Document xeD2q7a5Rg8k4ZnppOY4mNx0

Interoffice Communication To Safety Directors From Tom Grumbles Date March 12, 1984 Subject ESCAPE RESPIRATOR STUDY fconOCO) JJH: JCL TGG: RF n -e For your information, enclosed is a study of five commercially available escape respirators done by Haskell Laboratory. This report is worthwhile reading if you're planning on buying any such units in the near future. \ 0^ Thomas G. Grumbles ajo Enclosure > k S  <stnm> HASKELL LABORATORY FEB 1 ? 1394 bicii r -'A. J. 7 ESCAPE RESPIRATOR STUDY (Haskell Lab Report No. 573-83, MR-5526) SAi 00046360 1 TABLE OF CONTENTS Title Page I. SUMMARY II. BACKGROUND III. MATERIALS AND METHODS Choice of Respirators Test Equipment Test Panel Test Conditions Time Treadmill Speed Carbon Dioxide Buildup Oxygen Depletion Protection Factor (PF) IV. RESULTS Reliability & Quality Control Ease of Donning Completed Tests Protection Factors Personal Preference Individual Respirator Results ELSA Robertshaw Survivair Scram Ska Pak Ranking of Respirators 2- - | () () 0 0 4 6 3 & ^ TABLE OF CONTENTS (continued) V. DISCUSSION VI. ACKNOWLEDGEMENT VII. RECOMMENDATIONS VIII, REFERENCES Appendix A Tables 1-10 Figures 1-7 mi- 000046 / 3 HASKELL LABORATORY FOR DU PONT USS ONLY cc R. R. 3onczefc - Admin - D-9064 T. R. 3raun - mtl. - 0-4098-4 E. L. QeWhitt, Jr. - Conoco Oklahoma C. H. Evans - CSP - 3-6 21S ?. H. Fuller - Engg - L-13S1 P. G. Gilby - CSP - 3-6213 C. C. Griffith w. c. Haafi - Photo - FSFP - RSO-210 - 3-2302 E. n. Helmers - Engg - L-13S2 D. C. Hochstetler 3. w. Karrh, M.D. - Legal - SR - D--7083 - M--11543-3 R. N. Ligo, m.d. - SR - N--11400 R. L. Hanley S. M. Harks J. R. Martin - FSFP - 3-5202 - Rem. Arms - Bridgeport - ER - BOO-714 T. J. Nelson - F5F? - 3-5222 J. C. Olguin/s. S. Mileti - ?D J. G. Page, Jr. - ER J. L. Phillips - Legal - N--9498 - N-11543-2 - D-7083 3. Ramirez, M.D. - SR - N-11400 R. D. Richardson M. S. Sabla D. w. Smith - ACD T? - Admin - 3arley Mill 7 - N-3505 - 0-9063 H. A. Smith - PPD - M-5625 G. D. Thomely - FSFP - Barley Mill 9 January 24, 1984 C. F. RSINHAADT, M.2. CRSO HASKELL LABORATORY ESCAPE RESPIRATOR STUDY iaaa*aii Lao Report Mo. 57 3-33 ,'"mr-352S ) 4 I. SUMMARY Five escape type respirators were compared under moderate stress conditions on a panel of 16 male volunteers. Length of test, protection factor, carbon dioxide (CC^) buildup, oxygen (Cu) depletion and pulse rates were measured. Ease of donning, reliability and comfort were also studied. There was a large variation in the capability of the five brands of respirators to complete the test within the parameters chosen. Low airflow into the respirators appeared to be the critical factor for those units which most often did not meet the criteria for the tests. An overall summary of test results is given in Table 11. The chemical resistance of the respirator hood materials, although not investigated in this study, is considered in the Recommendations Section on page 17. II. BACKGROUND Several types of respirators have been developed \ for use in escaping from hazardous concentrations of aitf contaminants or from oxygen deficient atmospheres. They f range from simple mouthbit air purifying types for escape from low concentrations of chlorine to self contained breathing apparatus types. The choice of a given escape respirator depends on the anticipated hazardous situation. One type of escape respirator that is becoming popular employs a hood with a five- or fifteen-minute supply of air. These hood type supplied air escape respirators have several attributes which led to their popularity. They can be used in both contaminated and oxygen deficient atmospheres, can be used equally well by bearded and nonbearded persons, are relatively simple to don and use and are compact and storable for long periods of time. A survey of Du Pont plants in 1982(1) showed that several hundred hood type escape respirators were distributed for possible use at various sites. The predominant brand was the Robertshaw Air Capsule with a few Survivair escape units also in use. other escape respirators were the more traditional face mask/air cylinder type. Since that time, another unit (ELSA) has also been purchased by at least one plant site. One fact brought out in the survey report was that there were no documented instances where any of the hood type units had actually been used in an emergency. Although this is the hoped-for experience, it also meant that there were no field data that could be used to evaluate 5 the performance of the respirators in actual use. A purchasing decision had to be made based on vendor claims. National Institute of Occupational safety and Health (NIOSH) certification and a few published reports. NIOSH certification for escape respirators is described in 30 CFR Part II, Subpart H. It involves many tests, including four man-tests and a dummy head/breathing machine test for C02- The man tests include running, climbing and working in various combinations, and results are subjective, in that no measurements are taken. The respirator simply has to "satisfy the respiratory requirements of the wearer for the classified service time." CO- tests are done