wood construction or types of constructions with intermixed wood and other structural material systems. In the United States and primarily in the Western U.S. there are more buildings constructed of wood than any other structural material (Breyer, D.E. 1980). The widespread use of wood in the construction of buildings has both an economic and an aesthetic basis. The ability to construct wood buildings with a minimal amount of equipment has kept the cost of wood-frame buildings competitive with other types of construction. Besides that, in many instances exposed look of wood has an important architectural consideration, and it is the only renewable construction material. Chapter 1 - Structural Floor Systems 1. Types of Floor Systems. Selection of appropriate structure floor system for a given project is an important and difficult task that deals with many different structural and economic aspects. (Reid, E. 1984) Those aspects include : 1) General type of construction used in building. 2) Plan of the building. 3) Floor loads (static and dynamic) 4) Lateral resistance 5) Floor thickness 6) Fire protection 7) Heat and sound transmission 8) Weight 9) Mechanical equipment installation 10) Compliance with local code and registration 11 ) Cost. Besides the items mentioned above local conditions such as, availability of labor and materials, weather conditions, time constraints, etc., also make significant impact on selection of structural floor system. Structural framing of the building and floor system are closely interrelated and in most cases choice of the floor system is dictated by the choice of structural frame. Wood stud walls and partitions would naturally be associated with light wood joists and subfloors. Such floor systems might also be used with wood columns, beams, and girders; or a heavy timber floor system might be used. If a steel frame is used, the floor system might be one of several types of reinforced concrete plain or ribbed floor systems, except the flat-slab, flat-plate, or slab-band (Benjamin, B.S. 1982). This type of frame is also suitable for use with thick concrete slabs on light steel or open-web joists or on light-gage steel cellular, ribbed, or corrugated panels. If the framing can be arranged so that there are many identical panels, some form of precast reinforced concrete or gypsum panel may prove advantageous. If the beams and girders are to be of reinforced concrete, the floor system will be plain, ribbed or waffled reinforced concrete slabs and the columns usually will be reinforced concrete. Such columns will also be used with flat slab, flat-plate, and slab-band floor systems. Precast concrete or gypsum panels may be advantageous for the same conditions given in the preceding paragraph. Various types of floor systems can be used with masonry bearing walls, including light wood joist and subfloors, heavy timber floor systems, wood or concrete floors supported on steel I-beam or light steel or open web joists, reinforced concrete beams and slabs, reinforced concrete ribbed slabs, and precast concrete or gypsum slabs supported on steel joists (Cowan, H.J. 1991). There are three major ways that lateral loads can be resisted by the buildings: 1) Shear walls 2) Moment resistive frames 3) Braced Frames Some buildings, to resist lateral loads, utilize combination or even all three systems at once, however moment frames and braced frames are more common for high- rise buildings to resist wind and earthquake loads, while shear wall system more widely used for low-rise, heavy buildings. When shear walls are used to resist lateral loads, the function of floor system becomes very vital, because it acts as a diaphragm to transfer lateral loads of a particular story into shear walls bellow. For example: In type V wood construction, plywood floor sheathing blocked at the edges acts as a continuous horizontal diaphragm for lateral load transfer. Often shear walls are used in multistory buildings that lack rigid connection between columns and floor members. Buildings with flat-plate or slab-band floor systems qualify for such construction (Cowan, H.J. 1991). Since the seismic loads directly depend on the dead weight of a building, the advantage of wood (wherever permissible) in that respect becomes quite obvious due to its light weight. Whenever allowed by the design, using lightweight concrete instead of regular reinforced concrete might also help to reduce the weight of a structure. 2. Building Design The plan of a building is another major factor that influences the choice of floor construction system. If the spans are relatively short then wood joists and light timber construction are more economical. This is the case with most residential and light commercial construction. Structural concrete slabs and composite, corrugated steel deck floor systems are also characteristic of shorter spans. On the other hand if spans are relatively long then steel truss joists, concrete ribbed slabs and composite concrete-steel beam systems are more suitable (Hornbostel, C. 1978). 3. Resistance of Floors to Vertical and Lateral Loads Wood joist floors with plywood subfloors and some thick concrete slab floors supported by members such as open web steel joists, corrugated light-gage steel decks, etc., are appropriate only for light and medium concentrated loads which are typical for residential and light commercial construction. For buildings with heavier occupancy loads, reinforced concrete slabs over heavier supports such as steel frames and girders, reinforced concrete columns and beams, heavy timber beams, etc., are more appropriate (Benjamin, B.S. 1982). 4. Fire Resistance Considerations for Buildings. Fire resistance is another major factor contributing to selection of structural floor systems. In a building where the occupancy load is not too high, resistance to fire is not a decisive factor and such a building could be kept at a low budget. Wood frame construction, with wood joist floor systems, such as for residential construction, can be used. Contrary to that, in highly populated, downtown districts where the occupancy load is very high and with large floor areas, fire resistant construction methods and materials are required, such as sprayed fire proofing over structural steel, fire stops, fire proof, self closing exits and shafts, automatic sprinkler system, etc. (Cowan, H.J. 1991). Chapter 2 - Wood Construction 1• Framing and Framing Members. Although not permitted within certain fire zones, wood frame construction (often called also frame construction) has been, and remains, a leading technology in residential and light commercial construction. Wood construction, in general, is the type of construction in which the floors, walls, roofs and partitions are entirely or partially of wood or other combustible material. This type of construction is limited to three stories above ground. There are some wood buildings that are built higher than three stories, however special structural and fire provisions are considered in these cases (UBC, 1991), Except for post and beam construction most of the timber that comprises the structural frame of wood building, has a nominal thickness of 2in. Wall, floor and roof members most commonly are spaced at 16in o.c.. However spacing of 12in. and 24in. are often used too. Sizes and depths of main structural members such as beams, rafters and joists are determined by allowable stresses or allowable deflection, loading conditions and desired rigidity. Framing of exterior walls is usually covered with finished surfaces such as wood siding or concrete plaster which are applied over water proofing and (in case of concrete plaster) wire mesh. In some cases, different types of veneer materials such as brick veneer, stone veneer, tile etc., also are applied as finished surface covering. Roof construction consists of wood rafters, some sloped, set at a spacing of 16in, 24in., and sometimes even 48in. apart depending on load conditions and sizes of the joists used. Plywood sheathing (usually l/2in) is applied directly over joists with 8d nails and in most of cases edges of plywood between the joists are supported by horizontal wood blocking. Blocking are used to create a continuous, horizontal, plywood diaphragm for transfer of lateral loads (Andersen, L.O. 1987). There are two major types of ceilings used for construction of 10 dwellings: cathedral ceilings, and flat ceilings. In case of cathedral ceilings interior gyp. board or sheetrock is nailed to the underside of slopping rafters with roof insulation sandwiched in the depth of roof rafters. In this type of construction rooms on the second floor have higher ceilings, however the building lacks an attic space. From the structural standpoint, in cathedral ceiling arrangement, rafters act as a simply supported members and the necessity of ridge beam becomes apparent. In case of flat ceilings sheetrock is nailed to the underside of ceiling joists thus creating an attic space between flat ceiling joists and slopping roof rafters. If properly designed ceiling joists and rafters combined act as roof trusses with ceiling joists acting as tension members. In this arrangement roof insulation is laid directly over or between ceiling joist and only a ridge board instead of ridge beam is required. Wood floor construction consists of wood joists, spaced at 12in, 16in or 24in apart and covered with subfloor material which is nailed directly to joists; normally with lOd nails. Wood joists usually have nominal thickness of 2in, while the depth varies between 6 to 14in. In cases of heavier loading conditions doubling of joists or wider members are used. Originally, 1x6 wood strip boards were used as subfloor material diagonally nailed to TDP joists. With 11 the present technology and current code regulations, strip board subfloors are not considered an adequate method to create a continuous floor diaphragm for lateral load transfer. Because of that plywood subfloors are used instead (Cowan, H.J. 1991). The thickness of plywood subfloors varies, depending on joist spacing and loading. Generally it is between 5/8” to 1-1/8” thick. Plywood panels come in different grades, and it is up to the designer