Examination of the variability of the ASTM E 648 standard with respect to carpets

Abstract

EXECUTIVE SUMMARY PURPOSE AND SCOPEThe objectives of this research program were to investigate the variability of the ASTM E 648 Standard with respect to carpets and to make recommendations for revising the standard to improve the repeatability and reproducibility of the measured results. The need for such a study arose from NVLAP's concerns about the high coefficients of variation resulting from proficiency rounds carried out as part of the Carpet LAP.One carpet was studied to investigate the ignitability and flame propagation problem experienced by CRI in the summer of 1987.Several ignition sources and procedural variations were investigated to assure that flame propagation would occur; the most significant finding was that a propane line burner gave a more uniform burn pattern than the propane torch point source pilot burner and seemed to insure propagation away from the ignition area provided the critical radiant flux was less from the maximum flux from the panel.Eight carpets of varying fibers and constructions were examined using the standard procedure to determine which would result in the lowest coefficient of variation.One was chosen for a parametric study of the test variables. Based on the findings of this study, a "proficiency round" was carried out using this carpet and recommendations were made to ASTM Committee E05 on Fire Standards for revision of ASTM E 648. CONCLUSIONS AND RECOMMENDATIONSThe following conclusions resulted from this study:• Replacement of the point ignition source by a line burner improves the likelihood of flame propagation and provides more even burn patterns.• The rate of air flow through the apparatus has a significant effect on test results.• Contrary to previous experience with NVLAP proficiency rounds, good reproducibility can be achieved by paying closer attention to prescribed procedures.• Conditioning time up to at least 12 days appears to increase critical radiant flux values .• Imposing a time limit on the test is not a viable means for determining critical radiant flux of a carpet. iii • Use of 100 lb, or smaller, bottles of commercial propane can influence operating conditions during the course of the day if proper attention is not paid to making adjustments in the gas flow rate to maintain the panel operating temperature and flux profile.• The radiant panel requires 1 1/2 hours to approach equilibrium when propane is used for the fuel; equilibrium was achieved in only 1 hour with natural gas.Specific recommendations for further work which need to be addressed are as follows ;• Carry out a round robin with a wide selection of carpets, including as many laboratories as are willing to participate.• All future NVLAP proficiency rounds should be performed using the same adhesive supplied from a single source.• Determine the effect of specimen location on the carpet to assess variation in the product and determine the effect of carpet aging on critical radiant flux.Additional consideration should be given to the following in order to further the development of the Floor...

An evaluation of precision for the ASTM E648-91A standard test method for critical radiant flux of floor-covering systems

“…These changes included using a new line pilot burner. Results from this study showed an interlaboratory coefficient of variation of 11.5 percent, and some variations in CRF were found to be associated with conditioning time after glue down [1]. Of prime importance to this interlaboratory study was accuracy and simplicity.…”

“…The CV from this special NVLAP proficiency round for 10 of the 11 laboratories participating was 6.4 percent. In this round, one of the eleven laboratories was considered to be an outlier [1].…”

“…These directions allowed for the products to be glued to the test substrate in a fashion that would normally be used by a flooring contractor and followed by the testing laboratories. The limit on days of conditioning after glue down resulted from the finding in the 1989 study [1] that variations in conditioning time after glue down can alter the test results. The four to ten day window allowed a reasonable period of time for the laboratories to test their specimens without causing significant variations resulting from adhesive curing.…”

“…This research project is an extension of a previous study performed by the National Institute of Standards and Technology (NIST) and the Carpet and Rug Institute (CRI). Results from the earlier study are reported in NISTIR 89-4191, "Examination of the Variability of the ASTM E 648 Standard with Respect to Carpets" [1].^The current interlaboratory study (ILS) involved NIST, CRI and the American Society for Testing and Materials (ASTM). Results from this study will be used to improve the data base associated with the revised ASTM E 648-91 standard [2] and to prepare a precision and bias statement for the standard.…”

“…With some of the specimens, there was no flame propagation (NFP) away from the ignition point, and much of the data was not usable. Statistics for the carpet specimens which did propagate flames showed a coefficient of variation of 23.8 percent [1].…”