mechanically with a breathing machine which "exhales*1 air containing 5% CO- into the respirator facepiece. The respiration rate^is 14.5 respirations per minute with a minute-volume of 10.5 liters (10.5 Lpm breathi ng rate). Maximum allowable CO- concentration in the "inspired" air is 2.5% averaged over 3-4 respirations The averaged respirations are not taken at any particular time during the five minutes. f Respirator manufacturers may specify a minimum temperature at which their respirators are to be tested. All of the manufacturers of the hood type units used in this study requested a 20F test temperature. Therefore, the NIOSH certifications for the four hood models indicates that they are "approved for respiratory protection during escape only from oxygen deficient atmospheres, gases and vapors at temperatures above 20 F." The Ska Pak model (full face mask/air cylinder type) was tested at and certified for temperatures above -25 F. Studies on escape respirators have been done by Los Alamos Scientific Laboratory (LASL), National Aeronautics and space Administration (NASA) and the Air Force. One LASL study(2) compared ten escape respirators, including two hood types, for weight, comfort, ease of use and protection factor. CO- and 02 inside the masks or hoods were not measured. Tne report recommended that face mask type escape respirators be used in the pressure demand mode only. Hood type respirators provided less protection but had the advantages of being quickly donned and providing protection for bearded persons. SAL 00004 6 III. NASA has done some testing(3) of hooded escape respirators during which the test subjects were stressed on a treadmill. Measurements of 0* and CO., in the hoods were taken continuously. A qualitative fit test was done at the end of the walking exercise. In the NASA tests the ELSA unit was found to be effective and simple to use, the Survivair unit was found to be acceptable even though CO* concentrations were as high as 6%. The Robertshaw unit did not provide the protection that the other units did and was not considered acceptable for the particular escape conditions expected at the Kennedy space Center. The Air Force conducted a study(4) in which a hood type escape respirator (Robertshaw) was used to simulate an actual escape from a space launch complex. During the simulated escape, the test subjects' pulse rates were monitored, but no attempt was made to monitor CO* enrichment or protection factor. The report concluded that the Robertshaw Air Capsule was suitable for a slow (4 min., 19 sec.) escape from the complex but did not . have enough air flow to allow for a rapid escape. Our research involved the direct comparison of five escape respirators (four of the hood type) under a specific set of test conditions. By placing a treadmill inside a quantitative fit test booth, it was possible to measure the protection factor, CO* buildup, 0- depletion and pulse rate while the test subjects were under stress. MATERIALS AND METHODS Choice of Respirators Four currently available hood type escape respirators and one mask/air cylinder type were tested: - Robertshaw Model 5000 Air Capsule (referred to in report as "Robertshaw"). Survivair Five-Minute Emergency Escape Unit (referred to in report as "Survivair"). - Scott Scram Emergency Escape Breathing Device (referred to in report as "scram"). - ELSA Emergency Life Support Apparatus (referred to in report as "ELSA"). \ 000046366 7 Scott Ska-pak (referred to in report as "Ska PakM). See Figures 1-5 for pictures and detailed descriptions of the five respirators. Test Equipment The test equipment is illustrated schematically in Figure 6. A Dynatech-Frontier (DF) Model 259 H-l Laskin-type nozzle generator was used to generate a polydisperse di(2-ethy1-hexyl)phthalate (DEHP) aerosol (mean diameter 0.55 standard deviation 0.20 jum), maintained at approximately 25 + 5 mg/rrr in a 243 cubic foot DF Model 222-8/A test chamber containing a Trotter Model 320M Treadmill. A DF Model 259 H-2 Forward Light Scattering photometer was used to detect the ratio of chamber (C ) and respirator (C,) DEHP concentrations, resulting In a reported protection factor (PF)=C /C.. Sampled air from the respirator worn in the chamoer1was analyzed for CO, and O, content using Beckman Model j 864-38 Non-Dispersive Infrared Analyzer calibrated foi| CO2 and Beckman Model 755-02 paramagnetic Analyzers* The sample air was drawn at approximately 3.5 L/min. by the photometer pump and 1 L/min, from this primary line was HEPA-filtered and fed to the CO, and C>2 detectors at 0.8 and 0.2 L/min., respectively, via a Du Pont Model P-4000 personal sampling pump. Flowrates of the sample, calibration and zero gasses were measured with an external rotameter/manifold system. Two standard gases were used to calibrate the C02 and 0, detectors; one had 2.06% C02 and 5.2% O, , while the other had 4.97% CO2 and 18.8% 0, In addition, ambient air was used as a third^calibration point for o,. A Beckman Model ET-1204 audible alarm was tied into the of Detector and set to go off as O, fell below 18%. The CO, and 0, analyzers were connected to^a Hewlett-Packard Model 