to specify the type required to accommodate the design. Most commonly, plywood is laid with face grain perpendicular to floor joists and individual panels are staggered to create a more continuous diaphragm surface. To secure plywood, typically 8d nails are used for roof and lOd nails for floors. Minimum required nailing for plywood is as follows: 6in of spacing around edges and 10-12in in field. If the shear force in a plywood diaphragm is higher than the minimum allowable value then closer nail spacing will be required. Row of solid wood blocking should be placed under the unsupported edges of plywood to prevent differential deflection of adjoining plywood sheets and to provide continuity in the diaphragm. Shrinkage and swelling are important factors that can effect the performance of wood floors. During 12 construction, when the structure is still not enclosed, rains and changes in humidity of climate can initiate swelling of the entire floor surface, especially for strip board subfloors, which could seriously damage the building. To avoid these kind of problems spaces of about 1/8” to 1/4" should be left in between the adjoining pieces of subfloor sheathing (Cowan, H.T. 1991). Due to its nature, wood is an easy material to work with. It is light, it is relatively strong and, in most cases, does not require any fasteners other than nails. All these characteristics, combined with the relatively low cost of the product, make wood an irreplaceable asset to the construction industry and in particular to residential and light commercial constructions. Although very diverse at a glance, wood construction follows very strict rules and regulations. There are two types of framing systems that are commonly used for dwellings (Anderson, L.O. 1987). 1) Platform frame 2) Balloon Frame 2. Platform Frame is a framing type that is mostly popular in the Western U.S. and, due to its regional popularity, it is often also called Western Framing. In this type of framing system each floor of the building represents a platform, and, until the framing and 13 FIG-1 Platform Frame (National Lumber Manufacturers Assoc-) 14 subfloor of that particular floor are completed, walls above that floor cannot be started (Fig. 1). The existence of a first floor platform depends whether the dwelling has raised the slab foundation system. If a dwelling has raised floor foundation, then first floor platform is needed. Framing starts by placing sill plates over concrete foundation stem walls. First floor joists, that comprise the first floor platform, and are then being framed-in. After the joists are framed-in, plywood subfloor is nailed on top of joists and then the platform is ready to support the sole plates of the first floor walls above. The second floor platform is framed on top of double plates of first floor walls. After the second floor plywood subfloor is completed the same process is repeated for the second floor walls and the roof. On the other hand, if the dwelling has a slab on grade then the sill plates also become sole plates of the first floor walls and the first floor platform is eliminated. To provide temporary lateral rigidity to the framing of building, before the exterior sheathing are applied, 1x6 let-in or metal strap diagonal braces are usually provided. There are two types of bridging that are commonly used in platform frame floor construction. These are solid blocking and cross-bridging systems. Solid blocking usually are of same size lumber as the joists, and they are placed perpendicular to the joists, usually 8 feet on center, or sometimes even closer. Solid blocking is a better option, since it can be used to support the edges of adjacent plywood sheets and can also serve as a support to a partition wall above. In addition, it creates better distribution of vertical and lateral forces on the diaphragm, and provides resistance to torsional buckling or floor joists. 3. Balloon Frame. While platform frame is predominantly used in Western U.S., Balloon frame construction is the more popular method of construction in the Eastern part of the country. In difference to platform frame systems, in which exterior walls are interrupted by each platform, exterior wall studs in balloon framing extend in one piece from foundation to the roof. This is true for one as well as for two story dwellings. Joists of the second floor are nailed straight to the studs. At the same time, they are supported by 1x4 ledgers let-in which is also nailed to the side of the studs and is called ribbon (Fig. 2). Interior first story bearing walls are supported by girders which rest on concrete footing pads. These walls usually have single top plates which carry ends or second floor joists and support studs above. 16 R a fte r Joist Plate J o /s f __ F ire S to p p in g Stud- '.Single Sub Floor S topping ^ Jo/sf Ledgers' Girder -Bridging Sub Floor — Plate Si// ■ F o u n d a tio n FIG.2 Balloon Frame (National Lumber Manufacturers Assoc.) Roof joists and rafters are supported by double plates that sit at the top of wall studs. In other aspects such as blocking, wall bracing, double joists, plywood sheathing, etc. the technology for balloon framing is the same as for platform framing. There are advantages and disadvantages to both framing systems. The primary advantages of the Balloon frame over Platform frame is its lower vertical shrinkage. However, usually the Platform frame is chosen over the Balloon frame system because it is easier to built. The reason why Platform frame is easier to build is that new framing walls are built on the platform then they are being tilt up to their upright position. This method simplifies construction considerably because the framing is more accessible to contractors at each platform level. It is the only practical system for three story buildings. On the other hand, lower vertical shrinkage of Balloon framed dwellings offer a better quality of construction, especially when outside surfaces vulnerable to cracking are concerned, such as concrete plaster, stone or brick veneer etc. (Anderson, L.O. 1987). Chapter 3 - Competitive Systems Balloon and Platform framing construction systems use predominantly conventional lumber as framing material 18 for dwelling construction. However there are many other systems and materials that can be used for this type of construction. The selection process for proper system and materials is a very difficult one, because it depends on a number of considerations, such as availability of materials, labor, quality of construction desired, architectural considerations, and, very importantly, the cost of the project. Besides the joists, made of conventional lumber, there are some other types of joists available, that can also be used in the balloon or platform framing systems. However most of them require different methods and means of application. The materials mentioned below are used as an alternative to joists made of regular lumber, for dwelling construction (Hornbostel, C. 1978). 1. Structural, fabricated "I" joists with laminated wood flanges and plywood webs. 2. Structural, fabricated "I" joists with laminated wood flanges and tubular steel open webs ♦ 3. Light-gauge steel joists. Punched or solid, double nailable. 4. Open web steel joists. H or K series, with horizontal bridging, spans up to 30 ft. 1 • Structural fabricated MI, f joists, with laminated 19 wood flanges and plywood webs, are often used for residential,multi-family, institutional and light commercial applications. These types of joists usually come in depths of 9-1/2” to 30” in 2” increments. However for dwelling construction depths of 9-1/2” to 14” are most commonly used (Truss Joist Corp. Catalog, 1984). Top and bottom cords act as compression and tension members respectively, thus providing bending capacity to the joist. Plywood webs of these joists provide the shear capacity required and usually for webs structural I grade of plywood is used. Under the high shear and bearing loads local web buckling presents a problem for the structural fabricated ”1” joists. For this reason web stiff eners are used to support plywood webs at critical shear locations and at places of web connections. These joists can be used both in balloon and platform frame systems. 2. Structural Fabricated ”1” joists with laminated wood flanges and tubular steel open webs, commonly called open web series, are less used in dwelling construction. Because of their higher strength and ability to span up to 60 ft they become cost ineffective. These open web series are analyzed as pin connected trusses and, as in the previous case, laminated wood flanges serve as tension and compression members to resist binding stresses. Pin bearing, buckling, and shear forces are considered and resisted by slopping tubular web members. If properly designed this joist can be used in balloon and platform framing systems, however usually other support systems, such as steel beams with wood top nailers, are being used. 3. Light-gauge steel joists range from gauge No. 12 to No. 20 inclusive and are either single bent shapes or different bent shapes welded together. These shapes have a minimum yield point of 40,000 psi and are formed from flat rolled carbon steel. These joists come in depths of 4" to 12". If the heaviest gauge of steel (12 gauge) is used with 12" deep joists @ 12" o.c. then spans of up to 20 ft. can be achieved. For this reason metal stud and joist systems are being used for residential, motel, multiple housing, and light commercial types of structures, Light-gauge steel joists are not commonly used in platform and balloon wood framing systems. Light steel structures are more appropriate for these types of joists (Hornbostel, C. 1978). 