February 2000 Buro Happold Report - Fire Engineering of Steelwork
|2000-02-11||Buro Happold||World Trade Center Fire Engineering of Steelwork||Phase 1 Report||||33|
OCR Pages of BH Report.pdf:
OCR Software Used: http://code.google.com/p/tesseract-ocr/
Recognition that the member sizes used in the World Trade Center are heavier than the minimum UL listing. This means the steel will heat slower because it will be at a cooler temperature because of the larger W/D. This knowledge can be used to reduce thickness of the coating further.
Recognition that the structure has a greater ability to support loads during the fire load case. This has been effectively illustrated by the Cardington fire tests.
Buro Happold has been commissioned by the Port Authority of New York and New Jersey to conduct a fire engineering assessment of the fireproofing requirements of the open-web, steel, joists that support the floors in the tenant areas of Towers 1 and 2 of the World Trade Center. This Phase 1 report describes the work conducted to date and suggests how substantial cost savings can be made and increased value can be achieved.
There is a system currently in place at the World Trade Center for repairing and/or upgrading the fireproofing as tenants move out. When a full floor is vacated, the fireproofing is upgraded from 3/4 inch to 1 1/2 inches if this has not already been done. At full floors that have already been upgraded and at partially vacated floors, the condition of the fireproofing is assessed to determine whether it is more cost—effective to patch the fireproofing to match the existing thickness (1 1/2 inches or 3/4 inch) or to remove it and replace it. Where the fireproofing is removed, 1 1/2 inches are used for the replacement. This is summarised in Figure 1. Where existing fireproofing contains asbestos, the floors are abated. We understand that most of the
abatement has already taken place. The asbestos abatement process is outside the scope of this report.
Damage of existing fireproofing
Significant amounts of damage to the existing fireproofing occur during demolition after tenants move out. The product used in the past can be easily dislodged as ductwork, partitions, hangers, etc. are removed. Additional damage that happened during tenant fit-out or later modifications that may not have been repaired at the time. As we understand it, in the majority of cases, the existing fireproofing requires so much patching that it is more cost-effective to replace it.
It the damaged fireproofing is to be patched rather than replaced, the cementltious Monokote product is generally used regardless of what the in situ fireproofing material is. This can result in joists that are fire proofed by a combination of materials. This patching is generally done by hand rather than spray application. Repairs are made such that a constant thickness of fireproofing is provided to all joist members.
The Cafco Blaze—Shield flreproofing is readily removed using a high-powered water jet. The water from the jet is soaked up by the fireproofing which falls to the floor in a damp state. Usually all the water is absorbed by the fire proofing, however, on occasion water does leak onto the floor below. The floor is covered with protective sheets. The fireproofing is collected from these sheets and transported in waste buckets to a waste disposal area.
Any portions of fireproofing that are not removed by the jet are scraped away by hand. This often happens if there are sections of Monokote fireproofing and in the troughs that are formed at the bottom flange of the joist by the back-to·back angles. lf the Monokote cannot be removed by hand, it is assumed to be fixed in place and [ is covered by the new fireproofing applications.
When the fireproofing needs replacing, new fireproofing is applied to a thickness of ll/2 inches. While equivalent products are permitted, Cafco Blaze-Shield mineral fiber spray is generally used as the replacement fireproofing. it is estimated that 50-70% ofthe material is lost to overspray. lt can take 2-3 passes to apply 11/2 inches. if it is done in fewer passes, the fireproofing tends to fail the adhesion tests that are conducted after application. Sometimes ten feet tall off at once when tested.
2 Standard Fire Tests as Applied to WTC
All current building codes recommend that the structure of a building has an appropriate fire resistance period. In the case of the World Trade Center this is deemed as two hours. One method of proving the fire performance of a product is to test it in accordance with the "Standard Methods of Fire Tests of Building Materials} ASTM Designation E 1192 or equivalent.
2.1 ASTM E119
The ASTM standard fire test was developed primarily for establishing a method for comparing the relative performance of different construction assemblies when exposed to a controlled laboratory fire. The results of tests conducted in accordance with this standard do not necessarily indicate how these assemblies will perform under actual fire conditions. However, contemporary building codes reference the ASTM test and it is considered that in most cases the test gives conservative results.
The standard time-temperature curve
The Standard Time-Temperature curve was introduced into the ASTM testing procedure in 1918. This curve dictates the time-temperature regime to which all structural members are exposed to in the furnace. lt replaced the original specification of a constant 1700°F for the duration of the test. The standard fire curve was developed such that it would be readily reproducible in a laboratory and represented the committee's best judgement as to what constituted a realistically severe fire exposure. The applicability of the standard curve is discussed in more detail in section 6.
The standard fire test
The test requires a sample construction assembly to be exposed to a controiled laboratory fire defined by the standard time-temperature curve. The standard controls the dimensions of the test furnace and sample. The test specimen is to be either restrained or unrestrained. Restrained specimens are prevented from expanding due to thermal expansion and unrestrained tests are free to expand. The fire resistance of an assembly is the time, after the beginning of the test, when any of several endpoint criteria are exceeded. For both types of test, the structure must be able to support the applied load and the integrity of the floor system must be maintained for the duration of the test.
The endpoint criteria for beams, floors and columns all different. The table below shows all the endpoint conditions that are appropriate for the World Trade Center.
