Final Capstone Project Reports- Master of Science in Structural Engineering Program

Beane, Jeffrey “The Effects of Extreme Temperature on the Bond Performance of Epoxy-Bonded Threaded Rods in Unreinforced Concrete”

August 2012, 40pp, 5 references, figures, tables
Available for checkout
Abstract: The effects of extreme temperatures on the bond performance of threaded rods and structural adhesives in unreinforced concrete are not well understood. Manufactures of structural adhesives have conducted numerous tests to establish ultimate capacities of post-installed threaded rods embedded in concrete at room temperature loading. They also have established reduction factors of those capacities when the base material has been subjected to a constant elevated temperature for 24 hours. This program was conducted to better understand the interaction between bond performance and ultimate tensile capacity when the concrete base material is subjected to short duration extreme temperatures.Tests were performed on a total of thirty-five post-installed anchors. Test parameters include: 2-inch and 5-inch embedment depths, two structural adhesives and four temperature categories. Direct heat was applied to the base material surrounding the anchor until the desired surface temperature was reached, at which time constant tensile loads were applied directly to the anchor until failure.

Due to a limited number of test specimens, definitive conclusions to the interaction between bond performance and extreme temperatures were not determined. However, the tests were used for comparison to manufacturer baseline capacities and to highlight the importance of safety factors, as well as other reduction factors to consider when design professionals specify anchors that may be subjected to extreme temperatures.

Bernero, Alison “Improving the Seismic Capabilities of the Schools in Joyabaj, Guatemala”

May 2010, 97pp, 4 references, appendices
Available for checkout
Abstract: Most of the villages in the region of Joyabaj, Guatemala lack the funds and governmental support necessary to construct schools. NGOs have stepped in to build schools per a standard, but minimally engineered design. Construction is led by local masons and the building methods are traditional to the region. Earthquakes continue to be common in the area, and the seismic capabilities of these schools are a concern. The Joyabaj school structure is a textbook example of confined masonry construction. Masonry panels are framed by reinforced concrete confining members. In May of 2009, the author travelled to Joyabaj region to gather samples of the material used in typical school construction and to observe typical school construction methods. Flexural reinforcement available in Joyabaj was subject to tensile tests. The rebar used by MSOE-EWB possessed strengths similar to those of Grade 60 bars in the United States and can be classified as Grade 60 rebar. The bars in Joyabaj, however, were too weak to receive the same classification. They fall into the Grade 40 classification. Aggregate samples were subject to sieve analyses in-country and brought back to the U.S. for further visual inspection by a qualified Geotechnical Engineer. Fine aggregate samples were taken from distributor Que Maya and local construction sites. The Que Maya sand proved to be of significantly higher quality than the sand typically used by the masons, which was filled with strength-compromising mica. Coarse aggregate was taken from distributors Que Maya and Joyabaj Quarry. The coarse aggregate from Que Maya had several advantages over the Joyabaj Quarry aggregate used by EWB-MSOE and was overall more suitable for concrete. Mortar and concrete samples were made using Guatemalan-produced Cemento Progreso. Compression tests were performed on three batches of mortar cubes containing different combinations of Guatemalan and U.S. sand and cement. Results suggested that both the sand and the cement used in Joyabaj are of a lower quality than what is typical in the U.S. However, the mortar made with Guatemalan Cement, althougher weaker than U.S. mortar, still exceeded strength expectations. The sample which included Guatemalan sand was less than half the required strength. It was concluded that with the proper mix ratios and aggregate of a decent quality, Cemento Progreso can provide adequate mortar strength. The concrete, mixed using the “Volcano Method” by Joyabaj masons possessed a w:c ratio of approximately 0.57. The 70-day compressive strength of the concrete was found to be 825 psi. Further testing of this mix is recommended using ASTM C1019, The Standard Test Method for Sampling and Testing Grout. This paper proposed three options to achieve seismic capability. The first option, to achieve roof diaphragm action was found to be impossible due to connection strengths. The second option was to replace the steel roof with CIP concrete. This was found to be unaffordable. The final option, to improve the integrity of the non-diaphragm roof as-is was found to be possible assuming proper welding techniques and a long enough bearing length of the rafter and beam assemblies.

Blau, Andrew M. “Head and Bond Contributions to Anchorage of a Headed Reinforcing Bar”

June 2010, 94pp, 6 references, figures, tables
Available for checkout
Abstract: Headed reinforcing bars are becoming more popular in the construction industry; however, there is relatively little known about the interaction of the head and bond in resisting loads. As such, the ACI 318-08 (the 2008 version being the current version) code equation governing headed reinforcing bars appears to be a gross generalization of the capacity of a headed bar. The purpose of this project was to study the contributions of the head and bond as they change with varying bond lengths, clear covers, and head bearing areas to determine if a better model can be developed to represent a headed bar. Twelve headed bar specimens were cast, each one with a unique combination of bone length, clear cover, and head bearing area, and loaded until failure. Two failure modes were observed: bar yielding and side blowout. The loads resisted by the head and bond throughout the loading of each specimen were analyzed for trends. It was determined that headed bars follow one of two patterns: head-dominated and bond-dominated, and is dependent on bond length. As such, different models should be used for the capacity of each pattern. The specimen capacities were compared to models proposed by others, and the current ACI 381-08 code equation. The ACI 318-08 code equation was found to be a good fit with a minor modification for the bond-dominated bars. It was determined to not be possible to develop a model for head-dominated bars from this test program and that additional testing must be conducted.

Bogle-Boesiger, Larissa “Identification and Structural Properties of Untreated Lumber from the Joyabaj Region of Guatemala”

April 2016
Available for checkout
Abstract: The purpose of this research project was threefold:
1) Identify the species of wood used for construction by Engineers Without Borders (EWB) groups in the Joyabaj region of Guatemala
2) Provide grading information on lumber in the Joyabaj region of Guatemala, as well as a grading guide
3) Determine design properties of wood used for construction by Engineers Without Borders groups in the Joyabaj region of Guatemala.

The results of the project will aid in the design of wood structures by the EWB groups working in Joyabaj.

In this project specimens were collected and sent to the Forest Products Laboratory for species identification. Information on the species of wood collected was also taken from the sawyer at the mill where the specimens were collected. Visual grading rules were developed and field tested to determine the grade of the wood. The species options were then verified or rejected by results of small clear wood testing. Two types of tests were performed, static bending and compression parallel to grain, in order to obtain modulus of rupture, ultimate stress in compression parallel to grain, and modulus of elasticity. Specific gravity was also obtained from the specimens. The results showed that there were three different species collected. Design should be performed based on NDS (National Design Standard) values for a No. 3 or No. 2 Eastern White Pine, depending on the level of wood selection in the construction process.

Bolda, Jacob R. “Load Distribution and Load-Deflection Behavior of Hooked Reinforcing Bars Loaded in Tension”

May 2011, 89pp, 6 references, figures
Available for checkout
Abstract: The purpose of this capstone design project report is to discuss the behavior of a hooked bar in concrete carrying a tension force. The cover and bonded length are varied to observe the effect on the load distribution between the hook portion and the bonded length portion of a hooked bar. Each specimen had 1 or 2 inches of cover. Each specimen had a bonded length of 8, 12, or 16 inches. Straight rebar was also tested to provide a control and comparison to the hooked bars with similar cover and bond length variables.

All hooked specimens experienced a steel failure. The straight bar specimens with 8 inch bond length and the straight bar specimen with 12 inch bond length and 1 inch cover experienced a concrete failure in splitting. The straight bar specimen with 12 inch bond length and 2 inch cover and the straight bar specimens with 16 inch bond length experienced a steel failure.

The addition of the hook adds enough strength to prevent concrete failure when compared to the straight rebar with similar bond length and cover. With bond lengths less than 12 inches and minimal cover, the addition of the hook adds enough strength for the steel to reach yield. The hooked bar stiffness was larger than the straight bar stiffness in every specimen except for the straight bar specimen with 16 inches of bonded length and 2 inches of cover.
An interesting trend is the effect of bonded length and cover on load distribution between hook and bond. As the bond length increased, the bond took a greater portion of the load. At shorter bond lengths, the hook takes the greatest portion of the load. An increase in cover increased the load carried by the bond in all cases. This effect was greater at shorter bond lengths since the hook carries the greater portion of the load.

It is suggested in further work to continue to include a load cell, slip measurement, and strain gauges to allow for comparison of maximum loading, load distribution, and stiffness. The method for measuring lead and end slip might be improved. The lead slip in this capstone report was erratic or failed to record due to the LVDT slipping off the angle. The bulk of research conducted on hooks concentrates on a pullout failure. Further research and data will be required to make any statements regarding splitting-controlled hook configurations. The data obtained in this capstone report are not sufficient to create a model that can be applied to the code.

Bonnet, Jeffrey D. “Compression Chord Analysis Of A Cold-Formed Steel Joist Using Photoelasticity”

Nov. 2002, 76pp, includes references
Available for checkout
Abstract: Photoelastic analysis is an experimental tool which uses polarized light to determine stresses within a material. This project explores the validity and practicality of photoelastic analysis in structural engineering applications. This project was conceptualized from the destructive testing of the welded connection of a cold formed joist. In an attempt to utilize a less wasteful and potentially less expensive alternative, two non-destructive analysis tools are introduced — finite element analysis (FEA) and photoelastic analysis. The results from FEA and photoelastic analysis of an epoxy coated steel joist are compared to results from destructive. It was found that both FEA and photoelastic analysis did accurately predict the critical failure region from destructive testing. Also, a model of the steel joist was made using a rapid prototyping (RP) technique called stereolithography. The RP model was then compared to the actual steel joist using photoelastic analysis. It was found that if the RP model was epoxy-coated, the stress gradients behaved quite similar to that of the steel joist. In an attempt to explain any inconsistencies in the results between the steel and RP joists, a simplified model is created. Independent tests were performed isolating three variables: material, scale, and orientation. Several unexpected characteristics were discovered in these tests, mostly due to the residual stresses inherent in the RP models. The findings show that photoelastic analysis has potential for becoming a valuable tool in structural engineering, especially for specimens with complex geometries. Perhaps the most important finding was that the accuracy of the photoelastic testing depends greatly on the amount of residual stresses, and if these stresses are minimized, the analysis is more likely to remain linear and thus more quantifiable. Several ideas are recommended to guide those in the future who wish to more qualitatively explore the photoelastic characteristics of RP models.

Cervenka, Jan “Investigation of Strength of Glass Plates with Hole Subjected to Out-of-plane Bending”

May 2016
Available for checkout
Full-text available online
Abstract: Structural glass facades, distinguished architectural features, are widely used in modern building construction. Understanding the strength of such point-supported glass plates is crucial in the design process. As an analogy to this problem, this research investigates the strength of glass plates subjected to out-of-plane bending and the effects of stress concentration around a hole. The strength of fully tempered 1/2 inch monolithic and fully tempered 1/4 inch — PVB interlayer — 1/4 inch laminated glass specimens was determined using a destructive method. Four different monolithic glass specimen types were utilized: those without a hole subjected to four-point bending, those with a hole subjected to four-point bending, those with a hole loaded via a standard swivel fitting (three-point bending), and those with a hole with a chamfered edge subjected to four-point bending. This led to a deeper understanding of the behavior of glass plates near the fitting, and determined the influence of stress concentration on the overall strength of a glass plate. Ultimate stresses were obtained based on the stress concentration factor determined from the Finite Element Model (FEM), thick plate and thin plate analytical models, and readings from strain gauges applied to some of the specimens. The discoveries of this research demonstrate that the drilling method introduces a 3% reduction in the strength of a monolithic glass plate. If the load is applied through the fitting, the strength of a monolithic glass plate is reduced by an additional 13%. The breakthrough discovery made during this process is that chamfering the edge of a hole varies the peak breaking stresses, which results in 18% greater moment capacity.