4• Open web steel joists (also called bar joists) have minimum yield strengths between 36,000 to 50,000 PSI, and present a prefabricated lightweight trusses which have top and bottom cord made of T-angles and webs of tubular sections. There are many different types of open web steel joists. However the kinds that are commonly used for dwelling construction are H or K series with spans of up to 30 feet. Open web steel joists are available prefabricated or custom made for the lengths required. These types of joists generally are not used with conventional wood framing systems and require other types of supporting systems. All the joist types discussed above, with the exception of structurally fabricated ”1” joist with wood flanges and webs, are cost ineffective when they are used for projects with medium floor spans, such as dwelling construction. In comparison to these products, the use of conventional lumber, as joists, is more economical. Based on 1988 building construction cost data (Means, 1987) MIM joists with wood flanges and webs are the closest in price to the prices of conventional lumber, however they’re still between 25%-30% more expensive than wood joists. This leaves conventional lumber construction most economical for dwelling construction. Chapter 4 - Vibration Reduction Using Prestress In Wood Floor Framing In the following Chapters the technical aspects of wood floor prestressing are going to be discussed and, by using the data and results of extensive experimentation and testing, it will be shown that introduction of a simple prestress mechanism in wood floor framing reduces 22 vibration of the floor system by up to 50%. Conducted tests and experimentations have consisted of four major parts, and each part is organized into separate chapter. Our discussion will follow each of these chapters in the order they have been conducted, during the whole testing process. The four major parts of testing consisted of: 1. Building and testing of a static simulation model. 2. Building and testing of a full size (20,x20,) mock-up, with plywood still not applied at this time, 3. Retesting of the full size mock-up after shrinkage and plywood application. 4. Conducting a perception test on human response to vibration and noise of floor system. General Introduction With modern technology available, standard 2x12 (nominal) joist, which are the biggest standard sections available, span little over 21 feet (American Institute of Timber Construction, 1974). However, once clear span exceeds 16 feet, vibration and noise of the floor becomes quite noticeable. Although structurally safe, vibrating floors create a strong psychological discomfort for the 23 occupants. On the other hand, tests demonstrated that using prestressing devices in wood floors is very effective in reducing deflection and vibration. It is also inexpensive. In addition, it is easy and fast to install and does not require any special skills or instruments. The physics behind the idea of floor prestressing is very simple. When point load from a moving object, such as a person, is applied to the floor, the joists immediately below tend to act individually to resist that load. Although the blocking between the joists, as well as plywood, are useful in transferring lateral loads, for distribution of vertical point loads to the entire floor area, their help is insignificant. The reason why blocking do not provide adequate distribution of vertical point loads is because the nails between blocking and joists tend to come loose soon after construction. This is due to the cyclical nature of applied live loads and shrinkage of lumber. These conditions are also responsible for squeaky noise caused by the withdrawal of nails, due to differential deflection of the floor joists. The primary function of the floor prestressing is to press together adjacent floor joists, to make them act together as a group in resisting vertical loads. This is achieved by using a prestressing steel cable with end bearing steel plates which is placed parallel and next to a continuous row of blocking at the midspan and middepth of floor joists. Since the shear stress is 0 at midspan of the joist under the proposed loading conditions, and moment is max at midspan, but it is 0 at the neutral axis throughout the member, then one should realize that the point at the midspan and middepth of the joists is the point of 0 stress. Based on these facts, it is fair to conclude that a relatively small (1/2" diam.) hole at the above described location, in the joist, would not reduce the strength of the member.(Fig. 4) When prestress is applied by means of a turnbuckle or a similar prestressing device, the row of blocking starts to act as a compression member. The applied prestress force acting horizontally creates a friction between joists and blocking. If the prestress force is high enough it creates enough friction to prevent the slippage between joists and blocking in vertical direction without help of the nails. As expressed by the formula: F = M x F friction prestress, where Ffricti0n = friction force between joists and blocking M = friction coefficient of wood Fprestress = f°rce normal to the friction surface Slippaqe of blockings 1 Prestressed floor FIG.3 Behavior of Blockings under Dynamic,Impact Forces. FIG.4 Moment and Shear Diagrams. Floor prestressing produces two important results. First, the prevention of slippage between blocking and joists forces the joists to act in a group to distribute and to resist vertical point loads applied to any location of the floor. This group action of joists significantly reduces the local deflection of the floor. (Fig. 3) Secondly, the absence of slippage between joists and blocking creates much quieter floor systems with very little or no noise at all. The noise due to friction between members and withdrawing of nails are virtually eliminated when prestressing is applied to the floor. Chapter 5 - Testing of the Static Simulation Model 1. Materials and Scales Used for Modeling. As mentioned earlier, the first portion of testing was performed on a static simulation model. The model was made of Douglas Fir No. 2 lumber, which is commonly used for floor construction (Fig.5). The model was built to be six times smaller than the actual floor under consideration. The relationships between the technical parameters of the model and the prototype, such as size, loading 3 and physical characteristics of the materials used, are not linear. For example, if the model is six times smaller than the prototype, that does not mean that the 27 FIG.6 Strain Gage loadings used on the model should be also reduced six times from the loadings used on the actual floor considered. To accommodate for these uni inear scale reductions and to create a model that accurately simulates the real structures, the following three specially formulated scales were used: s* ~ Geometric scale = L /L = 0.167 - 1:6 a o s s = Strain scale = S model/S original =1= 1:1 s f - Force scale = P„/P„ a o s f = E l /E l x 1/(SJ2 x (S0 ) = 0.0278 = 1:36 aa oo t s E = modulus of elasticity I = moment of inertia m = subscript for model o = subscript for original structure By utilizing these three scales, the accurate representation of the problem, with different sizes and scales, was achieved. Another major consideration for testing the model, was the type and the magnitude of loading to be used. Since the testing was being done for static loading conditions, the dynamic impact loads were transformed into equivalent static loads (Popov, 1978) by means of the formula: % = + d + 2h/Dst)1/2] where Pdyn = Dynamic impact load = Gradually applied, known static force h = Distance that object falls before impact Dgt = Deflection caused by static force Pgt To see how the equivalent load transformation formula works we will consider an actual loading caused by people on the floor system. For example, a 200 pound (890 N) person would create an equivalent impact load of 600 pounds (2670 N) while walking on the floor. This is considering that the person puts 80 pounds of his weight on the front leg at the time of impact and the vertical distance that the foot travels before striking the floor is 2.2in. 2. Static Simulation Model The static simulation model (Fig. 7) consisted of 16 joists, 40 inches long (clear span), which were placed on wood frame acting as a continuous bearing support at the perimeter of the modeled floor. Steel cable .433 millimeters in diameter was ran perpendicularly through the midspan and middepth of all the joists and secured at one edge joist by steel washers against pull-out. The other end of the cable, after passing through all joists, was bent over a pulley and suspended from the edge of the floor model. By hanging weights from this second end of the cable, a desired prestress could be reached. A line 30 Plywood Joists Blocking Prestress cable FIG.7 Static Simulation Model. ‘ -Floor Plywood — Joists — Prestress Cable — Turnbuckle FIG.8 Model Section. of blocking was placed at midspan of the joists in such a manner that they would alternate on each side of the steel cable (Fig. 8). As a result, there was an equal number of blocking placed at each side of the steel cable. The blocking placed in a single alternating row have performed well during the model testing, however, for the full scale mock-up their arrangement was changed. This is due to the fact that the alternating blocking under high prestress loads were causing local bending of joists in the weak axis. New arrangement of blocking will be discussed in later paragraphs when the mock-up is presented. 