The maximum permissible design stress in steel is limited to approximately 60 percent of the yield strength. There has been extensive research into the properties of steel at elevated temperatures and it can be shown that steel still retains 60 percent of its yield strength up to a temperature of approximately 1000°F (538°C). Therefore, for structural members at 1000°F designed to carry the maximum permissible stress at ambient temperatures, the applied stress is approximately the same as the reduced yield strength of the member.
The structural performance of steel in fire conditions is related to the temperature of the steel. As the steel increases in temperature, it loses strength. Adequate fire performance is achieved by ensuring that the steel is sufficiently cool such that it maintains enough strength to support the required load. Since the load carrying capabilities of steel are related to its temperature, the less load that is applied, the less the fireproofing requirements. ln Allowable Stress Design the allowable stress is limited to approximately 60% of the yield or 9 buckling stress. This means that for the full design load at ambient temperatures, the steel is stressed to only about 60% of its capacity, giving a factor of safety of about 1.7. The fire condition is an extreme or accidental load case and as such it is recognised that it is appropriate to use a lower factor of safety when considering elevated fire temperatures. This reduced factor of safety is implicitly incorporated into the ASTM E119 test. It is assumed the steel member E, will be loaded to its code-allowed limiting stress, i.e. about 60% of the ultimate capacity. The limiting "F temperature of 1000°F corresponds to the point where the steel strength is approximately 60% of its ambient strength.
There has been substantial research conducted internationally that suggests that the fire performance of steel buildings is much greater than previously assumed. Broadgate @ in 1990 a serious fire occurred in the Broadgate Phase 8 Building in London. The building was still under construction and the fireprooting had not yet been applied to the steel frame. The fire spread over significant area of the floors yet the structure remained stable and repairs were made at a small cost. Traditional approaches and philosophies suggested that the unprotected structure, exposed to the fire conditions that it was, should have failed. This alerted the industry to the fact that composite steel-framed buildings have a greater inherent fire resistance than had been assumed previously. Therefore, six full-scale tests, as opposed to fire tests on individual members, were conducted on a purpose-built structure at Cardington in order to investigate behavior of real structures. BRE Cardington The test building was an eight-storey composite steel-framed structure built in the Building Research & Establishments (BRE) test facilities at Cardington. The tests were conducted by the BRE, the Steel Construction Institute (SCI) and British Steel. The frame was designed as a typical office building and contains no special features which might favorably affect its response during the fire tests. The following paragraphs are.quoted from an article by the Building Research Establishment published in the "Steel Construction" Journal?
• Previous tests have shown that the performance of steel frames in tire is signiiflcantiy better than that of the individual members which go to malre up the frame.
• The tests at Cardington Laboratory have demonstrated that modem composite floor systems possess a significant degree of inherent fire resistance which is not taken onto account in current design methods.
An important fact to consider with the Cardington tests was that in no situation was any protection provided to the beams. Atmosphere temperatures of up to 2200°F and steel temperatures of up to 2000°F were experienced in the tests. lt had been previously assumed that steel beams would fail at a temperature of approximately 1000°F. At 2-hours the furnace temperature in a standard furnace test is approximately 1850°F, 350°F lower than the atmosphere temperatures that were experienced at some of the Cardington tests. Performance based codes in the United Kingdom recognize the fact that stresses in steel beams are often lower than assumed in Standard Fire Tests, but they do not yet allow the engineer to fully account for the effects that were demonstrated at Cardington. However, there are approaches available that are capable of using some of the findings from Cardington.
Vulcan, a finite element analysis program that predicts the structural performance of composite steel—framed buildings in fire conditions has been developed at the University of Sheffield, England. The program, which was developed in conjunction with British Steel, the Steel Construction Institute and the Building Research Establishment, has been used to model the fire tests conducted at Cardington. The program has been subjected to extensive validation and many papers regarding its performance have been published in peer reviewed journals and presented at major, international fire conferences. By using Vulcan, FEDHA are able to account for many of the three dimensional effects that occur in the fire condition in composite steel—buiIdings. The computer analysis can be used to determine deflections, member forces and concrete cracking behavior. This allows the engineer to assess the safety of individual buildings and to calculate the exact amount of fire proofing required for actual buildings. This is often less than required by prescriptive design guides but not in all cases. Therefore the main advantages of this approach are that thea more cost effective structural fireproofing package can be provided and greater confidence in the final design is attained.
Vulcan offers the following advantages over other finite element packages:
• Non—Iinear material properties at elevated temperatures.
• Can assess large deflection and therefore can determine any load sharing that occurs.
• Can account for the benefits of the concrete slab and reinforcement that are not valid at ambient temperatures.
As a result of the post-analysis of the Cardington fire tests, new guidelines are being written in the United Kingdom. ln their current format, these codes recommend that for buildings with a 30—minute fire resistance period, fireproofing can be omitted from all beams, and for EO-minute buildings, fireproofing can be omitted from all secondary beams. All the analyses conducted in the development of these guidelines were done using Vulcan. This reflects the confidence in the use of Vulcan shown by the code writers. It also highlights the scope of Vulcan.
Port Authority of NY/NJ: Records For Reported WTC Renovation Work Destroyed On 9/11: http://www.911blogger.com/node/19889
Another amazing coincidence related to the WTC: http://911blogger.com/node/13272