Charlton, Nathan A. “Development of D45 Reinforcing Wire in Concrete”

May 2009, 51pp, 4 references, figures, tables
Available for checkout
Abstract: Reinforced concrete is a standard practice of construction throughout the world today. Traditionally concrete is reinforced with hot rolled steel bays but an alternative to this is steel wire reinforcing which is produced from an extrusion process. The overall geometry of the wire reinforcing is different than that of rolled bars, and therefore the reinforced concrete members act differently when subjected to loading. In order to examine these differences and determine the adequacy of steel wire reinforcing in regard to development length, a series of large scale tests were conducted and evaluated using a linear elastic transformed section analysis. It was found that all D45 test specimens experienced a failure caused by pullout, and the current ACI design equations do not accurately predict the bond stress at failure of D45 reinforcing wire.

Christensen, Heather “Determining the Validity of Design Provisions for HSS to Base Plate Connections with Corner Anchor Rods Subjected to Axial Tension”

August 2010, 147pp, 10 references, figures, appendices
Available for checkout
According to the American Institute of Steel Construction Hollow Structural Sections Connections Manual (AISC 1977), no published solutions currently exist to determine base plate thickness or weld design strength for an HSS column to base plate connection with corner-located anchor rods in a tensile loading condition. The manual suggests a design procedure, but it does not have physical testing to validate it. The equation suggests effective plate width, yield strength, base plate thickness, and the distance from anchor rod to HSS corner affect the calculated design strength. The objective of this research is to conduct testing to determine validity of the existing design procedure in the HSS Connections Manual or to adjust the equation as needed. To obtain experimental data, specimens with varying base plate and HSS column dimensions and thicknesses were subjected to an axial tensile loading to determine actual plate stress concentrations located between anchor rods and HSS corners. This research focused on the plate stress based equation, but weld strengths and strain were also investigated. Linear strain gages and strain rosettes were used to measure strain in base plates, HSS tubes, and welds. Plate deflection was also measured.

Experimental data and calculations showed that the elimination of a factor leftover from an error in the derivation of the equation, along with the addition of an HSS factor, R sub hss, produces comparable results for corner plate stress. The adjusted equation is: (equation). Weld results were inconclusive, but results are discussed. Contributing weld length should be further examined. Research focused on weld behavior is recommended.

Christiansen, Mary “An Analysis of Concrete Performance: Increasing Gypsum Content”

May 2008, 61pp, 17 references, figures, tables
Available for checkout
Abstract: Construction and demolition debris account for over 12% of the material in our nation’s landfills. Of that 12%, approximately one fifth is comprised of scrap gypsum wallboard. Gypsum wallboard is often sent to landfills because in many cases there is no strong end market for recycling it. The purpose of this project was to analyze the performance of concrete mixes when ground recycled gypsum wallboard is added in various amounts. Concrete is the most widely used construction material on Earth. Using gypsum as an additive in concrete could be an economical way to use a material that would most likely be thrown into the landfill. Multiple concrete mixes with varying gypsum contents (from 5 to 15% by weight) were tested for compressive strength over a period of ten weeks. Three of the concrete mixes reached the required strength of 4700 psi, though several mixes took longer than others to achieve the design strength. The batches that had more than 4% gypsum by weight added did not reach the required strength. Based on this research, it seems that for certain applications, such as backfill, foundation support, pavement base, conduit bedding, and architectural features or as fill for metal deck, a mix with a slightly increased gypsum content could produce acceptable concrete. Other properties, such as durability, shrinkage and expansion, and cracking need to be investigated further, but this experiment proves that design strength can be obtained if a long enough curing time is allowed.

Cichon, Christopher “Analysis of Combined Shear and Tensile Loading on All-Bolted, Single-Angle Shear Connections”

January 2011, 107pp, 21 references, appendices, figures, tables
Available for checkout
Abstract: In response to several recent building collapses following incidents of localized damage that developed into progressive collapse, the 2009 International Building Code has been revised to include a section regarding structural integrity. This section mandates, amongst other things, that beam to column connections must not only support the vertical reaction forces from the beam, but also a specified axial tension force. Simple shear connections, most notable the all-bolted, single-angle connection, are only designed to resist the end reaction of a beam, and their behavior when under axial tension forces have only been briefly investigated.This project investigates how the single-angle connection behaves when subjected to only an axial tension load and to simultaneous vertical shear and tension loads through a series of experiments conducted on full-scale connections. Two styles of the single-angle connections are tested; six specimens use a three-bolt configuration and three specimens use a five-bolt configuration. Testing is performed in the MSOE-CSEC test frame with the angles acting as a connection between a tension force applying actuator and the test frame. Some tests include a shear force applied by a second actuator oriented perpendicular to the tension applying actuator. This shear load is held constant through the duration of the test and is equal to 25% of the connection’s shear capacity. Overall, four of the three-bolt connections and two of the five-bolt are tested in the tension only condition. Two of the three-bolt connections and all five-bolt connections are tested with a washer between the bolt head and the inside face of the outstanding angle leg. The remaining two specimens from the three-bolt connection category are tested without the washer present. Removing the washer isolates the angle section and allows for further investigation of the angle’s behavior. The two remaining three-bolt angles and the one remaining five-bolt angle are tested under a simultaneous tension and shear condition.

Three main conclusions were reached. First, all of the tested connections met and exceeded the capacity requirements for end connections in steel structures as defined in Section 1614 of the 2009 International Building Code. Second, the simultaneous application of the axial tensile load and a constant vertical shear load did not influence the ultimate tensile strength of the connection. Third, failure of the connections was due to one of two limit states, one of them being the tension failure of the bolts connecting the angle to the test frame. The other limit state was the rupture of the outstanding angle leg on the bolt line, which resulted in the loss of stiffness of the connection and the eventual ruptures of the entire net section.

De Tennis, Danielle “Strength Testing and Analysis of a Stair Header Connection”

May 2010, 67pp, 16 references
Available for checkout
Abstract: This paper investigates the demand and calculated capacity of a post-installed stair header connection into concrete masonry unit (CMU) walls. The intent of this connection is to be able to resist similar stair configurations, up to a defined size.

The stair configuration was analyzed to determine the maximum force on the connection. Then, the capacity of the connection was determined with both elastic and inelastic analysis from manufacturer’s data for the capacity of the connecting expansion anchors. Finally, this connection capacity meets the required connection demand.

The connection was found to have less than half the capacity required to support the stair configuration. Additional testing may reduce this capacity further, as test results showed the configuration most thoroughly tested, which was thought to be the weakest configuration, was actually the strongest configuration.

DeLany, Heather “The Load-Deflection Behavior of Straight Deformed Bars in Concrete”

May 2009, 51pp, 7 references
Available for checkout
Abstract: This report covers the experiments and analysis conducted in the study of load-deflection behavior of straight deformed bars. The effects of clear cover, bonded length, and loading method on the stiffness of the load-deflection behavior were the primary focus.

Specimens that modeled the end of a beam in a knee joint were created and the reinforcement was loaded in tension, either monotonically or repetitively.

Of the twelve specimens tested, half of them had 2 d sub b of clear cover and half had 1 d sub b of clear cover. A third of the specimens had 16″ of bonded length, one third had 12″ of bonded length, and the other third had 8″ of bonded length.

Major findings include an increase in the load-deflection behavior stiffness by a factor of 1.5 for bonded lengths of 12″ and 16″ and a negative result for the hypothesis that repetitive loading would decrease the stiffness of the load-deflection behavior.

DeSimone, Richard “Effects of Bolt Spacing on Prying Action in Thin Flanged Specimens”

June 2012, 161 pp, 19 references, appendices, figures, tables
Available for checkout
Abstract: A large amount of research has been performed on prying action since the 1950’s which led to modern design procedures. However, many unknowns still exist. One of the unknown considerations is the effect that bolt spacing has on prying forces. This study focused on the effects of bolt spacing in a tee connection subjected to tensile force.The variable in the experiment was the bolt spacing, and the experiment determined how the prying forces change as the bolt spacing changed. A secondary consideration was to determine and/or validate the idea of the moment at the bolt line be greater than that of the moment at the WT shape’s web face, which is a variable in the current design procedure.

The American Institute of Steel Construction provides code provisions for which prying forces are determined. Based on those provisions, the bolt spacing for testing was determined.

In order to properly determine the effects of the prying force with regards to the bolt spacing, the bolt forces were determined using strain measurements from the bolts. Through data analysis, the prying force was determined and a comparison of five specimens was performed.

Results showed that the prying forces dropped considerably as the bolt spacing increased. The first test specimen had a bolt spacing of 5.25 inches and experienced prying forces larger than that of the AISC provisions. The cause of the large prying forces was due to alpha, a, being larger than 1.0. A significant drop in prying force occurred from the first specimen to the second specimen which had a bolt spacing of 7.00 inches. Specimen three had a bolt spacing of 8.75 inches and showed a slight decrease in prying force from specimen two. Specimens four and five which had bolt spacing of 10.50 inches and 14.00 inches, respectively, experienced approximately the same amount of prying force, which was less than that of specimen three.

It was concluded that the prying forces decrease as the bolt spacing increases, and the AISC provisions are conservative for specimens with bolt spacing greater than 7.00 inches, but unconservative for specimen one which had a bolt spacing of 5.25 inches.

Doocy, Steven A. “Externally Wrapped Carbon Fiber-Reinforced Concrete Beams: Analysis for Determining Feasible Use of External Carbon Fiber Reinforcement in New Beam Construction”

June 2012, 40 pp, 13 references, appendices, figures, tables
Available for checkout
Abstract: Carbon fiber has been used as reinforcement in concrete structures since the mid-1900’s. When used as an external reinforcement, carbon fiber has been used in construction; however, a majority of cases include confinement of existing columns and shear or flexural reinforcement to add or to regain capacity in cracked or failing sections. The purpose of the research conducted and presented in this paper is to investigate the use of carbon fiber as reinforcement for new construction, utilizing the workability of the fiber to produce preformed shapes for use as primary reinforcement as well as stay-in-place concrete formwork. The experiment consisted of testing simply supported flexural members with one of three reinforcing schemes: steel control, carbon fiber control, and carbon fiber and steel composite. The members created using these schemes were compared to current American Concrete Institute code design capacities to determine the feasibility of carbon fiber reinforcement for use in new construction. The results showed that externally bonded carbon fiber as formwork were able to support the pressures exerted by the wet concrete. They did not, however, produce the expected theoretical capacities predicted due to a slip that occurred during loading. This caused the concrete to fail and the fiber to buckle. The data did show that the strengths of carbon fiber and steel in combination were additive, and had the fiber reached its full capacity, the design equations would have been confirmed. Future experimentation should include application of strain gages at critical locations to better determine the stress levels in the respective materials and to determine if the carbon fiber could be used to span longer distances as stay-in-place forms under typical construction loading without excessive deflections.