3. Model Testing Testing of the static simulation model was performed as follows: First, static weight equivalent of an impact load was hung at the midspan of a single joist in the center of the modeled floor. Without applying any prestress to the floor, the initial deflection, due to the point load, was recorded, using a dial strain gage. (Fig. 6) Next, for each increasing prestress in the cable, the deflection readings were recorded. The vertical, static point load (weight) was kept constant throughout the whole test and was reapplied at each new test. For better reliability and certainty in the experiment findings, three separate sets of tests were performed. Test #1 consisted of 16 readings with constant, scaled point load of 5.6 pounds (200 pounds Actual), with an increasing prestress force from 0 to 56 pounds (2000 pounds Actual) in increments of 2.8 pounds (100 pounds Actual) first ten readings, and increments of 5.6 pounds (200 pounds Actual) for the rest of the readings. (Fig. 9) Test #2 consisted of 16 readings with constant, scaled, point load of 11.2 pounds (400 pounds actual), with an increasing prestress forces from 0 to 56 pounds (2000 pounds actual) in the increments of scaled 2.8 pounds (100 pounds actual) first ten readings, and increments of 5.6 pounds (200 pounds actual) for the rest of the readings. (Fig. 10) Test #3 consisted of 16 readings with constant, scaled, point load of 16.8 pounds (600 pounds actual), with an increasing prestress force from 0 to 56 pounds (2000 pounds actual) in increments of 2.8 pounds (100 pounds actual) first ten readings, and increments of 5.6 pounds (200 pounds actual) for the rest of the readings. (Fig. 11) For each of the three sets of data recorded graphs 33 POINT LOAD PRESTRESSING FORCE DEFLECTION STATIC (lb) flb) (in) SCALED REAL SCALED REAL SCALED REAL 5.6 200 0 0 0.0152 0.0912 5.6 200 2.8 100 0.014 0.084 5.6 200 5.6 200 0.012 0.072 5.6 200 8.4 300 0.012 0.072 5.6 200 11.2 400 0.0115 0.069 5.6 200 14 500 0.0113 0.0678 5.6 200 16.8 600 0.0109 0.0654 5.6 200 19.6 700 0.0103 0.0618 5.6 200 22.4 800 0.0104 0.0624 5.6 200 25.2 900 0.0103 0.0618 5.6 200 28 1000 0.0103 0.0618 5.6 200 33.6 1200 0.0102 0.0612 5.6 200 39.2 1400 0.0099 0.0594 5.6 200 44.8 1600 0.0095 0.057 5.6 200 50.4 1800 0.0092 0.0552 5.6 200 56 2000 0.0091 0.0546 FI G . 9 Static Simulation Teat Data #1 Calculation of MAX. Deflection Reduction Initial Deflection « 0.0912 in Final Deflection = 0.0546 in Deflection Reduction (% ) = 100% - [(0.0546 z 100) 1 0.0912] = 40.13% 34 POINT LOAD PRESTRESSING FORCE DEFLECTION STATIC (lb) O b ) (in) SCALED REAL SCALED REAL SCALED REAL 11.2 400 0 0 0.038 0.228 11.2 400 2.8 100 0.0355 0.213 11.2 400 5.6 200 0.033 0.198 11.2 400 8.4 300 0.0315 0.189 11.2 400 11.2 400 0.03 0.18 11.2 400 14 500 0.028 0.168 11.2 400 16.8 600 0.027 0.162 11.2 400 19.6 700 0.026 0.156 11.2 400 22.4 800 0.0258 0.1548 11.2 400 25.2 900 0.0248 0.1488 11.2 400 28 1000 0.0238 0.1428 11.2 400 33.6 1200 0.0226 0.1356 11.2 400 39.2 1400 0.023 0.138 11.2 400 44.8 1600 0.027 0.1362 11.2 400 50.4 1800 0.023 0.138 11.2 400 56 2000 0.0224 0.1344 FIG. 10 Static Simulation Test Data #2 Calculation of MAX. Deflection Reduction Initial Deflection = 0228 in Final Deflection = 0.1344 in Deflection Redaction (% ) = 100% - [(0.1344 z 100) / 0228]* 41.1% 35 POINT LOAD PRESTRESSING FORCE DEFLECTION STATIC (lb) O b ) (in) SCALED REAL SCALED REAL SCALED REAL 16.8 600 0 0 0.068 0.408 16.8 600 2.8 100 0.0622 0.3732 16.8 600 5.6 200 0.0555 0.333 16.8 600 8.4 300 0.0518 0.3108 16.8 600 11.2 400 0.0477 0.2862 16.8 600 14 500 0.046 0.276 16.8 600 16.8 600 0.043 0.258 16.8 600 19.6 700 0.0423 0.2538 16.8 600 22.4 800 0.0412 0.2472 16.8 600 25.2 900 0.0397 0.2382 16.8 600 28 1000 0.0385 0.231 16.8 600 33.6 1200 0.037 0.222 16.8 600 39.2 1400 0.0362 0.2172 16.8 600 44.8 1600 0.0355 0.213 16.8 600 50.4 1800 0.0351 0.2106 16.8 600 56 2000 0.034 0.204 F IG . 1 1 Static Simulation Test Data #3 Calculation of MAX. Deflection Reduction Initial Deflection = 0.408 in Final Deflection = 0.204 in Deflection Reduction (%)» 100% - [(0.204 x 100) / 0.408] = 50% showing prestress vs. deflection were plotted. (Fig. 12- 14). In addition a combined graph, showing all three curves overimposed over the same set of coordinate axis, was operated. (Fig. 15) 4. Results of Static Simulation Test Test results were very encouraging. With a scaled prestressing force of 56 pounds (2000 pounds actual), 50% reduction of the deflection was achieved for all three loading conditions considered. Prestressing forces of up to 292 (scaled) pounds (10,500 pounds actual) would still be allowable for the considered sections, however, based on the plotted curves for all three test datas, it has been established that further increase in prestress force would not be feasible. This is due to the fact that all three curves demonstrate substantial reduction in deflection for prestressing forces ranging between 0 and 33.6 (scaled) pounds. For prestressing forces between 33.6 (scaled) pounds and 56 (scaled) pounds, the recorded deflection reductions start to decrease. When prestressing forces pass over 56 (scaled) pounds, the mount of deflection for each trial becomes so small, that it can be considered negligible, and is not cost justif iable. One should keep in mind, that the discussed results are only for the static simulation model, and that the D E F L E C T IO N (in) DATA # 1 (point load = 200lb; L=20ft) 0.095 0.09 0.085 0.08 0.075 0.07 0.065 0.06 0.055 0.05 200 600 400 800 1000 1200 1400 2000 1800 PRESTRESSING FORCE (lb) FIG. 12 Results of Static Simulation Teat #1 D E F L E C T IO N (in) DATA # 2 (point load = 4001b; L=20ft) 0. 23; 0.22 0.21 0.2 0.19 0.18 0.17 0.16 0.15 0.14 0.13 200 600 1000 400 800 1400 1800 1200 1600 2000 PRESTRESSING FORCE (lb) FIG. 13 Results of Static Simulation Test #2 U) ( D O D E F L E C T IO N (in) DATA # 3 (point load = 6001b; L=20ft) 0.45 0.4 0.35 0.3 0.25 0.2 400 200 600 1000 PRESTRESSING ' 800 1000 1200 PRESTRESSING FORCE (lb) 1400 1600 1800 2000 FIG. 14 Results of Static Simulation Tost #3 L0 < £ ) D E F L E C T IO N (in) combined data #1, #2, #3; L = 20ft 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 200 800 1400 1800 1000 1200 1600 2000 0 400 600 PRESTRESSING FORCE (lb) DATA #1 -EEb DATA # 2 DATA # 3 FIG. 15 Results of Combined Static Simulation Tests #1,2, and 3 o limits of loading and deflections in other experiments performed could be somewhat different. The results acquired from testing the static simulation model were very convincing and they become a stepping stone for continuing the research with a full size mock-up. Chapter 6 - Testing of Full Size Mock-up Without Plywood 1. Full Size Mock-up. Based on the test results of the static simulation model, a full size mock-up (20,x20*) of an actual floor system was built and tested at the University of Southern California, as a part of continuing research work on the proposed idea. (Fig. 16-17) Testing of the Static Simulation Model had reveled a number of minor problems which were easily improved upon, and incorporated into the construction of full size mock-up. The behavior of the new structure and its improved parts were tested and are described below. One of the major improvements on the mock-up was placement of two parallel rows of blocking. (Fig. 18-19) Originally only one row of alternating blocking was used. However, that produced local bending in the weak axis of the joists due to eccentrically applied compression loads. These loads were exerted by the ends of 42 FIG. 17 Full Size Mock-up. " S S S 44 alternating blocking under prestressing load. With two rows of blocking 4" apart and the prestressing cable running in between, the bending of joists was prevented. (Fig. 20) 2. Prestress Mechanism The prestress device that puts a tension force into the floor system was revised during experimentations to make it easier and faster to assemble. (Fig. 21-22) The final device consists of: 1. f'diam. steel cable (Premium Whyte Strand 1WRC) SPC MacWhyte Co. 2. Large Ring Eye SA149-8 SPC MacWhyte Co. 3. 5/8"diam. adjustable threaded Stud bolt (I-Bar), or 5/8Mdiam. x 8" M.B. welded to 0 steel ring. 4. 2 - 6" x 6" x 3/8" Steel plates. 5. Easy-rig stud terminal ER7-8 MacWhyte Marine Products 6. Steel, tension spring a) body length = 4.248" b) overall length = 7.25" c) overall diam. = 3.00" d) wire diam. = .531" e) max. load capacity = 2500# Hardware for the prestressing device was designed in such a way that it does not require any field swaging or welding. The diam. steel cable is swaged into a large 45 Fig.22 1/4" 0 steel cable 3/4" or 5/8" T & G Plywood Double row of blockings See Fig.21 FIG. 20 Mock-up Plan View. 46 2x12 rim joist hex nut 6"x6"x3/8" thk steel plate easy-rig stud terminal 1/4" 0 steel cable 2x12 blockings FIG.21 Easy-Rig Terminal. 47 - A 2x12 joists ( § > 16" o.c. \_ 1/4" 0 steel cable large ring eye steel/tension spring 2x12 blockings 5/8" 0 stud bolt (I-bar) 2x12 ___ blockings hex nut 6"x6"x3/8"thk. steel plate 2x12 rim joist FIG. 22 Prestressing Mechanism. ring eye in the shop, and comes in unlimited length. The other end of the steel cable, after passing through the floor system was cut to the desired length and put into the Easy-Rig stud terminal. Easy-Rig Terminals create an attachment equal to the strength of the cable itself. A unique locking cone inside an Easy-Rig actually rips tighter as tension is put on the rigging. The wires and strands in an Easy-Rig swageless assembly have a progressive compression load upon them. (Fig. 24) Prestressing force is applied by tightening the nut on the threaded stud bolt to stretch the steel spring with bent hooks at each end. One of the hooks snaps into large ring eye, other than 5/8” 0 stud bolt (I-Bar). When the hex nut is tightened it causes the stud bolt to pull out, to stretch the spring so the load is applied. The amount of prestress applied is controlled by the travel length of the spring. Calibration of the spring in our experimentations was equal to 170# of prestressing force, for every 1/10” of travel distance. (Fig. 23) Two steel plates (6" x 6” x 3/8” thk.) at each end of the prestress device are placed to equally distribute and transfer applied prestress loads straight into both rows of blocking. 