Testing Flameproof Fabrics
A Study of the Field and Methods of Evaluating Material Treated With Compounds
ERNEST F. HARTMAN
THE necessity for a workable, easily understood and inexpensive laboratory method to test the efficacy of various fire-retarding preparations designed for use on or in fabrics has been long felt. The statistics of insurance companies show a loss of life, limb and property directly attributable to fires started by easily ignited fabrics in the form of curtains, drapes, rugs, tapestries and even clothing. The U. S. Dept, of Agriculture took a forward step in the field by issuing a bulletin for housewives informing them of a simple method for fireretarding fabrics. Commercial houses interested in selling patented mixtures have aided. Building codes and safety councils need requirements which would result in fewer fatalities in their communities. These resulted in a specification, drawn up by the U. S. Bureau of Standards, which served as a basis for the tests under discussion.
Tests at New York World’s Fair
The scope of the study of flameproof fabric tests was limited to the establishment of standards which fabrics should meet to provide reasonable protection against flame spread and panic hazards and to the evaluation of simple tests for determining the ability of fabrics to meet these standards. This study was undertaken at the request of the officials of the New York World’s Fair, 1939, Inc., when they found no existing standards that would fit into an interpretation of their slogan, “The World of Tomorrow.” The tests were especially desirable at the time, because of the presence in several World’s Fair buildings of materials, which, in their natural condition, are inflammable. For instance, the Cycloratna inside the Perisphere contains quantities of cardboard; many buildings have decorative draperies and bangings; and in a number of instances thatched roofs are employed for scenic effects. Test methods vary with the different materials. This article is limited to a discussion of flameproof tests of fabrics for interior use only.
Tests were made on fabrics treated with three different preparations, as follows :
- Treated fabrics, finished and ready for test, were obtained from one company representative of those now marketing “flameproof” fabrics. The methods of treating these fabrics are patented or secret, and are usually combined with the process of dyeing or finishing the material.
- One of the several chemicals prepared by manufacturers for the specific purpose of flameproofing was obtained and the fabric treated in accordance with the manufacturers’ directions. One pound of this salt was dissolved in one gallon of warm water and the fabrics immersed until thoroughly saturated, then wrung hand-dry, and dried at room temperature. The finishing of these fabrics consisted of ironing them when nearly dry with a moderately hot iron. If the fabric was too damp, or the iron too hot, the mixture deposited on the iron, necessitating wiping with a clean, damp cloth.
- A solution of 70 per cent borax and 30 per cent boric acid was prepared, and the fabrics treated with it. They were immersed until thoroughly saturated, then dried and ironed in the same manner as those treated under No. 2. This treating solution was considered one of the best in a long series of tests conducted by Ramsbottom and Snoad for the British Fabrics Coordinating Committee and reported to this body in a work which is considered the classic on flameproofing fabrics.
Scope of Investigation
In determining the flameproof qualities of the fabric under test the following reactions must be observed: Flashing flame during application of the igniting agent; flame after removal of the igniting agent; char; glow; and the permanence or lasting qualities of the treatment. By flashing flame is meant a flash of flame carried up the length of the test specimens by the fabric itself, and not the rapid burning of a fuzzy edge, which occasionally occurs when the specimen is not cleanly cut out of the sample.
If, when a fabric is ignited, there is an explosive flash of flame, the fabric must be considered as failing in its prescribed purpose, for the reason that such flashing may cause a panic. This is true, even though the flashing might be only momentary, though no flame followed the original flash, and though the fabric itself might be considered flameproof in the broad sense of the word.
Some fabrics will flame within the area covered by the igniting flame, but may be considered satisfactory, provided the flame does not continue for an appreciable time after the removal of the igniting flame.
Many treated fabrics will not flame, either during the time of exposure to the igniting flame, or after its removal and thus satisfactorily serve their purpose.
Charring of the material during exposure to the igniting flame is not important from the preventative point of view, since the material has served its purpose, if it has not contributed to the spread of flame.
Glow or Incandescence Important
The question of glow or incandescence is of more importance. At a meeting at the Testing Laboratories of Columbia University, which included representatives of the National and New York Boards of Fire Underwriters, the World’s Fair, the New York Fire and Building Departments, members of the Testing Laboratories, and other interested parties, it was the consensus of opinion that glow in the charred area would not ignite any combustible. A flame which would cause a flameproof fabric to glow would have ignited any proximate combustible material. Also, the glow of several samples, while igniting matches, was not of sufficiently high temperature to ignite paper or untreated linen damask.