7100B-14-24 dual pen strip chart recorder. The photometer was connected to an Omniscribe Model B5116-1 strip chart recorder and interfaced with Hewlett-Packard Model 2645A personal computer terminal via a Fluke Model 8502A Multimeter (analog to digital converter). Both strip charts and computer program hard copies were obtained for all but three of the total 80 respirator tests. For these three tests computer hard copies were retained. A Computer Instrument Corporation Model 2867-4067 pulse rate meter with * Note: To address possible H,0 interference associated with the CO, detector, tubing whicn contained water droplets was used to^ collect an ambient air sample; no interference was seen. 3 i L 0 0 004636 7 fingertip sensor was used to monitor subject pulse rates. When checked against the standard method of taking pulse, the pulse meter was accurate to within +. 2 beats per minute. Test panel A panel of 16 men with facial dimensions representative of the male working population as described by Hack, et al.(5,6) for testing full-face respirators assembled from a pool of volunteer Haskell Laboratory employees is demonstrated in Figure 7, Anthropometric Panel (full-face). Anthropometric sliding and spreading calipers were used to measure face width and length.* The numbers in each box of Figure 7 represent the number of men required to be within that facial dimension range. The small number of female volunteers precluded the possibility of assembling a male-female panel. Du Pont medical records for each volunteer were reviewed by the Corporate Medical Division to confirm the abilities of subjects to withstand physical stress conditions presented by the test protocol. All volunteers were granted medical approval based on a combination of spirometric, x-ray^J blood pressure and other records. Prior to testing, the test apparatus and laboratory were reviewed and approved by a process Hazards Review Committee; and the test protocol was approved by the Human Studies Committee. The subjects attended an orientation meeting designed to explain the objectives of and their roles in completing the study. The test protocol was explained and informed consent obtained. Each individual was then trained on the treadmill with and without a respirator (no DEHP aerosol was used), in order to select a treadmill speed which would approximate 60-80% of their individual maximum allowable pulse rates, based on age.(7) Ages of the test subjects ranged from 25 to 58, with a mean of 32 years. Fifteen of the men were tested at 4 mph, 0% grade; and one was tested at 3.2 mph, 0% grade. Test Conditions The objective in setting up the test conditions for this study was to determine how the respirators would perform while the test subjects were under stress, and yet not create an unsafe condition inside the test chamber. Walking on the treadmill at 4 mph and 0% grade was judged to Face width widest distance from cheek bone to cheek bone. Face Length = distance from fleshy area under chin to point of lowest nasal depression between eyes on profile. 9 be a "fast walk" by the test subjects. However, the stress level and resultant breathing rates were less than those for climbing stairs at a fast pace or running. Based on medical consultation and published literature the following test conditions were set. If any of the test conditions were exceeded, the test was stopped. Time A test time of 3.5 minutes on the treadmill was chosen to allow 1.5 minutes for donning the respirator, entering the booth, attaching probes and exiting after the test. Treadmill Speed A brisk walk (4 mph) was used for 15 of 16 test subjects. The sixteenth subject used a lower treadmill speed of 3.2 mph to keep his heartrate below the 80% of maximum (based on age) that was allowed. Carbon Dioxide Buildup J If CO- inside the mask or hood exceeded 3%, the test was stopped. The 3% limit was specified by our medical consultant to avoid excessive stress. The TLV--STEL for COis 1.5% and the NIOSH Criteria Document (1976) for CO-(8) recommended a top excursion value of 3% based on a 10-minute sampling period. in five supplemental tests under medical supervision the C02 limit of 3% was raised somewhat. Oxygen Depletion The cutoff point for oxygen chosen for the study was 18% minimum inside the mask or hood. The normal concentration of oxygen in air is 20.9% and OSHA requires at least 19.5% in any work environment. Impairment of mental performance is soon detected at a partial pressure of 120 torr, equivalent to 15.8% at sea level(9). Protection Factor (PF) As with all quantitative fit testing, the test was not continued if the respirator failed to maintain a PF of at least 10 during any part of the test. That PF was used to ensure test subjects were not exposed above the 5 mg/nr AEL for the DEHP in the test booth. protocol The test protocol used in this study is given in Appendix A. 10 IV. RESULTS Reliability & Quality Control in all 80 tests, none of the respirators malfunctioned or had any quality control defects which affected their use. There were some small brown spots on the hoods of some of the ELSA units, but they did not affect visibiity. The