3. Shrinkage Change in moisture content and shrinkage of lumber are two very important problems that could adversely mm FIG. 23 Prestressing Mechanism FIG. 24 Easy-Rig Terminal influence the performance of the proposed, prestressed wood floor system; Strong consideration was given to this. Although there are thousands of different types of trees, the wood cells that comprise the bodies of these trees consist of three major components. These components are cellulose, lignin and water (Breyer, D. 1980). Cellulose represents the frame and walls of the cell, while the lignin works like an adhesive and is responsible for binding and keeping cells together. The best of the cell’s body is filled with a water, and the amount of this water determines the moisture content of the wood. The moisture content is measured as the percentage of the moist weight to the dry weight of the wood (Breyer, D. 1980): Moist weight - Dry weight MC = --------------------------- x 100% Dry weight In some trees the weight of the water is twice the weight of solid, dry materials. That means, that these trees may have moisture content as high as 200%. In the United States, structural lumber arrives at the construction site with moisture content, sometimes, as high as 19%. However, as specified by the code, the 51 equilibrium moisture content (EMC) of the covered structure (dry conditions) should range between 7 to 14 percent. In the southern and western U.S., where whether conditions are relatively dry, moisture content is set to be 9%, while in the northern states it is specified to be 12%. As mentioned before, moisture content and shrinkage can play a significant role in the performance of the prestressed floor system. Because of shrinkage, the width of the floor structure shortens by a fraction of an inch, enough to decrease the prestressing force in the steel cable. The amount of shrinkage that a structure may undergo depends on the kind of shrinkage involved. There are 4 different kinds of shrinkages: 1. Radial 2. Tangential 3. Longitudinal 4. Volumetric Longitudinal shrinkages can be neglected due to insignificant change in length. Since we are only concerned with changes in the width of the structure, only radial and tangential shrinkages are of interest. Radial and tangential shrinkages relate to the orientation of annular rings in the lumber. To calculate any of the four shrinkages mentioned above the following 52 formula can be used: A = ( M C | ~ EM C ^ x D» s (3000/SV ) - 30 + M C j where: Dj = Initial Depth D2 = Final Depth MCj = Initial MC SV = Shrinkage Value. EMC = Equilibrium Moisture Content The only variable in the formula above is the shrinkage value (SV) , which is different for each of the shrinkages discussed. According to the American Institute of Timber Construction, the Shrinkage Value for Tangential Shrinkage is 7.6%, and that for radial shrinkage is 4.8%. Hence the shrinkage for lumber, with tangential orientation of annular rings, will be higher (Breyer, D. 1980) . For the full size mock-up under consideration, shrinkage of the blocking between the joists is longitudinal and can be neglected. On the other hand, shrinkage of the joists can be either tangential or radial depending on the lumber used. For that reason, we can calculate the shrinkage of joists for the width of entire floor using (SV = 7.6) for tangential shrinkage, to generate more conservative results. 53 For single joist shrinkage would be: A = (iff. I*5--- _ o.0391" s (3000/7.6) -30+19 For the entire width of the floor which involves 16 joists the total shrinkage would be: A = 16 x 0.0391" = 0.6256" = 5/8" stot. Based on the theoretical calculations, shrinkage for the width of the entire floor should not exceed 5/8". One of the major reasons why plywood was not applied to the mock-up in the initial stages of testing, is to have exact measured value of the shrinkage of the floor without plywood resistance. Plywood is very effective in resisting a compressive forces and would partly resist deformation of the floor due to shrinkage of lumber. To take the exact shrinkage measurement after initial testing without plywood, the mock-up was left for 2 months under the sun for natural drying under a constant prestressing force of 2,500 pounds. 4. First Testing of the Mock-up without Plywood. In this portion of research two sets of dynamic tests were conducted. The first set of testing of the mock-up was performed immediately after construction. At this initial stage the moisture content in the members was still very high. Speed-chart recorder was used to record the dynamic 54 response of the mock-up. (Fig. 25) This way, accurate readings were collected, which describe the deflectory motion of the mock-up continuously, throughout the whole process of testing. Although not needed for measurements of deflections produced by dynamic impact loadings, a dial strain gage was set up, at the expected location of maximum floor deflections to measure any long duration deflections caused by creep or changes in vertical direction due to application of prestressing forces. (Fig. 26) To produce adequate and constant impact force during the testing a special frame, made of wood, was created, by which the weights used for impact were supported. The cable, from which the weights were suspended, was operated manually. Each time the weight was pulled to a preset height of three inches, before letting it strike the surface of the mock-up by a free fall. To ease the job of the operators conducting the test, the cable with the weights hanging from it was passed through a series of pulleys. This reduced the weight being pulled-up considerably and made operation much easier and faster. Sand bags in a wood crate were used as weights. (Fig. 19) For the first half of Part 2 the weight of 175 pounds was used as a vertical, impact force. 5. Testing of the Mock-up. The initial strike was done and recorded with r 55 ■ m m m x m FIG.26 Strip Chart Recorder and Strain Gage prestressing force set to 0 pounds. Next measurement was done with prestressing force of 850 pounds. At the consequent strikes the prestressing force was gradually increased at the increments of 170 pounds which was achieved by stretching of the spring in the prestressing device, each time 1/10 of an inch. There were a total or 11 trials in this portion of testing, with a final prestressing force of 2,500 pounds. The speed-chart recorder was set to a following calibration during this portion of the experiment. Sensitivity = 5 millivolts/div. Speed = 25 mm/sec. 6. Test Results The results were in accordance with calculations and previously done tests with the static simulation model. Deflection reduction of 46.2% was achieved with 2380# pound prestressing force. (Fig. 27-29) Reduction of static deflection measured with the dial gage was 0.050 in. In other words, by applying a prestressing force of 2,500 pounds to the structure, the amount of sag of the floor due to the dead load had decreased by almost a 1/16 of an inch. After this portion of testing was completed the mock-up was left to dry naturally, under the sun for two full months. The prestressing force of 2,500 pounds was constantly applied to the mock-up during this entire 57 TRIAL POINT LOAD HEIGHT OF PRESTRESSING DEFLECTION NO. DYNAMIC (lb) DROP (in) FORCE (lb) (in) 1 175 3 0 0.21 2 175 3 850 0.163 3 175 3 1020 0.18 4 175 3 1190 0.19 5 175 3 1360 0.157 6 175 3 1530 0.152 7 175 3 1700 0.153 8 175 3 1870 0.137 9 175 3 2040 0.137 10 175 3 2210 0.127 1 1 175 3 2380 0.113 FIG. 27 Dynamic Teat Data #1 Calculation of MAX. Deflection Reduction Initial Deflection = 0.210 in Final Deflection = * 0.113 in Deflection Reduction (% ) - 100%-[(0.113x100)/0.210] * 46.2% D E F L E C T I O N ( i n) DEFLECTION AS A FUNCTION OF PRESTRESSINQ FORCE 0.21 0.2 0.19 0 . 1& 0.17 0.18 0.15 0.14- 0.13 0.12 0.11 1250 250 500 750 1500 1750 2000 0 1000 2250 2500 PRESTRESSING FORCE (lb) FIG. 28 Results of Dynamic Test #1 in oo 59 ^/UK- Actual Recording of Initial Deflection Actual Recording of Final Deflection i c m FIG.29 Actual Dynamic Recording (Test # 1) 60 drying period. 7. Second Testing of the Mock-up Without Plywood. Second testing of the Mock-up had started after the mock-up had undergone 2 months of natural drying under the sun. The procedure for testing was exactly the same as in the previous case. The only two differences were the increase of the vertical, impact force from 175 pounds to 200 pounds, and shrinkage. The increase of dynamic loading to 200 pounds was done to investigate the behavior of the structure under different loading conditions. Total of 9 readings made. Each new reading was done under increased prestressing force. 8. Shrinkage Before conducting the second portion of the experiment, shrinkage of the mock-up and loss of the prestressing force originally exerted on the structure, were measured. The shrinkage across the entire width of the floor structure was 5/8 inch; which was exactly in accordance with our theoretical findings. This amount of shrinkage can be considered as maximum, since there was no plywood applied to the mock-up to resist any shrinkage. Also, 61 prestressing force has been applied continuously throughout the whole process or drying. Shrinkage of 5/8" or 1/32" per foot, for the overall floor span of 20*, can be considered negligible, because it can easily be lost in spaces between plywood sheets and due to the movement of nails. Because of the above considerations , plywood layout with long direction of the parallel to the direction of the joists is preferable. Due to the shrinkage of the floor by an amount of 5/8", prestressing force applied to the floor structure had lost 1060 pounds of its original capacity of 2,500 pounds. To accommodate for loss due to shrinkage, an additional 55 of prestressing force should be applied for each foot of floor’s width, besides the originally applied prestressing force of 2,500 pounds. For example, the required prestressing force (P.F.) for a 20’ wide floor structure should be: P.F. = 2500 lb + (20 ft. x 55 lb/ft.) = 3600 lb 9• Deflection As in the first half of this experiment, deflection reduction of 46.2% was achieved with 2720# of prestressing force applied to the structure. (Fig. 30- 32) These results have proven two important points: 62 TRIAL POINT LOAD HEIGHT OF PRESTRESSING DEFLECTION NO. DYNAMIC (lb) DROP (in) FORCE (lb) (in) 1 200 3 0 0.26 2 200 3 850 0.24 3 200 3 1020 0.222 4 200 3 1190 0.212 5 200 3 1360 0.2 6 200 3 1700 0.19 7 200 3 2040 0.168 8 200 3 2380 0.164 9 200 3 2720 0.14 FIG. 30 Dynamic Taat Data #2 Calculation of MAX. Deflection Reduction Initial Deflection = 0.260 in final Deflection = 0.140 in Deflection Reduction (%) * 100% - [(0.140 x 100) / 0.260] = 46.2% D E F L E C T IO N (in) DEFLECTION AS A FUNCTION OF PRESTRESSING FORCE 0 . 