Faytarouni, Mahmoud "Static Seismic Analysis Methods"

May 2015, 123pp
Available for checkout

Abstract: The aim of seismic design is to reduce human and economic losses by designing earthquake resistant buildings. Codes provide standard methods for determining the earthquake demand on a structure. There are static methods, which represent the effects of ground movements by lateral forces applied to the building. There are also dynamic methods, which predict the structure’s response to time varying ground acceleration. These methods traditionally are based on linear elastic analysis, but there are trends to better include the effects of inelastic behavior and other sources of nonlinearity. This project compares the outcomes of different procedures for determining seismic demand on common structural systems. Structures are modeled using a common commercial finite element software. In this research, the Equivalent Lateral Force (ELF), the Modal Response Spectrum (MRS) and the Nonlinear Static Pushover (NSP) analytical methods were all applied to a Grohmann Tower building model in different ways. The building used has various characteristics and with different lateral force resisting systems. The building performance has been evaluated for all these different methods, and the NSP analysis alone enabled the ability to carry out further adjustments to the design to achieve the desired performance.

Frazee, Glenn “Design and Analysis of Timber Lamella Segmental Arches”

May 2011, 222pp, 16 references, figures, tables, appendices, bibliography
Available for checkout
Abstract: A lamella roof offers a unique architectural feature in its interwoven network of timbers. As a roof system, the stiffness created by the interlocking members results in a curved roof that uses less material than a traditional rafter and purlin design. The goal of this paper is for the reader to be able to create a preliminary design of a lamella roof that will be strong enough to withstand the loads stipulated by the most current ASCE 7-10 Minimum Design Loads for Buildings and Other Structures. This design is facilitated by load tables developed by the author using the finite element method and connection tables in compliance with the National Design Specification for Wood Construction 2005 Edition using the Allowable Stress Design (ASD) procedure. In reality, the values used for this preliminary design will give a conservative design that could most likely be lightened with a more in-depth structural analysis. Testing on a steel lamella model shows inconclusive results when compared to those predicted by the load table program developed by the author and should be investigated further.

Friedman, Adam D. “Axial, Shear and Moment Interaction of WT Connections”

May 2009, 157pp
Archival copy only

Gerloff, James R. “Cold-Formed Steel Slip-Track Connection”

Jan. 2004, 240pp, 16 references
  • Appendix A: Specimen Material Tests
  • Appendix B: MSOE Slip-Track Test Results
  • Appendix C: Summary of MSOE Slip-Track Test Results
Abstract: The slip-track connection is one of the most commonly used connections when designing curtain wall systems. There is little guidance in the American Iron and Steel Institutes’ (AISI) North American Specification for the Design of Cold-Formed Steel Structural Members (2001). This paper presents the nominal capacities for a slip-track connection as well as the effective distribution width of the track. Several methods of analysis were reviewed along with an example of each method. A parametric study of the slip-track connection was conducted as well as finite element modeling. A total of 108 test specimens were tested in different combinations with stud widths of 1 5/8″ and 2 1/2″. The specimens had a stud spacing of 16″ and 24″, a gap between the web of the track and the top of the stud of 1/2″ and 1″, and track thickness varying from 14, 16, to 18 gage. Seventy-two test specimens were tested in combinations with varying stud widths, stud spacing, and track thickness. Thirty-six test specimens were tested with alternate fastener spacing. These tests yielded higher failure loads, and for the sake of being conservative, only the tests with the fasteners located at the stud location are used in determining the nominal capacity. Finite element analysis was completed following the tests. A stud spacing of 24 inches was analyzed, with a stud flange of 1 5/8″, a slip gap of 1/2″, and track thickness of 18, 16, and 14 gages. Proposed design procedures based on the results of this project are provided.

Goswick, Zachary “Reinforcement of Axially Loaded Steel Compression Members”

May 2010, 76pp, 17 references, appendices
Available for checkout
Abstract: The reinforcement of axially loaded steel compression members (columns) is examined. Column behavior before, during, and after reinforcement is reviewed followed by a review of existing periodical literature on the subject. None of the existing literature either specifically addresses the behavior of reinforced columns or covers the topic in its entirety. Instead, the literature that is available typically addresses only portions of the entire subject. By reviewing the major theoretical aspects of the subject, an understanding of column behavior during each phase of reinforcement is provided. This aids in connecting the individual portions of the subject described in the existing literature. Even with a good understanding of this subject, missing information and the complexity of the topic require conservative simplifications be used by practicing structural engineers. The presented case study shows how to apply these simplifications to current design practices.

Gough, Arlo Jay, Jr. “Headed Reinforcement Lap Splices In Concrete Masonry.”

Nov. 2003, 69pp, 8 references
Available for checkout
Abstract: Headed reinforcing bars have been used in concrete construction in the past. Previous research has been conducted on the use of headed reinforcing bars as longitudinal bars in concrete applications. This experiment researches the use of splicing headed reinforcing bars in masonry construction. It is felt that a shorter splice length may be used to get the same capacity, or even more capacity, than code minimum splice length requirements for deformed bars. Tests were performed on nine specimens made with 8-inch units and three specimens made with 12-inch units to determine the effects of splice length and clear cover of headed reinforcement in masonry. Several limit states for headed reinforcement were reviewed and a few of them were analyzed. The limit states analyzed included: development, side blowout, and compression strut failure. A review of both variables (splice length and clear cover) for each limit state was provided. The limited number of specimens did not allow for additional variables to be tested. This experiment showed that splicing headed reinforcement in masonry warrants further review.

Graziano, Matthew J. "Investigation into a Proposed Methodology for Slender Element Analysis for Hot Rolled Steel Shapes"

May 2013, 105pp
Available for checkout
Abstract: The focus of this project is to compare the results of a proposed methodology for slender element analysis to different design methods, such as those produced by the Canadian Institute of Steel Design (CISC), and the American Institute of Steel Construction (AISC). A full parametric study is featured, following a review of relevant literature. The results of this project provide a basis for an understanding of the proposed methodology and contribute to the discussions necessary when considering the proposal for adoption into the AISC Specification. The parametric analysis for this report was conducted using Microsoft Excel. For each of the slender members under consideration, spreadsheets were assembled to determine the capacities of each particular member according to the specifications of AISC, CISC, and the proposed methodology. Each slender member under consideration was then analyzed, according to respective specifications, over various lengths in order to determine overall trends and to discover any outliers. Additionally, plots of individual slender members of the four members considered were created by producing a normalized y-axis with respect to yield force analyzed over slenderness ratios maximized at 200. This was completed in hopes of discovering particular trends for each specification pertaining to the four shapes under consideration in this report. The parametric analysis in this report of slender WT and single angle members provided key observations for unstiffened slender elements. It was observed that the proposed methodology provides more capacity for the unstiffened elements under consideration with respect to both the AISC and CISC Specifications. This observation is noted as an overall trend for the unstiffened elements. Through the parametric analysis of both slender stiffened elements in this project, it was observed that the AISC Specifications provided more member capacity with respect to the proposed methodology. In terms of the CISC Specifications, overall trends with respect to the proposed methodology pertaining to the slender stiffened elements in this project were not observed. This report recommends that additional parametric analysis of both stiffened and unstiffened elements must be completed. It is important that future research considers all slender HSS members in an effort to observe particular trends related to the slender stiffened elements. Also, additional shapes need to be considered in a parametric analysis, such as double angles, channels, and variations of built-up members in order to observe any inconsistencies between the methodologies. Due to the nature of the parametric analysis in this report, overall trends for the particular members were exclusively observed. Therefore, future research may also consist of statistical analysis in an attempt to compare the proposed methodology in an effort to provide necessary modifications to produce favorable trends consistent for all steel members. This statistical analysis could require member testing and analysis versus only a parametric analysis of particular members against the design specifications as seen in this report.

Grochala, Charles “Testing of a Local Fly Ash in Mitigating Alkali Silica Reaction”

May 2010, 41pp, 10 references, bibliography, appendices, figures
Available for checkout
Abstract: Alkali-Silica reaction is a fairly well documented phenomenon, caused by the alkali in Portland cement reacting with siliceous aggregates to form an expansive gel. Over time, the gel absorbs water and expands. This may cause expansion significant enough to compromise the host concrete’s structural integrity. 

The purpose of this research is to examine the use of fly ash as a means of mitigating this reaction. This experiment follows a modified version of ASTM C 1260, the mortar bar method for testing potential for alkali-silica reaction, with minor variations to allow for testing of a local fly ash in mitigating alkali-silica reaction.

The resultant data indicate that the addition of the tested Class C fly ash will reduce the effects of alkali-silica reaction with a 20% or more replacement of Portland cement with fly ash. The author recommends that a version of ASTM C 1293, the concrete prism test, be used in order to better determine an expected field performance.

Gross, Jason L. “Analysis of Wood Shear Walls with Alternating Spaces of Ventilation and Blocking”

June 2004, 85pp, includes references
  • Appendix A: Calculated Values for Fasteners and Sheathing
  • Appendix B: FEA Models
Available for checkout
Abstract:Shear walls have been used for many years as a lateral-force-resisting element for wood-member, light-framing construction. Ventilation requirements for attic roof spaces have been in place for just as long, but seldom has the topic of integrating the ventilation space with the connection between load carrying diaphragms with load resisting shear walls been addressed in the literature. The one reference found is the APA guide Introduction to Lateral Design, which mentions a technique to install blocking every other framing spece to accommodate venting in the spaces left open. It continues to explain that decreasing the nail spacing by half in the areas that remain, which results in twice as many nails in these spaces, can perform the same amount of shear transfer as a connection which has solid continuous sheathing. However, the guide does not mention any evidence to support the theory. This report describes similar design examples which have reduction values assigned for void spaces, analyzes the internal forces involved in placing vent spaces in different areas of the wall, lists the procedure and data from laboratory testing of shear wall specimens with and without vent spaces of two different capacities, and examines the difference between predicted and laboratory test results. The laboratory tests indicate there is not a reduction in strength due to the placement of void spaces in shear transfer connections, which proves the APA theory correct. A shear wall acts as a system of many parts, and for reasons not fully understood, the internal forces seem to distribute around openings to increase the overall performance of the wall. This report gives recommendations on future tests to validate the increase in strength that was observed.

Hayes, Megan E. “Robustness of WT Steel Connections during Quasi-Dynamic Loading”

May 2016
Available for checkout
Full-text available online
Abstract: One of the most commonly used connection types in steel frame buildings is a shear connection. These connections are typically designed only for vertical shear, though previous studies have proven that shear connections have some capability to resist an interaction of shear, axial, and moment. With recent events such as the Murrah Federal Building and the World Trade Center disasters, there is a continuing need to study the robustness of connections subject to unanticipated loading scenarios.
The purpose of this research is to qualitatively and quantitatively measure the effects of the interaction of forces during a quasi-dynamic loading of a flexible WT connection. Results from this research show how a WT connection will perform during a sudden collapse of an interior support column in a steel framed building. Through the study, the robustness of the connection through both rotation and accrual of load was observed and recorded. Results of the test demonstrated the flexural resistance of the WT connection and the presence of catenary action.
To properly study and analyze the connection, physical testing of three-, four-, and five-bolt configurations of WT connections was performed. For each of the configurations, one test was loaded under a quasi-static loading rate, and three tests were loaded under quasi-dynamic loading rate. Data collected from the testing were used to calculate forces at the connection, including shear, axial, and moment.
Testing showed that as the number of bolts increased, the flexural capacity of the connection increased but the amount rotation before failure decreased. To better compare results of the two different loading types, a comparison of energy absorbed by the connection was performed. Testing showed that as the loading rate increased the net amount of rotation the connection can withstand decreases.