It may be recalled that “flameless combustion” or glow, which is invisible to the eye. has been shown to be a factor in tests on heavier combustible materials. In the case of fabrics, this factor appears to be inconsiderable. by virtue of the fact that the porous nature of fabrics in general will not allow any invisible glow. Any tendency of the fabric to glow is made known immediately, inasmuch as the glow is visible to the eye. If this visibility is doubtful, as might be the case in a fabric colored similar to the red-orange of glowing charcoal, by simply feeling the specimen, its heat will verify the observation. It appears, therefore, that if the glow was not progressive, that is, did not continue outside the charred area, it could be disregarded, despite its duration.
Testing Durability of Treatment
The Building and Fire Departments of New York City make regular inspection of treated fabrics only once a year. In the case of the World’s Fair buildings, these fabrics would be tested before being used and would then be assumed to retain their flameproof qualities for one year. If, however, a fabric is treated in such manner that the fire retardents are merely present on the outer surface, where they may be brushed off in handling, blown or dusted off by moderate air currents present in the interiors of buildings, they may lose their flameproof qualities at any time under different circumstances, and may not, therefore, remain effective for a period of one year. For this reason it was felt that the fabrics should be finished, that is, the chemicals should be set or fixed before they are considered for testing; and that they should, even though finished, be subjected to a test to determine the permanence or durability of treatment, before making the flame tests. The following permanence or durability test was, therefore, developed :
A simple wind tunnel was designed, using a fan and a paper cone. In this tunnel the test specimens were exposed for five minutes, with the fan operating at a speed approximating 700 lineal feet per minute (eight miles per hour) through a seven-inch opening, as measured by a suitable anemometer. The specimens were suspended, one at a time, on a horizontal wire two inches above the center of the opening and whipped as a flag would be in a stiff breeze. Five minutes, with the air moving at this intensity, was considered sufficient time to whip the salts out of poorly treated fabrics which had no penetration or fixation.
The adoption of a five-minute period in the air stream was purely arbitrary. Experience in these tests shows that a shorter period at the same air-flow might give the same agitation to be expected over the period of a year in actual practice.
However, it is certain that the fiveminute period, while it may present greater difficulties to specimens, includes most possibilities and allows for a safety factor covering unusual applications. From several wind tunnel tests carried to ten and even 15minute periods, it becomes more apparent that some flame-retarding agents are relatively unaffected, there being only a slight loss of weight between the five and 15-minute periods. This loss is probably moisture, as borne out by the subsequent flame tests. Specimens failing the fiveminute test have an inherent weakness in that there is no fixation. This limitation has been commercially overcome so that there is no excuse for the use of such fabrics while they are known to be failing in this respect.
Tests on many specimens showed that this conditioning, or artificial weathering, actually threw off some of the salt, especially in the boraxboric acid mixtures. When subsequently tested with the whipped end exposed to the flame these samples tailed, and in some cases were entirely consumed. When the specimens were reversed, and the unwhipped ends exposed to the flame, better flame resistance was shown, especially in the height of the char. The flame resistance in such cases was comparable to that shown by specimens which had not been given a permanence test in the wind tunnel. Similar results, in a lesser degree, were obtained with the commercially prepared chemical mixture. The treated fabrics, which were obtained finished and ready for test, apparently lost mostly moisture in the whipping test, as evidenced bv low weight changes. These samples later passed the flame test successfully, showing marked resistance to fire.
Well-treated fabrics, therefore, passed this permanence test with facility, while inferior products failed and were eliminated at once.
Before entering into the actual flame testing of fabrics, a study was made of the work done on the subject, both in this country and abroad. It was recogtiiz.ed that the testing methods must be simple and effective, and inexpensive enough to create no hardship on the producer or the consumer. All available testing methods were, therefore, reviewed in the search for a simple test to meet the requirements of practical usage.
The testing method finally adopted was taken from the proposed specification of the National Bureau of Standards. Test specimens two inches wide and 12 1/2 inches long were cut from at least three different sections of the sample to assure a representative test of the fabric. These were numbered at one end, partly for identification of the sample and partly to differentiate the whipped and unwhipped ends of the specimens in the test to determine the permanence of treatment as before described.