2 6 1 0.24 0.22 0.2 0.18 0.16 0.14 250 500 750 1000 PRESTRESSING 1250 1500 1750 PRESTRESSING FORCE (lb) 2250 2500 2750 2000 FIG. 31 Results of Dynamic Test #2 64 Actual Recording of Initial Deflection j&L Actual Recording of Final Deflection FIG-32 Actual Dynamic Recording (Test #2) 65 First, that shrinkage has little if any influence on the performance of the proposed system, provided the initial prestressing force is increased as described above. Secondly, the consistency of three different tests discussed so far suggests that the proposed idea does work and can be readily used in wood floor systems to reduce deflection caused by applied loads. Chapter 7 - Testing of the Full Size Mock-up With Plywood Applied 1. Mock-up preparation. This test was the last one conducted to investigate the vibration and deflection reductions in the wood floor system. The major difference between this and the preceding tests was the application of 5/8” tongue and groove plywood to the top of the floor joists. (Fig. 33) Two main objectives of this test were: 1) to determine the degree to which the plywood would participate in the distribution of vertical impact loads to adjacent joists, and 2) to determine if the proposed prestressing mechanism would perform as effectively, when the plywood is applied to the floor. 66 / » FIG. 33 Mock:-up with Plywood Applied At the time of plywood application to the floor joists, the mock-up was under full prestress of 2500 pounds. Prestressing of the floor before plywood application should always be exercised, any time this product is used. This is done to avoid gaps and spaces between blocking and joists, which could later considerably reduce the tension in the cable once the prestressing is applied. 2. Testing of the Mock-Up The procedures for testing the mock-up with plywood, were the same as the procedures used for the previous tests. An impact load of 200 pounds was used, with prestressing forces gradually increasing from 0 to 2500 pounds. Readings were recorded by the speed-chart recorder, with sensitivity set at 5 millivolts/division and chart speed at 20 millimeters/sec. The recordings done by the speed-chart recorder had several advantages. The main advantage was the availability of the curve from which exact deflection measurements were made. Secondly, recorded diagrams presented the time history of vibration, starting with the main shock due to the initial impact and following by the vibration dissipation recordings. (Fig. 34) Knowing the speed of the recorder and counting the number of the cycles that follow the 68 Actual Recording of Initial Deflection Actual Recording of Final Deflection FIG.34 Actual Dynamic Recording (Test #3) 69 initial impact in a set period of time, the frequency of the floor oscillation can be established. The data gathered during the experiment has been analyzed and plotted, and it is discussed in the following paragraphs. 3. Test Results Performance of the prestressing mechanism has not been affected by the increased stiffness of the floor, due to the plywood application. Deflection reduction of 45.3% has been achieved with 2380 pounds of prestressing force applied to the Mock-up. There were 10 readings recorded and they are presented in tabular form. (Fig. 35) Graph of prestressing force versus deflection has been plotted to present the performance of the floor under prestressed conditions. (Fig. 36) Application of plywood to the mock-up had significantly changed the stiffness of the floor. Based on the theory, about 15% reduction of the deflection had been expected between the floor which has been tested without plywood, and the same floor after the plywood has been applied to it. The measured reduction in deflection (Fig. 37) was between 17 to 18 percent, depending on the portions of test data compared. In our opinion, slightly higher than expected stiffness of the floor, may have resulted from the fact, that T & G plywood had been used instead of regular plywood. 70 TRIAL POINT LOAD HEIGHT OF PRESTRESSING DEFLECTION NO. DYNAMIC (lb) DROP (in) FORCE (lb) (in) 1 200 3 0 0.211 2 200 3 850 0.17 3 200 3 1020 0.17 4 200 3 1360 0.156 5 200 3 1530 0.152 6 200 3 1700 0.152 7 200 3 1870 0.15 8 200 3 2040 0.139 9 200 3 2210 0.127 10 200 3 2380 0.115 FIG. 35 Dynamic Tact Data #3 Calculation of MAX. Deflection Reduction Initial Deflection = 0.211 in Final Deflection = 0.115 in Deflection Reducti on (% ) * 1 0 0% - [(0. 115 x 100) / 0.211] * 4 5. 5 % D E F L E C T IO N (in) DEFLECTION AS A FUNCTION OF PRESTRESSING FORCE 0.22 0.21 0.2 0.19 0.18 0.17 0.16 0.15 0.14 0.13 0.12 0.11 250 750 1000 1250 500 1500 1750 2000 2250 2500 PRESTRESSING FORCE (lb) FIG. 36 R esults of Dynamic Test # 3 D E F L E C T IO N (in) 0.28 0.261 0.24 0.22 0.2 0.18 0.16 0.14- 0.12 0.1 500 1000 1500 2000 2500 3000 PRESTRESSING FORCE (lb) DATA # 2 DATA # 3 FIG. 37 Results of Combined Dynamic Tests #2 and #3 <l t-o No change in static deflection, using the dial strain gages, has been recorded. In previous tests of the mock-up without plywood applied, reduction of static deflections due to the dead load of the structures, had been recorded. This reduction was due to the applied high prestressing forces. The reason why no reduction in static deflection was recorded in the last experiment is because the plywood, which was nailed to the floor, started to act as a tensile membrane that resisted any upright movement of the floor. Once again the experiment had shown that the proposed prestressing mechanism does work, and that its contribution to reducing the deflection of the wood floor systems, caused by dynamic impact loads, is very signif icant. Chapter 8 - Perception Test on Human Response to Vibration and Noise of the Floor System. 1. Human Perception of Vibration Vibration or unstable periodic movement of the environment around us is annoying and generally is very much disliked. For human beings the range of sensitivity to vibration is much greater than the range of hearing. Mechanical vibrations as low as 0.5 Hz and as high as 100 Hz are well in the range of human perception (Cowan, 1991). Although physically unharmful, mechanical vibrations create strong psychological discomfort. Many structures occupied by people are very vulnerable to mechanical vibrations of any sort such as natural frequency of building, building height, shape and materials used. Mechanical vibrations can also be induced in buildings by a operating machinery or by any other dynamic objects like passing cars, people walking on the floor, rolling carts, etc. (Burkhardt, L.R. 1961). The range of Mechanical Vibrations perceptible to humans can be subdivided into three main groups. Group #1 relates to the vibrations in the range between 0.1 to 3 Hz. Usually very large structures such as toll buildings, large ships, long suspension bridges are susceptible to vibrations in this range. Since the frequency of vibration is so low, the distraction and annoyance caused to inhabitants is slightly perceptible. Frequencies of oscillation of Group #2 range between 3-30 Hz. These frequencies are strongly perceptible and responsible for great discomfort of the occupants. Vibrations of this sort are very common in everyday life and they are usually caused by such things as passing cars and trains, operating machinery, building frames, building walls and floors, human traffic etc. (Guignard, J.C. 1971). These kinds of vibrations are usually accompanied by noise, and sometimes it is hard to tell 75 which one causes more disturbance. Frequencies of oscillation of Group #3 are 30 Hz and up. Vibrations induced at such high frequencies start to lose their mechanical characteristics while the noise criteria increases noticeably. Frequencies above 30 Hz are responsible for rattling of lightweight components of the structures such as windows, hardware, light wall partitions, lightweight room furnishings, etc. Although structurally safe in most cases, these higher frequencies are very disturbing because of their audible noise, which creates high level of resentment and destruction from activities among occupants. While annoyance or resentment is the natural reaction to such identifiable man-made disturbances, real fear of personal injury, as well as property damage, arise when vibration is felt in circumstances where there is a prevailing expectation of serious trouble such as military attack, earthquakes or violent storms (Parkin, P.M. 1979). 2. Vibration Sensory Mechanism in Humans The sensory mechanism that detects vibration in humans is very complex. According to J.C. Guignard, there are numerous, and each individual mechanism works over a limited frequency range. Here what he writes about these sensory mechanisms: Whereas man has only one kind of receptor for airborne sound, namely, the hair cells of the organ of Corti in the inner ear, he 76 appreciates vibration by means of several different sensory mechanisms, involving a variety of sensors distributed throughout the body. These receptors respond differentially to vibration in various overlapping frequency ranges. They differ both in the effective bandwidth of their response and in the degree of temporal integration or other processing of the information which they send to the brain Tl, 3, 4]. Moreover, the human body, viewed as an engineering construction, possesses a complex elasticity, allowing many modes of vibration. The damping is less than critical in some of these modes, so that differential motion of body parts with resonant amplification of vibration can occur [1, 5]. This reduces the effective sensory threshold at certain frequencies, notably in the range 1-10 Hz. Seeing and hearing the surroundings vibrate are examples of indirect sensory mechanism of vibration. The organs of direct sensation are of two main types. First, the organs of balance connected to the inner ear, act with their central connections in the brain as integrating angular and linear accelerometers working within a limited bandwidth. This bandwidth is related to their principal function of signalling tilts and turns of the head induced in the normal range of bodily activity and movement. The information which they send to the brain can be ambiguous (disorientation) and sometimes physiologically disturbing (motion sickness) if these organs are stimulated by sustained passive motion outside the normal range of activity. Second, vibratory forces and displacements are sensed by large numbers of small mechano- receptors distributed throughout the body. These again are of different kinds, having different primary functions. Some are slowly- adapting and respond to sustained or slowly