Helmin, Jessica “Crumb-Rubber as a Substitute Aggregate in Concrete”

May 2010, 54pp, 13 references, figures
Available for checkout
Abstract: Waste rubber tires are the most illegally disposed of solid waste in the country. Stockpiles of waste rubber tires are hazardous to the environment and to human health. Tire fires are the biggest concern because of the toxic fumes that are released from burning rubber tires and the difficulty of putting out tire fires. Using crumb rubber particles from recycled waste tires could benefit the environment and construction industry. This project analyzes the use of crumb rubber particles in concrete. Multiple mixes varying the size and amount by volume of crumb rubber particles were tested for compressive strength, flexural strength, and freeze-thaw effects. The experiment showed that compressive strength and flexural strength decrease as the amount of crumb rubber increases. Further, the modulus of elasticity also decreased with an increase of crumb rubber. The freeze-thaw cycles had no discernible effect on the concrete; these cylinders showed the same relationship of strength to crumb rubber content as the non-freeze-thaw cylinders. Because of the decrease in strength, crumb rubber concrete may be most appropriate for non-structural applications such as sidewalks, road barriers and sound barriers. Further investigation could be done on the bonding between the rubber particles and concrete as well as the impact strength of crumb rubber concrete.

Huber, Matthew L. “The Milwaukee School of Engineering’s Photovoltaic System Project: Final Structural Report”

Aug. 2007, 124pp, 17 references
  • Appendix A: CC Roof Structure Plan
  • Appendix B: AISC 3rd Ed. Table 5-17
  • Appendix C: Copyright Approval Letters
Available for checkout
Abstract: The Milwaukee School of Engineering has recently been awarded the funding to install a 29 kWH photovoltaic (PV) system, supporting the nation’s growing need for clean energy. For numerous reasons, MSOE has elected to install this system on the roof of the Student Campus Center (CC) building in downtown Milwaukee. Before proceeding with this project, a complete structural evaluation must be completed showing that the CC structure’s capacity is adequate under new design codes. The objective of this project is to provide MSOE with a final structural report for the CC building, including a design for a solar panel mounting frame. Final recommendations, based upon the evaluation’s findings, are necessary for MSOE to continue with their PV project. To complete this project, a firm understanding of the initial design intent is extremely important. After studying the CC structural plans, the new design codes, ASCE-7 and AISC’s 3rd edition LRFD, were applied to determine the capacity of the current structure. It should be noted that the design codes have evolved significantly since the initial design in the 1940’s. Using these design codes, a solar panel mounting frame was designed, increasing the roof demand even further. Through the structural evaluation, it was found that the CC structure had more than sufficient capacity to support current design codes. After designing the solar panel mounting frame, it was determined that the CC roof structure was able to support the frame as designed. However, the placement of these solar panel frames is restricted. MSOE may proceed with their PV project while following the strict recommendations of this report, and after receiving consent from a licensed Professional Engineer of the State of Wisconsin.

Huberty, Aaron M. "Numerical Modeling of Steel-framed Floors for Energy Harvesting Applications: Modeling Protocol, Verification, Parametric Study, and Optimization of a System"

September 2014, 94pp
Available for checkout

Abstract: Lightweight steel-framed floors have been known to be susceptible to vibrations. Pedestrian traffic on these floors is capable of exciting several vibration modes. Structural engineers typically design to minimize these vibrations. However, some of these modes may be targeted to harvest low-demand energy with the use of a Micro-Electrical Mechanical System (MEMS). Modal analysis is a useful tool for determining the resonant frequencies of a floor system. Recent advances in MEMS technology and vibration analysis, as well as increased demand for sustainable energy sources, have prompted the creation of intermediate scale energy harvesters with the ability to capture vibrations with low frequency content. In order to optimize the coupled system of steel-framed floor and harvester, an accurate numerical model must be created and evaluated. A numerical model of an existing experimental floor is presented. The existing floor system was selected for study because the modal frequencies, damping ratios, and mode shapes had already been determined, allowing for easy comparison to the numerical model. A parametric study was performed on the numerical model to optimize design of the coupled floor-harvester system. The study was done by varying the mass and stiffness of a harvester in the numerical model and measuring the resulting displacements, accelerations, and response frequencies. The process of the model’s creation, validation studies, and results are discussed. The numerical model was able to accurately portray the dynamic behavior of the experimental model as well as optimize basic parameters for the energy harvesters and show locations that would result in the highest performance from the harvesters. The model was able to replicate the dynamic behaviors of the floor accurately for the first 5 modes of vibration, with the largest margin of error being 25.6% (in the 3rd mode), and the smallest margin of error being 0.85% (in the 1st mode). Additionally, the model was able to show that the areas of the floor with the greatest acceleration occurred at the regions experiencing greatest displacement during the 3rd, 4th, and 5th modes. Finally, the model was able to show that during a 0 – 30 Hz slow sweep, the coupled harvester experienced accelerations greater than 0.1g during 31.74% of vibrations.

Jaeger, Jadon A. “Web Sidesway Buckling Of Crane Runway Beams”

Nov. 13, 2006, 67pp, 3 references
  • Appendix A: Web Sidesway Buckling Model Log File — Simply Supported, No Stiffeners, Two Loads
  • Appendix B: Web Sidesway Buckling Model Log File — Simply Supported, Stiffeners Included, Two Loads
Available for checkout
Abstract: Crane runway beams are a common fixture in warehouses and industrial facilities. Most crane trolleys apply two concentrated loads onto the runway beam. Few studies have been conducted investigating the web sidesway buckling capacity of beams subjected to two or more concentrated loads. The purpose of this report is to investigate the effects beam section, load spacing, transverse stiffeners, beam end fixity, and restraint type and location have on the web sidesway buckling capacity of crane runway beams. Finite element models employing shell elements were created in ANSYS to observe the deflected shape of the modeled beams and to determine the critical web sidesway buckling capacity. It was found that varying load spacing, the addition of transverse stiffeners, and the application of rotational restraints have only a marginal effect on a beam’s web sidesway buckling capacity. Fixed ended beams have a significantly higher capacity than simply supported beams. An increase in beam section size increases the web sidesway buckling capacity with the flange width having the most effect.

Kempfert, Michael T. “Lateral Load Capacity of Steel Joist and Joist Girder Seats.”

Nov. 2003, 92pp, 2 references, bibliography & appendices
Available for checkout
Abstract:Steel joist girders have been used for many years as an efficient system to resist gravity loads imposed on building roofs and floors. They may also be effectively used to resist lateral loads imposed on the building. In the past, extensive destructive testing has been completed to determine the allowable lateral loads that may be resisted by joist girders. Recently, the standard joist girder seat depth has been changed from 6-inch seats to either 5-inch for LH joists or 7 1/2-inch seats for joist girders. Therefore, the testing conducted in the past has become invalid. For this project, numerous destructive tests were performed to determine the lateral load capacities of the new joist girder seat depths. These capacities are then compared with the results obtained from finite element models of joist girder seats. It is found that the results from the finite element analysis closely match those found from the destructive testing. From the finite element analysis, the inflection point for the moment gradient of the joist girder seat is accurately found allowing the determination of the lateral load capacity. With the confidence gained in the accuracy of the finite element analysis, it is recommended that future destructive testing be limited, and that finite element analysis be used to determine lateral load capacities of joist girders with different configurations.

Knowles, John Alexander “First and Second Order Analysis of a Steel Gable Frame using Visual Basic Programming”

August 2010, 222pp, 12 references, bibliography, appendices, figures, tables
Available for checkout
Abstract: The purpose of this capstone design project is to develop a software program with Visual Basic (VB) programming language that analyzes the forces and deflections of a steel gable frame. The BV program finds the shear, moment, and axial forces in a four-member gable frame using either a first order static analysis or a second order single step Euler method. The results were compared to a MathCAD sheet and to the commercial software RSTAB, which uses the Timoshenko non-linear method for its second order analysis. The results for the first order analysis were identical between the VB program, MathCAD sheet, and RSTAB except for a 0.1% difference in moment. Approximating uniformly distributed loads with nodal point loads causes this difference. The second order results were similar for a practical building frame. The comparison between the VB program and RSTAB produced a 4.3% difference for global deflections, 7.8% difference for bending moments, and 6.1% for support reactions. For impractical building frames with large deflections, the results were also different (3.5% difference for global deflections, 3.7% difference for bending moments, 3.1% for support reactions, 9.8% for axial forces, and 3.3% for shear forces). The difference between the VB program and RSTAB can be attributed to the second order analysis method, Timoshenko (RSTAB) versus Euler method (VB). Also the VB program rotating loads on the deformed geometry causes result differences. The second order analysis results were identical between VB and MathCAD because they both used the Euler method.

Lasecki, Susan L. Caluwe, PE “The Capacity of Headed Stud Anchors Installed in Corner Applications”

Nov. 2007, 101pp, 19 references
Available for checkout
Abstract: Decades of research and testing have been performed to determine the capacity of headed stud anchors in tension and shear loading. Additional studies have been made on the capacity of these anchors in combined loading. All of these studies have investigated headed stud anchors that have been installed normal to the face of the loaded concrete surface. The purpose of the investigation was to determine the capacity of headed stud anchors cast into concrete at an edge where the anchor is at a forty-five degree angle. This type of installation is commonly utilized in many cast-in-place, precast and tilt-up applications. Equations which represent the observed capacities will be developed and proposed for use in future design conditions. Four types of tests were performed to determine the capacities of the headed stud anchors. All of the headed stud test specimens consisted of 1/2″ dia. x 4″ long headed studs welded to a 3″x3″x5/16″ steel angle at a forty-five degree angle. The first test type consisted of a single stud, the second type of test consisted of two studs spaced at 5″ c/c, and the third test type consisted of two studs spaced at 9″ c/c. The fourth test type was identical to the second test type except that confinement steel was added. The test specimens were each cast into an 8″ thick slab with a center recess. Two slabs were cast; each with eight test specimens, four on each side of the center recess. A 3/8″ thick steel plate was welded onto each of the steel angles. At the time of the test, a steel clevis was secured through the hole in this plate. A hydraulic ram was connected to a 3/4″ dia. rod which was screwed into the clevis. The hydraulic ram was supported by a steel frame. One of the columns of the steel frame was located within the trough and the other was located to the outside of the projected failure area of the concrete around the headed stud. The hydraulic ram was used to then apply a vertical force on the test specimens. The maximum breakout strength was recorded and the failure areas were measured and documented. The observed concrete breakout failure surfaces were similar to those documented in tension tests of headed studs near a free edge of concrete. The failure cones closely represented the 35 degree failure cone used in current design standards. The capacity of the single stud test was almost precisely one half of the test values of the two studs spaced at 9″ c/c. Studs spaced greater than twice the effective length of the stud have the capacity of a single stud multiplied by the total number of studs used. The ACI 318-05 and PCI Design Handbook – Sixth edition equations both provide an acceptable representation of the stud behavior, provided that the equations are applied properly. Proper application of these equations included correct determination of the effective embedment and the application of the eccentric load factor. New empirical equations were also derived from the actual test results which more closely represent the actual capacities of the connections. The use of these equations may be limited since the testing range was not broad enough to confirm their accuracy in conditions with longer or larger headed studs.