To obviate any difference in moisture content, and thus provide a fair comparison, the samples, or specimens. were dried for 24 hours immediately preceding the flame test in a room, having an average temperature of 70° F. and an average humidity of 40 per cent. This method of approximating an average moisture content for all specimens was adopted after oven drying had proven impractical, because of the rapid moisture assimilation of fabric and the possible instability of certain treatments in the drying oven, where prevailing conditions are abnormal to those encountered in the general use of the fabric. Moisture determinations of the fabrics were considered too technical for rapid testing practice. Some drying is accomplished in the conditioning tests, as shown byweight losses in untreated samples, where only moisture could be thrown off. To retain this dry condition, the specimens are placed in a closed, drycontainer immediately after the conditioning, and thus kept until used in the flame test.
The apparatus used in the flame test is essentially the same as used by the National Bureau of Standards. A Bunsen burner and ring stand, with an adjustable clamp, are enclosed in a metal shield 10 inches square and 20 inches high to protect the flame from drafts. The front of the shield is left open, as the time of exposure of the specimen to the flame is not sufficient for the operation of a sliding door, and the tendency for the heated shield to set up a draft by acting as a chimney, is thus eliminated.
The Bunsen burner used has a tube of 3/8-inch inside diameter. With the air supply completely shut off. the burner is adjusted to give a luminous flame 1 1/2 inches long. By specifying this certain flame, a nearly constant temperature can be maintained without the use of a pyrometer.
Proper Temperature for the Igniting Flame
As there was a question concerning the proper temperature for the igniting flame, tests on treated fabrics were run with the burner giving a blue gas flame of 1,800° F. Even in this flame specimens successful in the comparatively low temperature yellow flame contributed nothing to combustion. It is the consensus of opinion that most fires in which fabrics have a part are at low temperature, and the ignition point of most fabrics is well below the temperature of a luminous yellow gas flame.
The test specimen is suspended vertically with 1/2-inch of the numbered end in the clamp and 12 inches of length exposed to the flame. By adjusting the clamp on the ring stand, the lower end of the specimen is set fa 3/4-inch above the Bunsen burner. Thus, when the burner is set under the specimen in the shield, 3/4-inch of the fabric is in the flame. The flame is applied for 12 seconds, then withdrawn and the behavior of the specimen noted.
Tests on Representative Types of Fabrics
To make the study thoroughly comprehensive, tests were made not only on the three available types of treatment mentioned early in this report, but also on a large number of representative types of fabrics of various weights, namely, rayon, cotton, linen, silk, wool, and combinations of these, differing in use from curtain to upholstery materials. To set the limits of the various reactions of specimens to the igniting flame, these fabrics were run through the entire testing procedure.
Certain rayons and light cottons were found to be highly retentive of any salt used as a fire retardant, because of their porous weave. Velvets, on the other hand, have a nap which, if agitated, loses salt readily. Despite these differences in texture, however, it is possible to obtain equally successful flameproof treatments. The only material which could not be rendered flameproof by either the commercial mixture or the laboratory mixture was a type of synthetic silk which had been “filled” with a substance similar to airplane dope. This particular “filler” prevented absorption of either fire-retardant salt used in the laboratorytreatments. It is conceivable that the commercial treatment might be successful with this material, but this was not tried.
Analysis of Flame Test Results
An analysis of the results of the flame tests shows that the major difference between light and heavy materials, comparing the results of all the flame tests, is the tendency of the former to char over a greater area. However, since charred area is considered a resultant and not a factor in flame spread, no reason was found for varying the flashing, flame duration, or glow requirements according to the weight of the material. Heavymaterials, especially’ those having a nap, also tend to glow longer, but since this glow is confined to the charred area, it may’ be overlooked.
Flashing occurred in less than one per cent of the specimens. Specimens which failed in this respect were light muslins which had been treated with the borax-boric acid mixture and then subjected to the conditioning test. According to the definition of flameproof, any specimen which flashed was considered a failure.
Since any appreciable continuation of flame after the removal of the igniting agent is also considered failing, a limit of three seconds was set for this reaction. All commercially treated fabrics met this test satisfactorily, as did the majority of the specimens treated with the commercially prepared salt. Incidentally, a limit of less than three seconds could not be noted with any degree of accuracy. The laboratory mixture, that is, the borax-boric acid mixture, failed to meet this limitation in a large number of cases. Apparently this treatment, while successful where permanence is not a factor, should be limited to cases where less than one year of flameproof protection is required. The trouble does not appear to be in the salt, as shown by research conducted by the British Fabrics Coordinating Committee, but in the method of setting or fixing the salt in the fabric.
The test specimens, which passed the flashing and duration of flame requirements successfully, also passed the glow requirement, averaging well under 20 seconds of glow on the edge of the charred area. This figure thus appeared to be within the range of a well-treated fabric, but not sufficiently high to BE unduly lenient.
The certification and approval, marking, and costs of tests are in accordance with the requirements and the general practice of the New York Building Department and the Testing Laboratories of Columbia University.
Following are the recommended requirements for fire tests of flameproof fabrics in condensed form :
Recommended Requirements for Fire Tests of Flameproofed Fabrics
These requirements apply to fabrics used for decorative or other purposes on the inside of buildings or other structures.
The fabric shall be finished after the flameproofing treatment.
The flameproof fabric, when subjected to the test described in Section 9, shall meet the following requirements as to fire-retardant properties:
- Flashing—No flashing shall occur at any time on the length of the test specimen. A flashing due to fuzz or bad edges is to be disregarded.
- Duration of Flame—The average continuation of flaming shall not exceed three seconds.
- Duration of Glow—The average continuation of glow at the edge of the charred area shall not exceed an additional 20 seconds after the cessation of flaming. Glow confined to the charred area is to be disregarded.
Samples for test shall be selected by and tested under the supervision of the authority having jurisdiction. The selection shall, at its option, be at the point of production or the place of use. Samples of each variety of cloth and every kind of treatment shall be selected for testing. The number of samples required shall be discretionary with the testing authority, according to the yardage submitted for approval.
Specimens for test shall be two inches wide and 12 1/2 inches long, and shall be numbered on one end. Test specimens shall be carefully cut from at least three different places in the sample to provide three test specimens from each sample for testing.
Conditioning of Test Specimens
The test samples or test specimens shall he dried for 24 hours in a room of average temperature of 70° F., with an average humidity of 40.
Test for Permanence of Chemicals
Test specimens shall be suspended by folding and pinning 1/2-inch of the numbered end over a horizontal 8-gage wire placed two inches above the center of a seven-inch opening in a wind tunnel having a wind velocity of 700 lineal feet, plus or minus 50 feet, per minute as measured by a suitable anemometer on a line with the wire. In this position the test specimen shall be whipped tor five minutes. Immediately upon conclusion of this test the specimen shall be placed in a closed, dry container until required for the flame test.
Apparatus for Flame Test
The apparatus required consists of a ring stand 20 inches high, with an adjustable clamp, housed in a metal shield 10 inches square and 20 inches high, open in front and at both ends, and a Bunsen or Tirrill gas burner, which has a tube of 3/8-inch inside diameter.
A whipped test specimen shall be taken from the closed container required in Section on Test for Permanence of Chemicals and suspended vertically with 1/2-inch of the numbered end in a clip held in a clamp on the ring stand, so that the test specimen shall hang vertically within and near the center of the metal shield required in Section 8. The lower end of the test specimen shall be 3/4-inch above the top of the gas burner, on which the air supply is shut off completely and adjusted to give a luminous flame 1 1/2 inches long. The flame is applied for 12 seconds, then withdrawn. The result of the test for duration of flame and duration of glow of the three test specimens shall each be averaged and recorded. Flashing, if any, shall be recorded.
Certification and Approval
Certification of test shall be supplied from an approved laboratory. Approval for fabrics used in the interior of buildings shall be effective for no more than one year, at which time a re-test shall be made, or the fabric shall he retreated and re-tested. The right is reserved to demand check tests at any time, and material found not to conform to these requirements must be removed and replaced with fabrics that will fully conform to the requirements of Section 3.
Fabrics that have been approved shall be marked with a suitable rubber stamp, or other device, showing date of approval by the authority granting the approval.
Cost of Tests
The required tests shall be at the expense of the owner or contractor.