changing mechanical stimuli. They are found in the muscles, the tendons and the joint capsules and provide the feedback of information about load and position from the limbs and the trunk which the central nervous system needs for the normal regulation of static and active posture. Another kind of receptors is found widely distributed in the skin and in connective tissue inside the body. 77 These receptors, generally rapidly-adapting, form a band-limited vibro-tactile array responding most strongly to vibrations ranging from some 30 Hz up to well into the audio frequency range. They subserve the tactile sensation which, for example, enables us to appreciate surface texture, but they are also the organs mainly responsible for detecting ground and structure-borne vibration at the higher frequencies. There is still considerable debate among physiologists as to which kinds of mechano-receptor are of primary import in different ranges of frequency [1, 3, 4]. Based on the theory and the facts, brought earlier about vibration and its influence on people*s sensory system, the necessity of our research for reducing vibrations, noise and reflection of wood floor systems becomes quite obvious. Generally, vibrations of wood floors belong to the second group of vibrations described earlier, which have frequencies ranging bet 3-30 Hz. However floor vibration frequency is not the only factor causing disturbance. Because of the flexibility of wood members, especially for spans over 16’, the amplitude of the vibration generated by the cyclical deflection of the joists becomes a major factor for concern. The aim of prestressed wood floor system was to deal with the reduction of deflection and noise of flexural members (joists) in the floor system. Prestressing of the floor reduces the amplitude of vibration of joists which in turn makes the vibration of the floor less destructive and annoying for the occupants. 3. Perception test. There was a perception test conducted at USC as the part of continuing research process. A group of 43 students was asked to participate in the testing process. The testing procedure was as follows: First, students were asked to walk across the floor mock-up which had no prestress applied to it, and record their perception of vibration and noise on test sheet describing the strength of experienced sensations. (Fig. 38) The choices provided, were put in descending order from the strongest to the list noticeable one as follows. 1) Strongly perceptible 2) Distinctly perceptible 3) Slightly perceptible 4) Not perceptible All students, participating in the test, have walked across and have recorded their perceptions of vibration and noise of the unstressed wood floor system. That completed the first portion of the perception test. In the second portion of the test the floor system had been prestressed with 2500 pounds. The students were then asked to follow the same instructions as in Part 1. The data sheets were collected and analyzed. 4. Results of the Perception Test The results of the perception test were very 79 INSTRUCTIONS: W A L K ACROSS T H E FL O O R M O C K U P A N D R EC O R D T H E D E G R E E O F P E R C E P T IB IL IT Y OF FL O O R V IB R A T IO N A N D N O ISE IF A N Y . TEST 1 TEST 2 V IB R A T IO N N O ISE V IB R A T IO N N O ISE STRONGLY P E R C E P T IB L E DISTINCTLY P ER C E P TI B L E SLIGHTLY P E R C E P T IB L E NOT P E R C E P TI B L E FIG. 38 Perception Test Instructions to Participants TEST 1 ( P RE S T R E S S I N G F O R C E = O tt> ) TEST 2 ( P R E S T R E S S I N G FOfl C£=250ab) R A N G E O F PERC EPTIO N V IB R A T IO N (% ) N O ISE (% ) V IB R A T IO N (% ) N O ISE (% ) ST R O N G L Y PE R C E PT IB L E 27.9 18.6 7 7 D IS T IN C T L Y PE R C E PT IB L E 34.9 55.8 9.3 11.6 S L IG H T L Y PE R C E PT IB L E 32.6 20.9 58.1 51.2 N O T P E R C E PT IB L E 4.7 4.7 25.6 30.2 FIG. 39 Perception Teet Data 80 impressive. They have fully backed the theoretical and experimental findings and have proven that the idea has not only a technical advantages but it also has major significance for psychological comfort of the occupants. As it has been expected, vibration levels have dropped significantly when prestress have been applied to the floor system. (Fig. 39) Without prestressing applied, the majority of test participants have felt that vibration was in strongly perceptible and distinctly perceptible ranges. On the other hand, with prestress applied to the floor system the majority of participants felt vibrations in slightly perceptible or not perceptible ranges. Vibration fell from 27.9% to 7.0% in the strongly perceptible range and from 34.9% to 9.3% in the distinctly perceptible range. (Fig. 40-41) The perception test has also shown that the reduction of noise, using the prestressing floor mechanism, was also very successful. (Fig. 39) In the first part of testing, with no prestressing force applied, a majority of students experienced noise levels in the strongly and distinctly perceptible ranges, same as during vibration testing. However, after applying prestressing force of 2500 pounds to the floor system, the majority of participants in the test experienced noise levels in the slightly and none R A N G E O F PE R C EPT IO N (PRESTRESSING FORCE * 0) STRONGLY PERCEPTIBLE DISTINCTLY PERCEPTIBLE SLIGHTLY PERCEPTIBLE 100% PERCENTAGE OF TEST PARTICIPANTS FIG. 40 Results of Vibration Perception Test #1 C O R A N G E O F PE R C EPT IO N (PRESTRESSING FORCE = 2500 ft> ) DISTINCTLY PERCEPTIBLE 100% PERCENTAGE OF TEST PARTICIPANTS FIG. 41 R esults of Vibration Perception T est # 2 CD to R A N G E O F PE R C EPT IO N (PRESTRESSINQ FORCE ~ 0) 18.6% STRONGLY PERCEPTIBLE SLIGHTLY PERCEPTIBLE 100% PERCENTAGE OF TEST PARTICIPANTS FIG. 42 Results of Noise Perception T est #1 00 LO R A N G E O F PE RC EPT IO N (PRESTRESSING FORCE * 2500(b) STRONGLY PERCEPTIBLE DISTINCTLY PERCEPTIBLE SLIGHTLY PERCEPTIBLE 100% PERCENTAGE OF TEST PARTICIPANTS FIG. 43 Results of Noise Perception Test #2 00 4^ perceptible ranges. According to collected data noise levels in the strongly perceptible range had fallen from 18.6% to 7.0% and in distinctly perceptible range they had fallen from high 55.8% to a surprisingly low 11.6%. (Fig. 42-43) The perception test, involving the average human reaction and response to vibration and noise of the floor had become one more proof to already theoretically and technically proven parts that the prestress in wood floor systems works and that its contribution to achieving better structural floor systems can be very considerable. CONCLUSIONS The problem of vibrating floors had been of great interest to architects and civil engineers for long time. Different structural floor systems have been discussed. Vibration and noise of wood floor systems have been investigated. Four major experiments conducted in this research, were very successful. Deflection and noise reductions of over 45% were achieved with the proposed prestressing device for structural wood floor systems. Tests have been conducted both for static simulation 86 model and (20,x20’) full size mock-up (with and without plywood). In addition, human perception test has been conducted to investigate human response to noise and vibration of the floors. The consistency in noise and deflection reduction, in the experiments above, prove the adequacy and need for the proposed prestressing device. The cost efficiency of the proposed prestressing system is another major factor justifying its usefulness. At the present time, a patent application has been filed for the proposed device, for prestressing wood floor system, with the U.S. Department of Patents and Copyrights, with help of USC*s patent department. 87 References American Institute of Timber Construction (1974), Timber Construction Manual. 333 West Hampden Ave. , Englewood, Co. Anderson, L.O. and Taylor F. Winslow (1987), Wood - Frame House Construction. Craftsman Book Co., Carlsbad, Calif. Baxter, R.L. Dynamic Balancing. Sound and Vibration, Vol. 6, April 1972, pp. 30-33. Benjamin, B.S. (1982), Building Construction For Architects and Engineers. Ashnorjan Benzaleel Publishing Co., Lawrence, Kansas. Breyer, D.E. (1980), Design of Wood Structures. McGraw- Hill, Inc. N.Y. Burkhardt, L.R., Vibration Analysis For Structural Floor Systems. Journal of the Structural Division, Preceedings of the American Society of Civil Engineers, Vol. 87, No. St. 7, October 1961. Clough, R. (1986), Construction Contracting. John Wiley & Sons, Inc., N.Y. Cowan, H.J. (1991), Handbook of Architectural Technology. Van Nostrand Reinhold, N.Y. Dietz, A.G.H. (1974), Dwelling House Construction. 4th Ed., The MIT Press, Cambridge, Massachusetts. Griffin, M.J. and E.M. Whitman, The Discomfort Produced By Impulsive Whole-Body Vibration. Journal of the Acoustical Society of America, Vol. 58, 1980, pp. 1277-1284. Guignard, J.C., Human Sensitivity to Vibration. Sound and Vibration, Vol. 15, 1971, pp. 11-16. Hornbostel, C. (1978), Construction Materials. John Wiley & Sons, Inc., N.Y. Hoyle, R.J. (1979), Wood Technology in the Design of Structures. Mountain Press Publishing Company, Missoula, MT. 88 KC Metal Products, Inc. (1978), Rough Carpentry. Wood Framing Systems. San Jose, CA. Mens (1987), Building Construction Cost Data 1988. R.S. Means Company, Inc. Publishers, Kingston, MA. National Forest Products Association (1977), National Design Specifications For Wood Construction. Washington, D.C. National Lumber Manufacturers Association, Manual for House Framing. Washington D.C. Parkin, P.M., M.R. Humphreys, and J.R. Cowell (1979), Acoustics. Noise and Buildings. Faber & Faber, London, Boston. Plunkett, R. , Shock and Vibration Instrumentation. Shock and Vibration Digest, Vol. 14, Sept. 1982, pp. 3-5. Popov, E.P. (1978), Mechanics of Materials. Prentice-Hall, Inc., Englewood Cliffs, New Jersey. Rao, S.S. (1986), Mechanical Vibrations. Addison-Wesley Publishing Company, Inc., Menlo Park, Calif. Reid, E. (1984), Understanding Buildings. The MIT Press, Cambridge, Massachusetts. Trus Joist Corporation Catalog (1984) TT International Inc., Boise, Idaho.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The response of high-rise structures to lateral ground movements