airflows lasted for at least as long as advertised by the manufacturers. Ease of Donning Generally, the respirators were easy to remove from their cases and don, except for persons wearing glasses. The elastic or rubber neck bands were hard to pull over glasses, especially glasses with large frames. The Robertshaw unit carrying case had a Velcro fastener that looked like it was part of the shoulder strap; several subjects had difficulty locating the ' fastener without prompting. Also, the Robertshaw model had a drawstring around the neck area that was difficult to operate and appeared unnecessary. The Survivair unit has an activating ring that can accidentally be pulled when the hood is removed from the case. That actually happened prior to the test program when one of the units was being unpacked to install the Quantitative Fit Test (QNFT) probe. Donning the Ska pak unit was slow in some cases because hair had to be held away from the facepiece. Completed Tests Table 1 is a summary of the test results. It shows that the number of completed tests ranged from 0 of 16 for the Robertshaw model to 13 of 16 for the Scram. Reasons for not completing the test are also shown in Table 1. The principal reason for terminating tests was that the 3% CO- limit was exceeded; 14 Survivair unit tests were stopped because of CO- only, while 14 of the Robertshaw Air Capsule tests were stopped because of both CO_ and 0* limits being exceeded simultaneously and 2 because of C02 only. Tne number of tests not completed because of CObuildup or oxygen depletion seems to be directly related to airflow as can be seen by Table 1A. 0 0 00 463 /'1 11 protection Factors Table 2 summarizes the protection factors achieved by the respirators during the tests. All of the hood type devices achieved PF1s of at least 100 with 83% of them having at least a 1000 PF. The Ska Pak's protection factors covered a wide range with three failing to provide a PF of 10 and one providing a PF of more than 10,000. Personal preference In addition to the measurements of respirator fit and breathing air analysis, the subjects rated the respirators by personal preference; they were asked which would be preferred in an emergency situation. Scores assigned by subjects to the respirators are tabulated in Table 3 - Personal Preference of Test Subjects. The mean scores are listed in the last row. The significance of the rating scores will be discussed later. Individual Respirator Results ELSA Table 4 summarizes the test data for the ELSA respirator. Twelve of sixteen subjects completed the 3 1/2-minute test within the parameters chosen. Of the tests that were not completed, one was stopped because of low oxygen and three were stopped because of excess C09. The minimum protection factor for the ELSA was over 3000 with the median value being > 10,000*. Users found the ELSA unit to be easy to don and relatively comfortable, although some complained of the tightness of the neck strap. The units, using regular 2216 psig air cylinder, were simple to refill and clean, which eases the training burden. Robertshaw Air Capsule Table 5 summarizes the test data for the Robertshaw model respirator. None of the test subjects completed the 3 1/2-minute test within the parameters chosen. In most (12/16) of the cases. All measured values for PF over 10,000 are reported as 10,000 in this report, since this respresents the reliable limit of sensitivity of the photometer. ;ml 0004>0463'? 1 12 the oxygen depletion to below 18% and the carbon dioxide buildup to more han 3% occurred nearly simultaneously. Since none of the Robertshaw units completed the original tests, the question was raised as to how they would perform against less stringent criteria. Under constant medical supervision one of the test subjects tested three additional Robertshaw respirators beyond the previous CO- and 0- limits. In these tests there was no DEHP^in the QNFT booth and the doors were left open. All three tests were terminated in two minutes or less when the O- level dipped below 16%. At that point the CO- levels were between 4.7 and 5.1%. The test subject was somewhat uncomfortable at the end of the tests and felt relieved after removing the hood, but he was not dizzy or unduly stressed. See Table 6 for data from supplementary tests. As mentioned previously, the Robertshaw was somewhat difficult to remove from the case and don. Most of the test subjects complained of a choking sensation while wearing the unit, due to the tight elastic neck band. Many had to be prompted on how to loosen the added drawstring around the neck at the end of the test. Since the Robertshaw uses a 5000 tubing air supply resevoir, it can be with special equipment. The training start/stop valve but the regular unit turned off once it is activated. psig steel refilled only model has a cannot be The protection factors for this respirator ranged from 350 to 10,000, with the median value being 2600, Survivair Respirator Table 7 summarizes the test data for the Survivair unit. Fourteen of sixteen subjects did not complete the test because of CO- buildup exceeding 3%. The oxygen levels all remained over 21% because this respirator utilizes air enriched to 28% 02* As with the Robertshaw respirator, some supplemental tests were run on the Survivair* A test subject tested two additional units under constant medical supervision (Table 6). In one case, because of a timing error, the test was stopped at 3 1/4 minutes. At that point the CO^ p pi j () () ()() 4 6 J / 13 level was 4,3% and the 0- level was 23.2%. In the second test, the levels after 3 1/2 minutes were 2.9% for CO- and 25.3* for n . tests indiclte that The supplemental would have been an increase in the number of Survivair units completing the original tests, if the allowable CO- level could have been safely extended to 4 or 5%. The protection factors for the Survivair respirators were excellent, with just one unit below the 10,000 level. The median protection factor was > 10,000. Generally, the respirators were easy to don and comfortable. The twin cylinders are not refillable and must be replaced each time the unit is used. Scram Table 8 summarizes the test data for the ; Scram respirator. This unit had the highest number (13/16) of completed tests. The three teats not completed were stopped because of CO2 buildup. In all cases the o2 level remained at or above 21%. Protection factors for the scram were variable. The range was from 113 to 10,000 with a median value of 2550. The Scram unit was easy to don. Almost universally the subjects complained of the heat buildup in the hood because of the oxygen generating reaction. The Scram unit is not rechargeable, although the manufacturer does sell a training model that allows the trainee to pull a resettable pin and place the hood over his/her neck. No air or oxygen is flowing in the test unit, but holes in the neck grommet allow air to enter the hood during the training exercise. Ska Pak Table 9 summarizes the test data for the Scott Ska Pak. Ten of sixteen persons completed the 3 1/2minute test. The Ska Pak was the only respirator which failed to achieve even a 10 PF for some test subjects (3 of 16). The PF1s ranged from less than 10 to more than 10,000 with a median value of 320. 14 Those who did achieve a good fit seemed to like the ska Pak, and some mentioned that they felt "secure" using it. This unit uses a refillable 2216 psig cylinder and a standard full face mask. Ranking of Respirators The test results for CO-, 0-, time, PF and personal preference were evaluated using Friedman's Test(10,ll), a nonparametric technique for comparing more than two matched groups. Matching was considered important due to conditioning and individual differences among the test subjects. Using this test, a significant difference was found to exist among the five respirators in their performance during the tests and in the personal preference for them by the test subjects. The Studentized Range Test(12) was then used to determine which respirators ranked higher and which ranked lower. Table 10 shows the ranking groups. V. DISCUSSION All escape respirators have a single purpose: to provide respirable air for a short period of time in emergency situations. The quality of the air in the respirator need not necessarily remain the same as normal air over the entire escape period, but it has to be good enough to allow for escape under stressful conditions without permanent after effects. It was not the purpose of this study to determine just how poor the air inside a respirator can get and still be useful for escape, but to see how various escape respirators performed under moderate stress conditions within a common set of criteria. The results allow us to make some useful observations: Airflow into a hood is critical in maintaining low carbon dioxide content (see Table 1A) for respirators that do not scrub out the C0-. The use of oxygen enriched air allowed tne Survivair unit (28 Lpm airflow) to maintain the O- level in the hood but did not prevent the CO- level from exceeding the 3% limit. The Rooertshaw model had the lowest airflow (25.5 +_ 2 Lpm). It did not complete any tests. {\ o o ft i 15 VI. Comfort is relatively unimportant unless it gets to the point where potential users would shun a particular respirator in an emergency. The Scram respirator drew almost universal complaints from the test subjects because of heat buildup inside the hood, but it also had the highest percentage of completed tests. The very short use time for this type of respirator may allow us to discount the temporary discomfort involved. Training is very important in any program which uses emergency escape respirators. Not only how to use them but when to use them, should be addressed. Training is necessary to overcome the natural apprehension about using a device which covers the head and to provide information about the limitations of the devices. - The protection factors provided by the escape respirators varied considerably. The adequacy of the PF depends on the anticipated use of each device. - The results of these tests point out the inherent problem with mask type respirators used for escape (Ska Pak). They do not provide a universal fit. To be useful, mask type units should be used only by those individuals who have passed a fit test. Beards and other interferences with a good seal make consistently passing such a test difficult. For those who do get a good fit, the mask/air cylinder escape respirator still appears to be a good choice. The fact that the hood type respirators were certified by NIOSH for use above 20F raises the question about their use in very cold locations. The useful low temperature for these devices may be much lower than 20F? this issue must be addressed when making a decision about their use. RECOMMENDATIONS Each potential purchaser of escape respirators should determine the anticipated time required to escape from a typical emergency situation and the required protection factor. 00046375 16 VI. VIII. The information in Table 11 should be used to determine the best candidate respirators for the intended use. The cost per unit is summarized in Table 12. If one of the tested respirators is already in use or is being considered for purchase and the median values in Table 11 for time and PF do not meet those required for the intended use, then a field test should be done to determine its adequacy. The field test should be done only after consulting with the departmental occupational health coordinator. Haskell Laboratory's monograph Chemical Resistance of Protective Clothing should be consulted for use situations where chemical resistance problems are anticipated. Haskell Laboratory can assist if additional chemical tests are required. ACKNOWLEDGEMENT The work of L. P. Percival in the initial planning and coordinating of this study is acknowledged, as is the assistance of R. L. Trivits in conducting the testing, the assistance of G. J. Graepel in doing statistical analyses and the work of J. A. Perry in computerizing the fit test readings. REFERENCES 1. Haskell Laboratory Report No. 327-82, Escape Respirators, MR-5468-001, June 2, 1982. 2. Hack, A. L., et al., Evaluation and Performance of Escape Type Self-Contained Breathing Apparatus," Report No. NUREG. CR-1586, LA-8432-PR, Los Alamos Scientific Laboratory, October 1, 1978 - September 30, 1979. 3. Private communication, D. F. Doerr (NASA) to R. W. Pell (Haskell Laboratory), May 20, 1983. 4. Myhre, L. G., "Field Study Determination of Ventilatory Requirements of Men Rapidly Evacuating a Space Launch Complex, Report SAM-TR-80-43, USAF School of Aerospace Medicine, Brooks AFB, Texas. 5. Hack, A. L., et al., "Selection of Respirator Test Panels Representative of U.S. Adult Facial Sizes," Report No. LA-5488, Los Alamos Scientific Laboratory, December, 1973 - March, 1974. 17 3 A L. 0 0 0046 3 / 6 6. Hack, A, L., et al., "Respirator Protection Factors: Part I - Development of an Anthropometric Test Panel," Am, Ind. Hyq. Assoc. 2i=970-975 (1978 ). 7. "Exercise and Your Heart," U.S. Dept, of Health and Human Services, Public Health Service, N.I.H. Publication No. 81-1677, May, 1981. 8. U.S. Department of Health & Human Resources (NIOSH): Criteria for a Recommended Standard, Occupational Exposure to Carbon Dioxide: Publication No. 76-194, p. 2 (1976). 9. Beard, R. R: "Inorganic Compounds of Oxygen, Nitrogen and Carbon," Industrial Hygiene and Toxicology (F. A. Patty^ ed.), Third Revised Edition, Interscience Publishers, New York, NY, Vol. II, p. 4054 (1982). 10. Conover, W. J., Practical Nonparametric Statistics, John Wiley and Sons, 1971. 11. Miller, I. and Freund, J. E, Probability and Statistics for Engineers, Prentice-Hall Mathematics Series, 1965. 12. Snedecor and Cochran, Statistical Methods, Iowa State Univ. Press, 1980. 00 A /, / > 80 18 Report by: James Wj Kayser Consultant Industrial Hygiene Approved by Stephen W. Dixon Section Supervisor Industrial Hygiene K. Frieda Fisher Industrial Hygienist (LJL 0, ru^ Frank A. Bower Associate Director Field Services Division JMK/KFF/dar Date Issued: January 24, 1984 Haskell Laboratory Report 573-83 Attachments: Appendix A, Tables 1-10, Number of pages in this report: 40 Figures 1-7 19 9AU 0 0 '*1 APPENDIX A TEST PROTOCOL 1. Determine that subject has completed preliminary requirements: a. Medical Approval b. Consent Form c. Orientation d. Training e. Maximum pulse rate for age determined. 2\ Require subject to have resting pulse rate checked at Medical. 3. Determine from random number chart the order in which the five respirators will be tested for the particular' subject. 4. Explain test to subject. 5. Check instrumentation, 6. Perform QNFT: a. Record "baseline'' at anticipated use range. b. Record "full scale" (100% range). c. Have test subject don respirator, enter booth and plug in. Note any problems with donning and record time when respirator was activated. d. Have test subject start treadmill and begin walking. e. Record "normal breathing" at range used for baseline for 3.5 minutes. f. Stop test short if: 1. Requested by subject for any reason. 2. Respirator or equipment malfunctions. 3. Oxygen level falls below 18%. 20 4. Carbon dioxide levels goes above 3.0%. 5. Pulse rate goes above 80% of maximum allowaDle. g. After 3.5 minutes, ask test subject to "unplug," leave booth and remove respirator h. Record baseline as in #1 above. i. Immediately record full scale. j. Record the time that the respirator has a noticeable drop off in air supply. 7. Note: a. Ease & quickness of donning. b Special training requirements. c. Vision restrictions, fogging. d. Comfort. 8. Allow 5-10 minutes rest between respirator tests. 9. After all tests are done, ask the subject to rate the respirators from 1 to 6, personal preference. Note comments. JMK/KFF/dar HL 573-83 00004*380 21 DATA SUMMARY: TA3LZ 1 SUM3SR Or COMPLSTS/INC5MPL2TS (16 tocai testa) Respirators ELSA* Robertshaw Survivair Scram* Ska Pak* Complete Test 12 0 2 13 10 ?? (only) 0 0 0 0 3 Reason for Incomplete Test 3 2 (only) 1 CO^ (onlv) 2 3ot.n 3-, and C3> 4 0 2 L4 0 14 0 03 Q 1 TABLE 1A RELATION OF AIRFLOW TO INCOMPLETE TESTS Respirator Ska Pak9 Scram* ELSA* Survivair Rocercshaw Airflow (vendor data ? On demand 30 Lpm (recirculated) 33 Lpm (continuous) 23 Lpm (continuous) 25.5 * 2 Lpm (continuous' Tests Not Completed Because of 0i-.-- -CO i- or 31 otn -- 3 3 15 ;) 00 0463 Q:!. TA3L5 2 PROTECTION -ACTOR <?F? DISTRIBUTION < 10 10-100 SLSA* 00 Robertsnaw Q 0 Survivair 0 0 Scram-* 3 0 Ska ?ak* 3(25* 3(1 ) 101-1,000 0 3 0 3 4(1) 1,00 1-10,000 2 11 1 2 5(0) > 10,000 14 2 15 $ 1(0 ) Approximaca median ?? Values 10,000 2,500 10,000 2.500 300 Mumper in parentheses shows cr.ose in jroup with beards. TABLE 3 PERSONAL PREFERENCE RANKING Subject ELSA 11 21 32 42 53 62 71 84 91 10 2 11 3 12 3 13 2 14 2 15 2 16 1 TTOQ Score 1 * most preferred 5 = least preferred Robertshaw 4 4 4 5 4 5 3 5 5 5 5 4 4 4 4 5 4.38 Survivair 3 3 3 3 2 3 2 2 2 3 1 2 3 1 1 2 2.25 Scram 5 5 5 4 5 4 4 3 4 4 4 1 5 5 5 4 T7T9 Ska Pak 2 2 1 1 1 1 5 1 3 1 2 5 1 3 3 _3_____ 2.19 Mean HL 573-83 - 24 - SAI. 4 6 TABLE 4 INDIVIDUAL TEST RESULTS FOR THE ELSA RESPIRATOR ;Ct No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Time (min.) 2.73 3.50 3.50 2.33 3.50 3.50 3.50 3.50 3.50 3.50 3.50 1.45 3.50 3.50 2.65 3.50 CO, (Max, %) 2.75 1.45 1.95 3.15 2.45 1.75 2.90 1.80 1.20 1.15 2.30 3.10 1.95 1.52 3.25 2.50 2 (Min. %) 17.9 19.8 19.5 18.0 18.7 19.6 18.2 19.4 20.0 19.8 18.6 17.8 19.2 19.4 18.0 18.8 PF 10,000 10,000 4,780 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 3,448 10,000 10,000 10,000 10,000 Median 3.50 Median 10,000 HL 573-83 - 25 - , "i f'j O 0 4 & 3AI. TABLE 5 INDIVIDUAL TEST :RESULTS FOR THE ROBERTSHAW AIR CAPSULE Subject No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Time {min .) 0.83 1.11 1.00 1.23 0.95 0.45 0.62 0,30 1.03 1.08 1.12 0.67 0.38 1.63 0.60 0.96 CO^ (Max. %> 3.35 3.30 3.30 3.75 3.20 3.20 3.30 3.30 3.55 3.25 3.20 3,60 3.55 3.20 3.40 3.20 0,, {Min. %) 17.2 17.2 17.3 16.5 18.0 18.5 17.5 17.9 17.5 17.2 17.6 16.5 17.9 17.5 16.5 17.0 PF 3,333 5,263 2,083 351 7,692 935 1,667 3,226 1,613 6,250 1,220 1,351 10,000 5,556 943 10,000 Median - 0.96 Median = 2,600 HL 573-83 - 26 - TABLE 6 SUPPLEMENTARY TESTS FOR & C02 (one test subject, treadmill only, no DEHP, medical supervision) <#> CM O Test Robertshaw 1 Robertshaw 2 Robertshaw 3 Survivair 1 Survivair 2 Time, Minutes 2 2 1/4 2 3 1/4 3 1/2 15.5 15.5 15.5 23.2 25.3 002, 4.7 4.7 5.1 4.3 2.9 HL 573-83 - 27 - SAL- 0 0 O t.V ' 1 TABLE 7 INDIVIDUAL TEST RESULTS FOR THE SURVIVAIR ESCAPE UNIT Subject No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Time (min.) 1.43 2.15 3.50 1.04 2.51 2.80 2.13 1.74 3.50 2.15 3.50 2.07 1.85 1.98 1.08 1.71 CCU (Max. %) 3.15 3.25 2.80 3.80 3.30 3.30 3.35 3.10 2.78 3.30 4.90 3.30 3.20 3.35 3.60 3.05 0^ (Min. %) 21.2 21.5 21.4 22.5 21,0 21.6 21.5 21.3 21.5 21.0 22.5 22.0 21.4 21.5 21.3 21.4 PF 10,000 2,500 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 10,000 Median = 2.10 Median = 10,000 HL 573-83 - 28 - TABLE 8 INDIVIDUAL TEST RESULTS FOR THE SCRAM RESPIRATOR Subiect No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Time (sec.) 3.50 3.50 3.50 1.75 3.50 3.50 0.96 2.05 3.50 3,50 3.50 3.50 3.50 3.50 3.50 3.50 C02 (Max.%) 3.20 2.30 1.85 3.50 0.65 2.90 3.95 3.10 2.90 1.95 2.00 2.20 2.15 1.60 2.60 2.80 22 (Min. %) 21.1 21.5 21.0 21.0 21.4 21.6 21.5 21.4 21.5 21.0 22.0 21.8 21.3 21.0 21.0 21.0 PF 10,000 4,800 749 588 363 377 10,000 500 163 10,000 485 4,348 10,000 10,000 113 10,000 Median = 3.50 Median = 2,500 HL 573-33 - 29 - ^ [_ () () 0 0 4 6 3 3 3 TA3L 9 IN3IVI3UA1 T5ST SSS'JITS :03 TH z 3KA ?&;< 3SS?I3aT0R Sud*ect So1 2 3 4 5 0 / 3 9 10 i_ i 12 13 14 15 IS Tia* (ain.) (cctaolata test = 3.5 nin.1 3.50 1.50 1.69 2.50* 3.50 3.50 o.so 3.50 0.00 3.50 3.50 3.50 3.50 3.50 2.03 Oi/ I CO-, (Max. 2. 15 2.70 3.55 2.50 2.10 2. 40 2.15 2.30 1.90 2.43 2.30 2.30 2.45 1.90 2.55 2.00 0, 1 Min. 31 19.9 17.5 17.9 13.2 19.2 19.7 19.5 13.5 19.5 13.3 13.5 13.0 13.4 Id . 6 1.3. 2 13.7 ?? 