Lawson, David “Creep Characteristics of Post-Installed Epoxy-Based Anchors Subjected to Tensile Stresses”

May 2008, 79pp, 16 references
Available for checkout
Abstract: Industrial strength epoxies have become common use in construction for installing anchor rods into cured concrete. However, connections that rely upon the use of epoxy-based adhesives do not fare as well under tensile stresses as a cast-in-place or mechanical anchor connection. Such materials are prone to creep strains, which can eventually lead to failure in the connection if unchecked.
In this report, the creep characteristics of four commonly used adhesives for post-installed anchors were evaluated. The adhesives were used to install 3/8″ diameter threaded rods into a concrete beam. The rods were then placed under a constant tensile load using a system of levers. Displacements in the anchors were measured daily over a period of eight weeks. Based on the data acquired and the shear stresses placed on the anchors, it was found that even when loaded within the manufacturer’s approved allowable limits, the epoxies within the connections experienced creep. This raises concerns about using epoxy-based adhesives to design connections under a continuous tensile load.

Leonardelli, Anthony Jon “The Effect of Rigid Board Insulation on Screw Connections in Cold-Formed Steel”

August 2014, 131pp
Available for checkout
Abstract: Changes in model energy code requirements have placed a greater emphasis on creating a complete building envelope by including a layer of rigid board insulation between screw fastened steel plies, i.e. the building framing members (cold-formed steel stud) and the exterior sheathing (sheet metal panel). This layer of insulation creates a gap between the fastened plies affecting connection stiffness, strength, and failure mode. The current cold-formed steel provisions do not take into account the presence of such a gap. In this study, shear tests were conducted to explore the effect of rigid board insulation, steel ply thickness, and screw diameter on screw connection strength and behavior. Connection failures are categorized into system failures and fastener failures. Test results indicate that, regardless of failure mode, connections with rigid board insulation thicknesses of 1 in. or greater exhibit reduced capacity and experience large displacements prior to failure. A design procedure is proposed in which connection geometry predicts connection failure mode, and strength adjustment factors are proposed for system and fastener failures. In addition, a simple model is developed to predict both displacement and stiffness based on connection geometry.

Lesser, Jacklyn E. “Robustness of Shear Plate Connections under Quasi-Dynamic Loading”

May 2016
Available for checkout
Full-text available online
Abstract: Simple shear connections have been commonly used in steel-frame construction because they are quick to erect and relatively cost effective. The connections are used for shear resistance, but studies have shown that these connections are capable of sustaining measurable rotational demands and axial load. This enables the connection to help resist collapse in the event of a column failure.
The purpose of this research is to evaluate the robustness of single plate “shear tab” connections when subjected to quasi-dynamic loading scenarios. The research outlines the shear tab’s capacity, its ability to utilize catenary action as a source of secondary load transfer, and the connection’s innate ability to sustain rotational demands and axial forces in both statically and dynamically loaded connections.
Eleven full-scale tests were conducted, consisting of two wide flange beams connected to a wide flange column stub with single plate connections of three-, four-, and five-bolt configurations. Two of the 11 tests used galvanized bolts. The column stub was pulled vertically downward simulating the compromise of a central column in a building. Axial forces and moments in the connection were calculated from measured strains. Beam rotations were calculated from displacement measurements. Applied load was measured by means of a force transducer.
The shear tab connection resists the applied quasi-dynamic load by means of flexural resistance and catenary action, similar to what was seen in prior static tests. Systems that undergo quasi-dynamic loading achieve failure at a lesser rotation than do those subjected to static loading. An analysis of the work done showed that both statically and quasi-dynamically loaded systems dissipate similar amounts of energy.

Lieffring, Craig “Pottery Cull, a Cement Replacement”

June 2010, 47pp, 19 references, bibliography, tables
Available for checkout
Abstract: Kohler Co. manufactures vitreous sanitary ware at numerous facilities around the world. Kohler Co. only accepts the best product, resulting in less desirable pieces becoming waste product. This waste product, cull, accounts for nearly 3,200 metric tons per month. Cull is currently stockpiled in landfills to be used as fill or sub-base for infrastructure improvement projects. Often, the cull is not used and is eventually buried.The object of this paper is to determine if ground cull can be an effective substitute for cement. Cull from Kohler Company’s vitreous product line in Spartanburg, South Carolina was used in this experiment. The cull was ground to a size passing through a 325 mesh sieve. Once in powder form, the cull is referred to as pitcher. Mortar cubes were made using various mixtures in accordance with ASTM C109. The mixtures were comprised of cement, pitcher, sand, and water. The control sample did not contain any pitcher. In the other samples, a portion of the portland cement was replaced with pitcher. Specimens with five, ten, twenty, and thirty percent pitcher, by weight were tested. The results of the experiment show that this cull can be used as an effective replacement for cement while maintaining or increasing the compressive strength of a mortar mixture. However, further tests are needed to determine the feasibility and sustainability of using cull as a replacement for cement.

Loehrl, Tyler "Waste Glass and Class C Fly Ash as Material Replacements in Concrete"

May 2013, 43pp
Available for checkout
Abstract: Every year, millions of tons of waste glass are disposed of into landfills without being repurposed. It is common knowledge that waste glass is a recyclable material, so this is disconcerting. One way to repurpose waste glass without reusing it in the production of glass items is to mix concrete with glass as aggregate replacement. This project evaluated the performance of concrete with the addition of waste glass as a fine aggregate and Class C fly ash as a supplementary cementitious material. Test batches of concrete were mixed based on currently used concrete mixes from the Wisconsin Department of Transportation and Prairie Material concrete provider to test the performance of concrete regarding workability, expansion, and strength. Test data showed that good workability can be achieved with the proportions of waste glass and fly ash used. Additionally, expansion was proven to not be an issue of concern for failure. Lastly, the introduction of waste glass and fly ash did not adversely
affect the strength of concrete enough to eliminate it as a viable option for structural applications. The test batches of concrete performed well above the specified strength they were mixed for and the test batches mixed from the Wisconsin Department of Transportation’s design actually outperformed the control mix. More research could be done on the freeze-thaw effects of concrete with waste glass, water-cementitious material ratio, and using Class C fly ash versus Class F fly ash to control expansion of the concrete.

Markgraf, Ryan “Finite Element Analysis of a Rivnut Connection”

May 2011, 105pp, 5 references, figures, tables, bibliography, appendix
Available for checkout
Abstract: This research project focuses on the tensile capacity of Rivnut connections. A Rivnut is a type of fastener used in steel construction that requires “blind” connections, often associated with Hollow Structural Section (HSS) tubes. The research consists of finite element analysis, using ANSYS Workbench, and physical testing conducted at the Milwaukee School of Engineering.

For the finite element analysis, two types of finite element models were constructed. The first was used to explore the stresses created by the Rivnut installation process. The second was a simplified model used to explore the load versus displacement relationship of a Rivnut under tensile loading.

Destructive physical testing was conducted to compare results to the finite element analysis. The tests were simple tension tests that were designed to limit the tested variables to those related only to Rivnut tension failure. Loads were applied using a hydraulic piston and displacements were measured using a linear variable displacement transducer (LVDT).

The results of both the finite element analysis and the physical testing are analyzed separately and then compared to one another. The results of the finite element analysis closely matched those of the physical testing through the elastic range. It was also able to predict an initial yielding load that was near the load at which the physical specimens began to yield. The specimens were found to behave similarly in the plastic range. Each specimen reached its yield plateau, gathered post-elastic stiffness, and failed due to Rivnut flange shear. It was concluded that the finite element analysis is valid for modeling Rivnut tensile stiffness through the linear elastic range.

Recommendations for future research exploration of different Rivnut loading directions, finite element analysis of the plastic loading range and fracture analysis using ANSYS, effect of various plate thicknesses and hole sizes, and use of automated mechanical loading for further testing.

Marshall, Lucas B. “Distortional Buckling of Cold-Formed Steel Studs with Multiple Compression Flanges”

May 2012, 127pp, 23 references, appendices, figures, tables
Available for checkout
Abstract: Reflecting a lack of relevant literature, the 2007 edition of American Iron and Steel Institute North American Specification for the Design of Cold-Formed Steel Structural Members provides insufficient guidance on how to determine the moment strength of members subjected to weak axis bending. Specifically, the specification is silent on distortional buckling of such members. Recent literature on testing of shear walls, suggests that distortional buckling can be a critical limit state for cold-formed members in weak axis bending, so practical methods to predict it are needed.The objectives of this project are to investigate analytically the distortional buckling behavior of cold formed members subjected to bending stress about the weak axis, and to explore the application of the “reduced web thickness method” for finite strip analysis for members with standard slotted web holes bent about their weak axis. A finite element method eigenvalue buckling analysis is used to explore the distortional buckling limit state. The accuracy of the finite element model is established by comparing its results to work reported by other researchers, to the closed form equations from the specification, and to a finite strip analysis.

The results of the finite strip and finite element analyses are compared to the yield moment to show that distortional buckling is a valid limit state for some members in weak axis bending. Comparisons of finite element results for specimens with and without holes show that standard slotted holes do affect the distortional buckling strength of members in weak axis bending. The strength reduction is proportional to the ratio of the depth of the specimen to the height of the hole, and is significant for some such ratios. Results also show that the reduced web thickness method to account for the effect of holes in a finite strip analysis, verified in the literature for strong axis buckling, requires adjustment with a constant multiplier to provide better accuracy for specimens in weak axis bending.

Matarrese, Vincent "Finite Element Buckling Analysis Comparison with AISC Effective Length Procedures"

June 2015, 123pp
Available for checkout

Abstract: This thesis project is focused on the use of the effective length method as found in ANSI/AISC 360-10 Steel Manual as a reliable stability analysis method for practicing engineers. The objective is to compare various moment frame stability results with a finite element buckling analysis. Recommendations and findings are based on the comparison results for frames that involve load patterns, leaning columns, effects of inelasticity and other general stability topics.

To fully understand the specifics of results, a few baselines were developed with regard to the use of SAP2000, which was the software chosen for finite element analysis. Before any frames were analyzed, accuracy tests were carried out to choose model element totals and a relative joint restraint factor for pinned column supports was determined for use in the effective length method.

Once baselines were developed for percent error discussions, frames were chosen to bring certain stability features into focus. Various load patterns were applied to frames with a variety of first and second story loading ratios, leaning column frames were developed with two stability-to-leaning column ratios and inelastic stiffness reductions were applied to the first-story of a frame to test inelastic effects. To measure influences on stability and to understand how the effective length method takes into account these stability influences, percent differences were created with regard to the finite element analysis.

In almost all cases it was found that the effective length method, using the alignment charts, yielded results that were more conservative than the finite element analysis. Since the alignment charts found in ANSI/AISC 360-10 only account for the sidesway buckling mode, the largest percent difference results occur on the second-story columns when controlled by other modes.

The overall conclusion of the project is that the finite element buckling analysis is the best solution for stability analysis for practicing engineers. If finite element analysis methods are not easily accessed, then it is acceptable to use the effective length method for simple, sidesway governed frames. When frames become more complex due to load patterns or the presence of leaning columns, it is best to ensure a finite element analysis is conducted.

Meier, Austin “Effects of Bolt Spacing on Prying Action in Thick Flanged WT Shapes”

July 2012, 135pp, 15 references, appendices, figures, tables
Available for checkout
Abstract: Prying action is a phenomenon that occurs in bolted angle and WT shapes subjected to tension. Numerous studies have been conducted to learn how prying forces act in connections. One variable that needs additional study is the relationship between bolt spacing and prying force. The objective of this research is to determine the effects of bolt spacing on prying actions through experimentation and then comparing to current design provisions.Research on five unique specimen configurations were developed with varying bolt spacings of 1.5b, 2.0b, 2.5b, 3.0b, and 4.0b, where b is the distance from the edge of the stem to the center of the bolt holes. Such spacings were chosen to test two bolt spacings that are within the AISC upper limit to tributary length to compare to current design provisions, and then to extrapolate past the 2.0b limit to determine the relationship between bolt spacing and prying force. Each specimen was connected with two instrumented bolts for means of collecting strain within the bolts from being subjected to 100 kips of axial tensile load. From the recorded strains, bolt forces and prying forces present in the bolts were determined. In addition, deflection was measured at three key points: system (total) deflection, stem deflection and flange deflection.

Results show an inverse relationship between bolt spacing and prying forces. As spacing increased between the bolts, the prying force decreased, and although more deformation occurred in the flange along the orthogonal length of the specimen, it wasn’t significant enough to cause any further impact on the prying action forces in its respective direction. It was concluded that AISC provisions are conservative with respect to prying force estimation. It was also found was that the tributary length for each bolt tended to be considerably larger than what is allowed by AISC provisions. An effective thickness modification factor is recommended that allows for more efficient designs in prying action models.

Moorhead, William “Comparison of modeling in metal-plate-connected wood roof trusses”

Dec. 2008, 82pp, includes references, appendices, figures
Available for checkout
Abstract: Several editions of a national standard for metal-plate-connected wood truss design have been issued in the United States going back to 1960. However the current standard does not include language to precisely define how the heel joint region of a roof truss should be modeled in analysis to predict truss performance. The Canadian national standard for trusses does include provisions for the heel joint analog. This study compares results from trusses modeled under three different heel joint analogs, and includes the results from proprietary analysis and design software title. Certain truss types are reported to have much lower combined stresses in the first top chord panel adjacent to the heel joint when the Canadian provisions are used. Truss deflections are not appreciably affected by the various analogs compared. It is common that stresses in the first top chord panel adjacent to the heel joint are most influential in determining required chord grade, and the lower stress levels in the Canadian variations suggest that using the Canadian analog could provide designs with generally lower lumber grade requirements, and hence more economical trusses.

Moser, Timothy C. “Rotational Stiffness of Single-Bolt Sleeve Connections”

May 2010, 101 pp, 26 references, appendices
Available for checkout
Abstract: Novum Structures uses a single-bolt sleeve connection to connect hollow tubular members to spherical nodes at the joints of its Kugel Knoten space frame system. A sleeve connection is a shop-assembled connection consisting of four parts: a sleeve, essentially a modified hexagonal nut with an unthreaded hole and a slot or hole in two opposite faces; a bolt with a pin-hole; end-cones that are welded to the ends of the tubular members; and a small locking pin. The sleeve conceals the unthreaded portion of the bolt shaft left exposed by the end-cone, and the pin holds the connection together by preventing the bolt from sliding back inside the tube. Turning the sleeve tightens the bolt. Novum designs its space frames as space trusses, neglecting the moment capacity and rotational stiffness of the sleeve connections. There are instances, however, where bending moments are applied at the joints of the space frame due to eccentric loading. The least cumbersone and perhaps most economical solution is to take advantage of the strength and stiffness of the sleeve connections. In order to use the inherent strength and stiffness of the sleeve connections, the moment-rotation behavior of the sleeve connections must be understood. An analytical method already exists to calculate the moment capacity of a sleeve connection. No equivalent method exists to compute the stiffness of the connections. The purpose of this capstone project is to determine the rotational stiffness of three typical sleeve connection sizes, one small, one medium, and one large. A finite element method is employed to model and analyze the connections under zero axial force conditions. Results of the finite element analyses are presented and compared to actual experimental tests of the same connections. A secant stiffness model is suggested for determining the stiffness of the sleeve connections from the moment-rotation behavior predicted by the finite element analyses. The softest secant model accurately predicts the moment-rotation behavior of the medium and large connection sizes. A stiffer secant model is required to accurately predict the moment rotation behavior of the small connection size. Additionally, the computed stiffnesses are compared to the American Institute of Steel Construction’s stiffness limits for connection classification. The Kugel Knoten connection behaves either as a simple connection or a partially restrained connection depending on the flexural stiffness of the connected member. For practical tube lengths, the medium and large connections qualify as simple connections based on AISC’s criteria. The small connection qualifies as a partially restrained connection for practical tube lengths.

Nordling, Daniel W. “Interaction of bond and bearing for headed and hooked reinforcement”

Jan. 2009, 59pp, includes references, figures, tables
Available for checkout
Abstract: The interaction between bond and bearing for headed and hooked reinforcement in concrete is not well understood. Previous research has established the ultimate capacities of headed and hooked reinforcement in concrete. This project was conducted to better understand the interaction between bond and bearing between initial load application and ultimate capacity.

Tests were performed on a total of 16 specimens. Of these specimens, 5 were headed reinforcing bars, 5 were straight reinforcing bars, and 6 were hooked reinforcing bars. The reinforcing bars were ASTM A706, Grade 60, #7 bars. Concrete cover and bond length were the variables tested. Embedment depth remained constant for each specimen. Limit states of steel failure, breakout, pullout, side-face blowout, and splitting were analyzed to help determine the interaction between bond and bearing. Also, bonded length was used to analyze the bond strength over a partial development length.

Due to a limited number of test specimens, definite conclusions to the interaction between bond and bearing were not determined. However, the tests were used as verification for capacities found using ACI 318-02 code equations and proposed design equations from Thompson, Jirsa, and Breen for headed reinforcement. ACI 318-02 equations found a more conservative capacity for headed reinforcement than the proposed equations, but the tests indicated the proposed equations are more appropriate for capacity calculations. It was also found that the proposed equations cannot be applied to hooked reinforcement.

Patton, Jaime. “Effect of Splice Length of Headed Reinforcement Bars in Masonry Cells”

May 2004, 51pp, 4 references
Available for checkout
Abstract: The use of headed reinforcement is relatively new in masonry walls. Research is being conducted to determine if splice lengths can be reduced when heads are present. Previous research tested anchorage of headed bars in concrete and effects of splice length and clear cover in masonry. This project was conducted to further previous research by testing effects of headed reinforcement on splice length in masonry cells. Tests were performed with a total of 12 specimens. The specimens were constructed using ASTM C 90 dimensions of standard CMU block. There was a combination of 8-inch block cells and 12-inch block cells. Variables of splice length, development length, compressive strength and material confinement or clear cover were tested. Limit states of single blowout, compression strut or a combination of the two were developed and analyzed to determine the capacity of the specimens. Due to the limited number of test specimens, definite conclusions to the effect of headed reinforcement on lap splices in masonry cells could not be detrmined. Development length had no direct effect on the capacities, more cover led to higher capacity and longer splice length led to increased capacity. An equation was developed using previous research for in-plane and out-of-plane blowout failures. It was determined that the in-plane equation was the best fit for longer splice lengths. More research is needed to determine the effects of splice length on capacity for short splice lengths.

Peterson, Matthew "Experimental Testing of a Proprietary, Lightweight Concrete Panel System for Housing Applications in Haiti"

May 2014, 71pp
Available for checkout

Abstract: In 2007, Rod Ingram began developing concrete panels with a composite cross-section for housing construction in Haiti. The panels are constructed of cement mortar, expanded polystyrene concrete, and welded wire mesh. A mathematical model has been developed to determine the flexural capacity of the panels based on their various variables including cement mortar mix proportions, water to cement ratio, amount of flexural steel, and EPS concrete strength. Experimental testing was performed on these panels to validate the capacity estimates generated from the mathematical model. The main results generated from testing conclude that the concrete mortar strength does not significantly influence the overall capacity of the panels, however it does have a direct relationship with panel ductility. Upon completion of the experimental program, it is recommended that changes be made to the current design of the concrete panels to ensure adequate performance when used in construction in Haiti.

Pinkerton, Elise “Axial, Shear, and Moment Interactions of Single-Angle Connections”

September, 2014, 88pp
Available for checkout
Abstract: Simple shear connection such as shear tabs, single angles, double angles, and tee connections are commonly used in steel framed structures because they are relatively simple and economical in terms of fabrication and erection costs. These connections have typically only been considered to resist vertical shear loads. Their ability to resist the interaction of shear, axial, and moment forces has not been well studied and their behavior under combined loading is not well understood.
The purpose of this research is to investigate the behavior of single-angle connections that are subjected to the combination of shear, axial, and moment forces due to a simulated column failure. This thesis presents historical background on single-angle connections, current industry standards and provisions, previous research on the interaction of forces on shear connections, and a brief discussion on catenary action.
Six full scale tests simulating a column failure were conducted on single-angle connections for three-, four-, and five-bolt configurations. Experimentally measured strain and deflection data was used to calculate the shear, axial, and moment forces at the bolt line of the connection, as well as the end rotation of the beam. In addition to the experimental testing, a finite element model of the five-bolt configuration was created in an attempt to replicate the behavior of the connection.

Price, Andrew M. “Lateral Strength of Cold-Formed Steel Slip Track Connection”

Dec. 2006, 220pp, 4 references
  • Appendix A: Elastic Slope Data Plots
  • Appendix B: Original Data Plots
Available for checkout
Abstract: This project was based on the initial testing of the Cold Formed Steel Slip Track (CFS-ST) – to – Cold Formed Steel Stud connection and design procedure by James Gerloff. His analysis and conclusions allowed for this topic to be revisited by analyzing a new slip-track design. The variables used in determining the strength of the CFS-ST were design gap, e, thickness (gage) of material, and either the use of clips or no use of clips. All of these variables were standardized to analyze the effects that each variable had on the overall strength of the CFS-ST. Safety factors were also determined based on testing results and were compared with the original testing done by James Gerloff. The summary of results are determined for the nominal strength of the CFS-ST specimens.

Rescorla, Matthew R. “Partially restrained moment connections : classification of connections relative to stiffness, strength, and ductility”

July 2009, 85pp, tables, figures, appendices, 9 references
Available for checkout
Abstract: Integral to the design of all structures is connection design. Therefore, one must know the stiffness, strength, and ductility capacity, along with the moment-rotation characteristics of the connection that is being utilized. To that end, it was the author’s intent to investigate two different types of connections in hopes of obtaining results that could be used to classify the connections as simple, partially restrained, or fully restrained. In addition, the connections’ defining characteristics and the moment-rotation capacity of each connection were established.

The investigation was conducted using five test specimens. Three set-ups consisted of cold-formed c-channels bolted to vertical column tops and two set-ups consisted of wide flange members bolted to the face of a hollow structural section via a moment end-plate. Each specimen was placed and bolted into the testing apparatus such that the hydraulic ram could engage a beam cantilevered from the column approximately three feet. Loading was applied in a cyclical fashion; the load magnitude was recorded while linear differential variable transducer (LDTV) measured deflections near the joint.

It was established that both types of connection were within the “simple” range of two different connection classification schemes. The prevailing factors that contributed to the results are lack of enough bolts in conjunction with slightly oversized slots in the case of the c-channel tests and a very stiff beam compared to a relatively thin tube wall thickness for the moment end-plate connection.

The classification schemes chosen for this project are based on those found in the American Institute of Steel Construction (AISC) Steel Construction Manual and in publications by Bjorhovde et al., and Christopher and Bjorhovde.

Richard, Russell “Finite Element Analysis of Square HSS Section to Base Plate Connections in Tension”

September 2010, 126pp, 13 references, bibliography, tables, appendices
Available for checkout
Abstract: In the AISC Hollow Structural Sections Connections Manual base plates of HSS sections are discussed. Specific to this project, the equations for a base plate in axial tensile loading appear on page 7-9. As explained in the manual, “published solutions to determine the base plate thickness or weld design for the base plate-anchor rod connection […] do not exist.” It is the purpose of this Master’s Capstone Project to determine how correct the given equation set is and suggest alternatives to the equation if necessary. The method by which this is accomplished is through the theoretical testing of many variations of square HSS and base plate thickness combinations in the finite element analysis program ANSYS 12. These finite element specimens are put in uniform tension applied to the end of the HSS to simulate uplift on a column. The force is resisted by four bolts. The models are observed for stress concentrations and possible failure modes. The models are also observed to determine if there is correlation between the given equations that would be applicable for each component and the model results. The finite element analysis results are somewhat limited due to the limitations of the academic version of the program that was used for experimentation; however, individual results demonstrate that the equations do not accurately describe the behavior of the connection. Specifically, the weld equations do not take into account all of the variables that are required to accurately describe the behavior of the weld. The weld strength equation only takes into account the size of the weld. The thickness of the plate, the radius of the corner of the HSS section and the size of both the plate and the HSS section affect the strength of the weld. Also, the weld strain equation is unneeded. The plate equation set takes into account all of the proper variables; however, the equation is used to describe one quarter of the equation, when as it is written, it should describe the entire connection. Based on the results of the finite element analysis, the equations that describe this connection should be altered to reflect all of the variables that affect the connections. The weld equation especially needs to be changed to include the variables that affect the strength of that part of the connection.

Rinke, Brian J., P.E., S.E. “Reinforcing Steel Beams With Steel of Different Yield Stress for Flexure”

Aug. 2007, 116pp, 14 references, bibliography
Available for checkout
Abstract: Reinforcing steel members by welding on additional steel pieces has been a practice for as long as welding has been available. Structural engineers regularly investigate a variety of older buildings with steel framing that require reinforcing of the members. Information about combining the assumed lower yield steel of the original member with the higher yield steel of the new pieces seems to be difficult to find. Will the higher yield strength of the new steel offer any additional load carrying capacity to the system? To explore this condition, four steel beams were obtained and additional steel pieces were welded to them. These shapes were configured in a manner that is common to the industry for reinforcing steel that is currently in place in a building’s structure. Each steel beam was placed in the test frame one at a time, supported at each end with half-round supports, and a hydraulic ram in the center of the test frame provided the force. The applied force was spread to two equal point loads, each eight inches off the centerline. Although the steel of the original beam and the reinforcing pieces had different yield strengths, only the material that was furthest away from the neutral axis of the beam had any influence on the load capacity of the system. In all of these shapes this was the top flange of the original beam. For configurations like those used, it is recommended to use the yield stress of the original beam when designing reinforced shapes. It was also observed that the deflections of the beams were greater than the calculated values, possibly caused by a slow and steady shift of the neutral axis of the systems throughout the course of each experiment, even before the beams began to yield. In addition, all of the beams had about the same load limit before the rotational stability was exceeded, regardless of the thickness of the flange or the web of the beam or how it was reinforced.

Robinson, Jonathan. “Analysis and design of masonry spanning 2-ways between in-wall pilasters”

May 2006, 183pp
Available for checkout
Abstract: Some of the most common buildings are one-story buildings and one of the most common construction methods for these one-story buildings is the use of load bearing masonry walls. A system used for this type of construction is one with pilasters that work by having unreinforced masonry spanning between them. In this project the focus was a pilaster system that included pilasters that did not project from the wall. Another specialty of this project was that the unreinforced masonry between the pilasters was allowed to span in two-directions. This specific system set the stage for a computer application to be created. This report goes through both reinforced and unreinforced masonry designs that were used to design this type of system. Along with this is an explanation of how the computer application was created. One special topic in this project is how the masonry pilaster system was affected by in-wall pilasters when thinking about the distribution of load for the unreinforced masonry spanning in two-directions.

Rudolf, Jerod James, P.E. “Headed Bar Lamp Splices: Effects of Clear Cover and Splice Length on Splice Capacity”

July 2011, 121pp, 10 references, figures, tables
Available for checkout
Abstract: Research has been performed on headed bar anchorage over the last three decades, and provisions for headed bar anchorage development were included in the ACI 318-05 code. In the most recent decade, research began on the use of headed bar lap splices in reinforced concrete masonry; however, little-to-no research has been performed investigating the use of headed bar lap splices in concrete sections. For this reason, the following project was established to investigate the possibility of reducing lap splice lengths by using headed bars. To narrow the scope of the research, the bond length along the bars was eliminated in order to isolate the effect that only the bar heads have on the splice capacity.Twelve concrete specimens with #7 headed bar lap splices, representing splices with three clear cover values and three splice lengths, were cast and tested until no additional load could be sustained. Based on the test results, three theoretical limit states were investigated: side face blowout in two directions and the formation of a compression strut between the bar heads. The test results were compared to two equations from previous research for side face blowout, and theoretically derived equations, based on previous research and code recommendations, for all three limit states in an attempt to predict the splice capacities and determine the controlling limit states of the tested specimens.

Due to the small number of specimens tested, the main conclusions from this research are only relevant for use as a starting point for future research and may not match conclusions from more extensive testing. In general, lap splice capacity increases by increasing clear cover, and splice capacity doesn’t necessarily increase by increasing lap length. The only clearly visible limit state observed during testing was a face blowout failure for the two shallowest clear cover specimens and the two longest lap splice specimens. The derived equations and equations from previous research did not provide any solid evidence towards a clear relationship between clear cover and a controlling limit state, and splice length and a controlling limit state.

With respect to real-world design, that headed bar lap splices with clear covers of 1.50 inches or less cannot develop the yield strength of the bar, relying only on the bar heads. It was determined, however, that an 8-inch lap splice can fully develop the yield capacity of a #7 bar with clear cover greater than or equal to 3 inches. Based on the findings of this research, the use of headed bar lap splices can result in a significant reduction in steel compared to standard ACI 318-05 lap splices used in engineering practice today.

Schmitz, Robert P. “Fabric-Formed Concrete Panel Design”

Oct. 25, 2004, 80pp, 19 references
Available for checkout
Abstract: Concrete wall panels have traditionally been cast using a rigid formwork. Straight-forward methods of analysis and design are available for the traditionally cast concrete wall or floor panel. This is not so for the panel cast in a flexible fabric formwork. The purpose of this report is to develop a design procedure that allows one to design a fabric cast concrete panel. A four-step procedure for analytically modeling the fabric formwork is developed in order to determine the final shape of the concrete wall panel. The structural analysis program ADINA was employed to analyze the complexities of a flexible fabric formwork and the concrete panel cast in it. Analytical modeling and design techniques are developed in this report that will allow the design community another way to express themselves through the use of flexible fabric formwork.

Schultz, Joshua Adam “Design of Fully Tempered Monolithic Structural Glass with Point Supports Based on Ultimate Stresses and Stress Distributions”

May 2008, 79pp, 20 references, figures, tables
Available for checkout
Abstract: This paper investigates stresses in a typical application of fully tempered glass as a point supported structural balustrade. Loadings are defined by building codes, but literature on strength or other limit states for tempered glass in this application is limited. The paper compares finite element analysis, equations from plate theory, and test results. The goal of the paper is to establish an accurate and efficient method of design that uses a manual calculation to select a preliminary design thickness and employs the calibrated finite element method for the final design. The experimental results serve as a baseline against which the finite element models are compared in order to determine accuracy.

Experimental strain data is gathered from six fully tempered, monolithic glass balusters loaded to both service and ultimate conditions. Strain and deflections are measured and used to select the finite element model which most accurately idealizes the test results.The effect of element type, integration schemes and boundary conditions on finite element analysis results and computational processing time are compared, leading to the conclusion that for this application the model with solid linear elements and “pinned” boundary conditions provide the best results. Consideration of plate theory, including the effects of stress concentrations, results in an equation for field stresses and adjustment factors for stresses in glass and around holes. The resulting calculation provides maximum stresses based on plate thickness which compares favorably with the linear finite element results.

Experimental, finite element and manual results all show that the stresses at a 200 lb load are approximately 8,000 psi and about 24,000 psi at failure. These test results calibrate the finite element and plate theory predictions and suggest a design procedure.

Sell, Justin “Evaluating the Effect of an Alternative Epoxy Coating on Bond of Steel Reinforcing Bars”

June 2011, 62pp, 11 references, figures, tables
Available for checkout
Abstract: This report focuses on the effect an alternative polyurethane coating has on the bond performance compared with plain rebar. Twelve beam-end specimens were cast with six concrete specimens each containing one #8 plain reinforcing bar and six concrete specimens each containing one #8 coated reinforcing bar. Development length, cover depth, and presence of a coating were variables during testing. It was observed that coating thickness had the greatest effect on the overall strength capacity of the concrete specimens. The thick alternative polyurethane coating resulted in a strength decrease when compared to plain rebars. An increase in development length caused an increase in both strength and stiffness in coated and uncoated rebar. Concrete specimens with uncoated rebar saw an increase in capacity as cover increased but concrete specimens with coated bars saw a decrease in strength capacity.

Smith, Joshua R. “Selection of Steel Joists Subject to Snow Loading- a Computer Application”

May 2012, 103pp, 3 references, appendices, figures, tables
Available for checkout
Abstract: Steel joists have been a major building component for the last 60 years. Due to their relatively cheap materials and standardization of sizes and capacities, they are very economical structural component for roofs. It is up to the engineer of record of a building to determine the design loads that will be applied to the steel joists and determine a size that will be adequate for the specific building conditions, based on minimum SJI (Steel Joist Institute) requirements.These design loads have many factors that are use-dependent, location-dependent and importance of the building dependent. All these things have to be considered when calculating the minimum design loads that the steel joists will have to accommodate. One of the most complex and time consuming factors for joint design is snow load, and more specifically, drift snow. Where applicable, joists may have to be designed for trapezoidal loading because of drifting snow. This can be a very tedious task to do by hand.

This project focuses on joint design and creating a computer application to aid the engineer in the calculations. By using Visual Basic programming language, a simple, yet powerful software program was developed to help aid engineers in an important design task.

The result is a program that can calculate the minimum required loads on a steel joist per the latest building code and industry standards. The program is capable of designing a typical roof joist with standard loading on it, as well as joists that have one of two different types of drifting snow loads applied to it. The program selects a size of joist from the Steel Joist Institute’s database and displays the joist size, allowable load and other critical information that pertains to that specific joist.

In conclusion, this program is successful in performing the design of a simple, standard loaded joist, as well as being able to select a joist for two different drifting snow conditions. The program will help any engineer in the efficiency and accuracy of their calculations.

The program could easily be built upon to have additional features, such as the ability to analyze existing joists and suggest reinforcing for the joist if required. It could also be upgraded to output loading diagrams in a drawing file for the engineer to input into the construction documents.