PDF

Investigation of sloshing water damper for earthquake mitigation

PDF

Investigation of seismic isolators as a mass damper for mixed-used buildings

PDF

Return to the moon: MALEO, Module Assemby in Low Earth Orbit: A strategy for lunar base build-up

PDF

Energy performance and daylighting illumination levels of tensile structures in an extreme climate

PDF

Statistical analysis of the damage to residential buildings in the Northridge earthquake

PDF

The Indian Himalayan building energy code as a step towards energy conservation

PDF

A computer teaching tool for passive cooling

PDF

Computer aided value conscious design

PDF

Comparison of lateral performance: Residential light wood framing versus cold-formed steel framing

PDF

Computer aided design and analysis of anticlastic membranes and cable nets

PDF

A proposal for the Indian National Lighting Code

PDF

Computer aided form-finding for cable net structures

PDF

From nescience to science and beyond: A critical investigation of 'building' in cyclone prone areas

PDF

Passive cooling methods for mid to high-rise buildings in the hot-humid climate of Douala, Cameroon, West Africa

PDF

A proposed wood frame system for the Philippines

PDF

Terrace housing: A conscious solution

PDF

Computer aided design and manufacture of membrane structures Fab-CAD

PDF

Behaviour of hipped roof dwellings in response to wind forces: A comparitive study of thatch against conventional roofs

PDF

Hebel design analysis program

Asset Metadata

Creator Ayrapetyan, Ruben (author)

Core Title Vibration reduction using prestress in wood floor framing

Degree Master of Building Science

Degree Program Building Science

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

Tag engineering, architectural,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Schierle, Gotthilf Goetz (committee chair), Ambrose, James (committee member), Vergun, Dimitry K. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c20-299491

Unique identifier UC11259161

Identifier EP41427.pdf (filename),usctheses-c20-299491 (legacy record id)

Legacy Identifier EP41427.pdf

Dmrecord 299491

Document Type Thesis

Rights Ayrapetyan, Ruben

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA