Thermal measuring and testing – Flammability testing
Reexamination Certificate
2001-10-17
2003-03-25
Gutierrez, Diego (Department: 2859)
Thermal measuring and testing
Flammability testing
C374S031000, C436S147000, C252S607000
Reexamination Certificate
active
06536943
ABSTRACT:
BACKGROUND
Building codes in the United States place certain fire restrictions on cellular plastics used in construction. Sheathing products must at least meet the Class II requirements of the ASTM E-84 test: flame spread ≦75 and smoke ≦450. Class I E-84 criteria (≦25 flame spread and ≦450 smoke) are desirable, but not specifically called for in the building codes.
Roofing products, in addition to other tests, must meet the Factory Mutual Class I Approval as described in Factory Mutual Procedure 4450/4470. An important component of the Factory Mutual 4450/4470 procedure is the so-called FM Calorimeter test (hereinafter referred to as the context requires as the Calorimeter test, Calorimeter testing, or Calorimeter). In the Factory Mutual (FM) Calorimeter test, a 22.5 ft.
2
section of the particular roof construction is subjected to a heptane-fired flame for thirty minutes. According to the FM 4450 standard, the intended sample exposure is 1650 BTU/min./ft.
2
The heat release rate and total BTU content of the sample is determined and compared to the acceptance criteria. The FM Calorimeter test reports the peak heat release rates for the worst 3-minute, 5-minute, and 10-minute periods, as determined from the time/temperature curve. The average heat release rate, equal to the sample (total BTU/448), is also reported and used as a pass/fail criteria.
Approvals obtained under FM 4450/4470 are specific to the roof construction tested. The approval is limited to a minimum thickness of the insulation, and to the particular construction. For years, the ‘worst case’ construction has been the ‘4-ply glass’ built-up roofing (BUR) system. In this construction, the foam insulation is applied directly over a fluted steel roof deck. The steel decking in the test sample is not continuous. It is composed of two pieces, overlapped by about ¼″ and not secured with stitch screws. The top surface of the foam insulation is then mopped with Type 3 asphalt. Seven overlapped pieces of glass ply-sheet are applied with alternating moppings of asphalt to secure the ply sheets. A final asphalt mopping or ‘flood coat’ is then applied to the top surface of the roof deck. Starting in 2001, FM instituted routine weighing of the roof deck to determine the total amount of asphalt applied. The asphalt application rate is reported by FM to be included in any subsequent approvals from the particular sample. Asphalt application rates are critical because our results indicate that there is not enough BTU contribution available from a typical 1.5″ PIR foam board to fail the Calorimeter test.
Prior to this invention, the results of Calorimeter testing by others has been made available to see if such information would enable the development of superior test method. Such Calorimeter testing has included diagnostic tests, including the use of test decks using commercially-produced cover-boards and no asphalt. Also made available as background information were the results of extensive determinations of the total available BTU content of all roof deck components, using bomb calorimetry. Such data, summarized in Table 1, allows the researcher to calculate the approximate total BTU content of any Calorimeter test deck. The BTU content of each component is shown in BTU/lb. as well as the total pounds and BTU typically installed on the 22.5 ft.
2
test deck. The BTU content should be determined for each foam formulation, and for any alteration in type or supply of the deck components.
Also made available was information based on BTU determinations of the char remaining after the Calorimeter test, and calculations of the net BTU consumption in the Calorimeter. It had been found that typically the total BTU content of the deck, as reported by the Calorimeter, is 10-35% higher than the available BTU as determined by the bomb calorimeter, even when it is assumed that the entire 22.5 ft.
2
of the Calorimeter test deck is consumed. Actual measurements indicated that approximately 19 ft.
2
, or 85% of the total Calorimeter test deck area, participates in the test. This is in contrast to the 16 ft.
2
, or 71%, of the total test deck are assumed in the Calorimeter test procedure and calculations.
Results of, and conclusions from, investigations on the reproducibility of the Calorimeter test were also provided as background information. These investigations had been conducted with repeat tests of the same board in identical constructions. The data indicates that 1 standard deviation for the average BTU/ft.
2
min. for a typical ‘4-ply glass’ BUR construction is on the order of 50 BTU/ft.
2
min. The 2&sgr; (95%) confidence interval then is 285
+
/−100 BTU/ft.
2
min.
Recent measurements show that temperatures of 1500-1600° F. (820-870 C) can develop on the underside of the steel deck during the Calorimeter test. Similar temperatures have been recorded for research Calorimeter tests, using constructions that employed a cover board and no asphalt. The temperatures recorded as part of the Calorimeter test were measured in the exhaust flue by an array of 12 thermocouples and averaged. Measurements showed that the actual exposure temperature of the test deck can be up to 400° higher. Temperatures are lower in the exhaust flue of the Calorimeter due to mixing of stratified airflow.
Under the test deck exposure conditions, the ideal roof insulating material should provide a thick intumescent char, with minimal cracking, that would protect the asphalt in the BUR test sample. The insulation in the test sample is composed of four separate pieces that are butted together in the form of a square when viewed from above. For this reason, the ideal insulation also will not shrink or contract when exposed to the test conditions. Shrinkage will open up seams at the butt-joints and allow melted asphalt to flow through, eventually reaching the inside of the firebox and contributing to the observed total BTU. Laboratory screening tests focus on three key attributes of the foam: thickness retention, lateral shrinkage, and char integrity.
Several prior laboratory flammability methods are available to the researcher. The most common and easiest screening procedure is the so-called ‘hot-plate’ test. The roof insulation product is cut to 4″×4″ samples and placed on a pre-heated laboratory hotplate. Samples may be from faced production boards or from un-faced laboratory foams. A 900 gram weight maintains contact of the foam on the sample stage and prevents curling. The sample is heated on one side in the same manner as in the Calorimeter. Typically the hotplate is heated to 850° F. and the sample is exposed for 30 minutes. In fact, the temperature should be 870° C. (1598° F.) to simulate the actual exposure in the Calorimeter. Commercially available laboratory hotplates will generally not reach the same temperatures as the Calorimeter. Historically, a hotplate temperature of 800-850° F. has been used and correlated empirically to the Calorimeter.
Prior to October, 1999 this correlation served fairly well. Under the old Calorimeter correlation, results of 15% or less thickness loss in the hotplate are considered good and the material is a good candidate for Calorimeter testing. Results of 16-25% thickness loss in the hotplate are considered marginal candidates for Calorimeter testing. Those boards with >25% thickness loss in the hotplate are generally not considered for Calorimeter testing.
Statistical treatment of hotplate data showed that the 95% confidence limit is fairly broad. For this reason, hotplate data has traditionally been used only as a screening method. In spite of the large error bars associated with hotplate results, trends became obvious when large data sets of similar density foams were compared for different flame retardant combinations.
Weight loss and dimensional change have generally not been used as criteria for hotplate screening of materials for Calorimeter testing. Wide variation in thickness change is seen with samples that have substantially the same weight loss. Dimensi
Albemarle Corporation
Gutierrez Diego
Spielman, Jr. Edgar E.
Verbitsky Gail
LandOfFree
Method and apparatus for testing flammability properties of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for testing flammability properties of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for testing flammability properties of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3015576