Thompson, Scott L. “Axial, Shear and Moment Interaction of Single Plate “Shear Tab” Connections”

May 2009, 182pp, 23 references, figures, tables
Available for checkout
Abstract: The construction and engineering fields have long used simple shear connections such as single plates, single angles, and t-sections as supports for framing members in steel framed structures due to their relative simplicity and cost effectiveness of fabrication and erection. Historically these connections have been thought to support vertical shear loads exculsively; however, knowledge of their ability to support axial forces and moments has been speculated but seldom verified.

The purpose of this research is to provide observations and numerical verification of the single plate “Shear Tab” connection’s ability to support the combination of shear, axial, and moment forces as a result of a simulated column failure. This research presents a historical background of the single plate connection’s development along with providing insight into the connection’s ability to utilize catenary action as an inherent secondary load transfer mechanism.

Nine full scale tests simulating an interior column failure have been conducted for various depths of single plate connections. Shear, axial, and moment forces, as well as beam end rotation values have been derived from experimentally measured strain and deflection data to provide numerical evidence of the various observed connection rupture failures. A preliminary bolt force analysis technique has been developed to provide an understanding of the connection’s behavior prior to failure as well as to provide comparisons between the observed failure mechanisms and those expected using the current steel specification.

This research has shown the single plate connection has a low level ability to transform from a shear and flexural response to catenary tension. The experimental data suggest the shear tab connection alone could not support its intended design level shear load in the event of a catastrophic loss of a supporting column.

Vande Zande, Brian James “Structural Glass Fin Connections”

November 2011, 60pp, 23 references, tables, figures
Available for checkout
Abstract: For centuries, glass has existed as a component in exterior cladding systems because of its transparency. The use of glass in building facades increased as new technology allowed glass panels to be produced larger, and with greater precision. As technology continues to develop stronger glass, and more efficient ways of holding it up, architects are insisting on increasingly transparent structures. Recently, the trends have gone from structural steel supports for glass walls, to structural glass supports for glass walls.

Glass is extremely strong, but, unlike steel, its failure is governed by fracture, and it does not reach an inelastic range before failure. This means that special considerations, such as higher factors of safety, must be taken into account when using glass as a structural material. Currently, glass fins are the most popular use of glass as a structural element. In this situation, the fin acts as a wind strut for what is commonly a multi-story glass wall.

This paper will address the connections which either transmit forces to the glass fin, through the glass fin, or from the glass fin to a support. Two types of bolted connections will be analyzed, one in which forces are transmitted through bearing on the glass, and another where forces are transmitted through friction.

Whittinghill, Chad J. “Partial Fixity in Steel Single Plate Connections to HSS Tube Columns”

April 2009, 129pp, 10 references, appendices, tables, figures
Available for checkout
Abstract: When analyzing a structure, beam to column connections are classified as simple, fully restrained, or partially restrained. The type of end connection used in the analysis greatly affects the design of the beam and the column. Previous research has suggested that single shear plate connections, which are commonly designed as simple connections, may be stiffer than originally assumed. The scope of the previous research was limited to examine the accuracy of modeling single plate connections of beams to HSS tube columns as simple connections. The results of this research are used to demonstrate how modeling single plate connections as partially restrained, rather than simple, can lead to a more accurate analysis and possible cost savings.

Connections were made between a W24x76 beam and HSS tube columns of varying size and thickness. A horizontal force and beam deflection were measured simultaneously throughout a predetermined load cycle. Each tube column section was tested using three separate bolt configurations. From this data, a rotational stiffness was determined for each trial connection. The rotational stiffness values were then used in sample problems to compare true moment and deflection behavior with the assumed behavior.

The test connections all displayed moment-rotation behaviors that were well within the AISC recommendations for designing as simple shear connections. However, each connection was able to transfer moment to the column. While this moment transfer may be insignificant in many designs, understanding the true connection behavior may be valuable for beams with minor end moment demands.

Willis, David Charles, P.E. “Embedment of cast-in-place anchors in concrete : minimum embedment depth”

October 2009, 104pp, 13 references, appendices, tables, figures, nomenclature
Available for checkout
Abstract: Anchor bolts have been a major building component for a long time. However, design of anchor bolts has only recently been introduced into building codes. ACI 318-02, Appendix D, presents the first building code requirements (for non-nuclear facilities) for anchoring to concrete. Anchor design is controlled by the lowest value of the applicable failure modes. For an anchor in tension, these include the steel strength of the anchor, the concrete breakout strength, the pullout strength, and the concrete side-face blowout strength. Splitting failure of the concrete must also be considered in design.

When the anchors fail as a result of the concrete, a sudden brittle failure is experienced. It has long been considered good engineering practice to design for ductility where possible. A ductile failure associated with the anchor steel does not produce a sudden rupture. This report explores the failure modes associated with a single anchor in tension in an effort to produce a recommendation for a minimum embedment depth of an anchor in concrete to assure the steel failure modes govern. Eighteen anchors made from 3/4″ threaded rods were embedded in concrete as varying embedment depths. Some of the embedded anchors included a nut at the embedment depth, while the rest were embedded as plane threaded rod. A tensile load was applied to the anchors until failure.

The results of the anchor tests narrowed the range in which the embedment depth is sufficient to fully develop the steel strength of the anchor. Tests also showed that for short embedment lengths, the headed and non-headed threaded rods failed at similar loads. Although threaded rod appears to be closely related to deformed bar, the development and interaction with concrete proves to be much closer to plane steel bars. Analysis suggests that bond strength between the concrete and the anchor play a critical role in the ultimate strength of the anchor.

Establishing a minimum embedment depth for a steel anchor would allow for engineers to provide quick and economical anchor design with added confidence. A steel controlling failure mode design would incorporate the ductile elastic properties of steel into all anchor designs. With additional testing and analysis, an established relationship between the anchor type, anchor size, and concrete strength could be developed to establish a minimum embedment depth for anchors in concrete.

Wilsmann, Joshua “Weld Behavior in a Rectangular HSS Base Plate Connection with Corner Anchor Rods Subjected to an Axial Tensile Force”

July 2012, 81 pp, 10 references, appendices, figures, tables
Available for checkout
Abstract: The Hollow Structural Sections Connections Manual contains a proposed design procedure to determine the strength of an HSS base plate connection with corner anchor rods under an axial tension load based on base plate thickness and weld design strength, but no known supporting test results are provided to validate the procedure. Previous research by Christensen recommended adjustments to the equation based on base plate thickness, but found inconclusive results about weld behavior and recommended future research be done focused on weld strength.The purpose of this research project was to investigate the weld behavior in an attempt to validate provisions found in the Hollow Structural Sections Connections Manual relating to weld strength. To achieve this, physical testing was conducted on 12 test specimens, varying the base plate geometry and HSS size. Data were collected during the testing to measure strain along the length of the weld, base plate displacement, and column elongation.

Testing showed that a strain concentration develops at the HSS corner weld and distributes along the weld length in a parabolic pattern and that HSS wall thickness affects the strain development along the weld length. Test specimens did not develop the recommended four inch effective weld length used in the proposed equation based on weld strength. Future research is recommended to determine the effect of connection geometry on effective weld length. The proposed equation based on weld strain provided overly conservative results and the author recommends this not be considered when determining the connection capacity.

Witt, Zane Frederick "Modeling Cold-Formed Steel Screw Connections with an Insulation Layer Using Finite Element Analysis"

June 2015, 114pp
Available for checkout

Abstract: This study focused on creating a model using finite element analysis to accurately predict the maximum force, maximum displacement, and failure mode of a cold-formed steel screw connection with an insulation layer. These connections use cold-formed steel studs, a layer of rigid insulation and a metal sheathing panel with a screw connecting all three layers. The analysis includes geometric and material nonlinearities. To determine the accuracy of the model, the predicted maximum force, maximum displacement and failure mode were compared to previously completed tests of screw connection. The results show that the finite element model can predict the force and displacement for connections using thin studs and panels with some degree of accuracy. For connections using thicker studs and metal sheathing panels, the model predicted failure at very low force and displacement when compared to the test results.

Wurtz, Jacob “An Investigation of the Effects of Aggregate Replacement with Pottery Cull in Concrete”

May 2011, 58pp, 20 references, tables, figures, nomenclature
Available for checkout
Abstract: The Kohler Company manufactures a wide range of kitchen and bathroom fixtures around the world. The Kohler Company only produces products that are the best possible quality; all less desirable products become waste products. All waste products are collected in a Kohler-owned landfill at a rate of nearly 3,300 metric tons of pottery cull per month. Some portions of this waste are currently being used as a sub-base layer for infrastructure projects. If pottery cull waste is not used, it eventually is buried in the landfill.

The purpose of this paper is to determine if pottery cull can become a viable substitution for fine and coarse aggregate in concrete. All pottery cull used in this experiment was pre-crushed, by the Kohler Company, to fine and coarse aggregate sizes. Concrete cylinders were made using various mix designs in accordance to ASTM C192. Mix designs included the following materials: water, cement, class F fly ash, natural coarse limestone, sand, coarse pottery cull, and fine pottery cull. Along with these mix designs, two control mix designs were made that included no pottery cull aggregate substitutions. In all other mix design samples, a portion of coarse natural limestone and sand were replaced by coarse and fine pottery cull, respectively. Pottery cull replaced these natural aggregates at 15, 35, 50, and 100 percent levels using the volume replacement method described in ACI 211. All concrete cylinders were then tested using ASTM C39 to determine the compressive strengths.

The results of the experiment proved that pottery cull is an adequate aggregate replacer for maintaining the necessary compressive strength of concrete. It is recommended that further tests be analyzed for adverse freeze-thaw conditions and for alkali silica reactions. In addition, abrasive durability must also be assessed.

Zimdahl, Jamie L. “The Behavior of Headed Reinforcement in Concrete Members with an Applied Tensile Load”

May 2009, 67pp, 6 references
Available for checkout
Abstract: As the exploration for additional uses of headed deformed bars continues, it becomes more necessary to determine the behavior of reinforcement. Currently, little information is available on a method to predict the behavior or capacity of headed reinforcement. This project will focus on determining if it is possible to analyze the individual behavior of a straight bar and a bar head, and then combine them to find the relationship of a headed bar with bonded length. Twelve beam-end specimens, consisting of a pair of matching specimens for each of the six different variable combinations, were tested to find the effects of bonded length and clear cover. Headed bars with 0″ bonded length were tested, and the results were superimposed onto the results of straight bars of certain bonded lengths. The results of the superposition were compared to the test results of headed bars with matching bonded lengths. Test results showed that a percentage of the stiffness of the headed bar with 0″ bonded length is combined with a percentage of the stiffness for a straight bar of a certain bonded length to make up the stiffness of a headed bar with bonded length. After testing a variety of percentage combinations, the combination that accurately models the behavior of headed reinforcement could not be identified for certain. It was determined that the assumption that linear superposition of the stiffness of a headed bar of 0″ bonded length and a straight bar with bonded length is not accurate. A more accurate behavior could be found by documenting and graphing the independent behavior of the head and the bond length at each stage of loading. With the completion of further testing focused on the effects of clear cover and bonded length, the behavior of headed reinforcement could better be understood.
Contact the Library

(414) 277-7180


500 E. Kilbourn Ave.
Milwaukee, WI 53202

Library Facebook

Library Twitter

Ask a Librarian

Library Staff

Library Director:
Gary Shimek

Librarian I:
Sarah Rowell

Librarian II and Archivist:
Denise Gergetz

Senior Assistant Librarians:
Karen Bolton
Pam Gorzalski