736 22 10,000 753 1,232 9,336 < 10 2,053 < 10 4,369 133 303 1.353 12 24 1 Median 3.50 Medi an 320 v Ran 3u: of air cecause face nask was difficult to con, o., ar.d C3, and ?f vera c:<iy at t.te enc of one test tine (2 1/2 nin.'T * TABLE 10 GROUPINGS OF STATISTICAL SIGNIFICANCE Tested Condition Personal preference Protection factor CO2 buildup to 3% *-- Best ELSA Ska Pak Survivair ELSA Survivair Ska Pak O2 depletion to 18% Time (complete tests) Scram Survivair ELSA Ska Pak Scram Groupinqs* Robertshaw Scram Robertshaw Scram ELSA Scram ELSA Ska Pak Survivair Worst Ska Pak Robertshaw Survivair Robertshaw Robertshaw * Each group is significantly different from the next, but there is no significant difference between respirators within a group. HL 573-83 - 31 - 0 0 0046 39 0 TABLE 11 TEST TIME AND PROTECTION FACTOR SUMMARY Time within Test Criteria 3.5 minutes max.) Protection Factor ELSA Robertshaw Survivair Scram Ska Pak Range minutes) 1.45-3.50 0.30-1.63 1.04-3.50 0.46-3.50 0.0-3.50 Median 3.50 0.96 2.10 3.50 3.50 Ranqe 3450-104 350-104 2500-104 110-104 < 10-104 Median Temperature at which Certified 104 Above 20 F 2660 104 Above 20 F Above 20 F 2550 Above 20 F 320 Above -25 0 F Test Criteria - 3.5-minute brisk walk (4 mph), > 18% 02, < 3% C02, PF > 10, no malfunctions, adequate comfort. HL 573-83 - 32 - SAL > 0 0 4 6 3 9 Respirator ELSA Robertshaw Survivair Scram Ska Pak TABLE 12 ESCAPE RESPIRATOR COSTS Approx. Unit Cost (1984) $217 $215 $236 $225 $485 Approx. Service Cost Per Use $10 (on site) $12.50 (vendor refill) $38.00 (new cylinders & cleanir (not reuseable) $10 (on site) 8 AI... 0 00 0 4 63 V >: 33 Manufacturer: NIOSH Certification: Serial Numbers of Test Units: Description: International Safety Systems, Inc. 480 Perry Street Lawrenceville, GA 30245 TC-13F-111 02906, 03775, 04536, 04041, 03725, 03757 A five-minute compressed air breathing apparatus consisting of an aluminum cylinder containing 7 ft. of air at 2216 psig; a valve arrangement consisting of an on/off handwheel, reducing valve, pressure gauge, and flow control oriface? a flexible supply hose? and a clear polyurethane hood with an elastic collar. The entire unit is stored in a bright red flexible bag with a Velcro closure. The air flow to the hood is 38 liters/min. Polyvinyl chloride hood also available. HL 573-83 Figure 1 34 000046393 Manufacturer: NIOSH Certification Serial Numbers of Test Units: Description: Robertshaw Controls Company 333 N. Euclid Way Anaheim, CA 92803 TC-13F-28 3090015 through 3090020 An escape device that supplies five minutes of air to the user through a clear, polyurethane hood, which has an elastic neck closure as well as a draw cord. The breathing air is stored in coiled, stainless steel tubing at an operating pressure of 5000 psig. Airflow to the hood is actuated by pulling a start ring. The Air Capsule is stored in a rigid plastic case with a Velcro closure. The flow rate of air is 25.5 + 2 liters/min. at 1000-5000 psig. The unit is reuseable buc require* special equipment to recharge to 5000 psig. Polyvinyl chloride hood also available. HL 573-83 Figure 2 - 35 - Survivair Five-Minute Emergency Air Escape Unit P/N 0028-00 Manufacturer: U.S.D. Corporation Survivair Division Santa Ana, CA NIOSH Certification: TC-13F-86 Serial Numbers of Test Units: None - all from Lot #17530 Description: The emergency escape unit consists of a transparent polyurethane hood, an integrated cylinder manifold and regulator assembly and a protective case. (Both hard case and soft case models are available.) The twin cylinders are prefilled with 28% oxygen enriched air. When initially assembled, the outlet of each air cylinder is sealed with a frangible copper disc. When the user pulls the actuating ring, a cam-accuated activating piston pierces this disc and allows air to flow into the pressure reducing regulator. The regulator maintains a constant outlet flowrate of approximately 28 liters of air per minute, which is then directed into the hood. Once the actuating ring has been pulled, the unit will provide a constant flow of air into the hood until the cylinders are depleted. New cylinders must than be purchased. HL 573-83 Figure 3 - 36 ,00$ A 63 Scott Scram Emergency Escape Breathing Device P/N 802300-02 Manufacturer Scott Aviation (Division of ATO) Lancaster, NY 14086 NIOSH Certification TC-13F-88 Serial Numbers of Test Units: Not serialized. Description A 15-minute escape device consisting of a loose fitting Tef1on-coated fiberglass hood attached to a life support unit. In the life support unit low pressure oxygen is produced by an exothermic reaction which produces sodium chloride and oxygen from a sodium chlorate core. The oxygen is fed to the primary nozzle of an ejector device which aspirates C02-lac^en hood gases from the hood through a lithium hydroxide scrubber, moisture absorber and filter? then back to the hood. Oxygen duration is 15 minutes, 02 output is 90 liters and the circulation rate is 50 liters/minute. The unit is not reuseable. It is stored in a rigid plastic case. A training model is available. HL 573-83 Figure 4 37 tl. 000046396 Manufacturer NIOSH Certification Serial Numbers of Test Units: Description: Scott Aviation (Division AT0 5 Lancaster, NY 14086 TC-13F-68 No serial number on mask. Cylinder serial numbers were A31048, A33092, A32602, A32059, A32546 and A32582 A five-minute escape device consisting of a cylinder containing 7 ft. of air at 2216 psig, pressure regulator, flexible hose, Scott-o-vista mask with mask-mounted regulator operating in the demand mode. To operate, turn on the cylinder valve, place the mask on the face and breathe normally. HL 573-83 Figure 5 38 SAL 000-463 * PF \\ L HL 573-83 Figure 6 Schematic of Apparatus 39 SAL 0000' LENGTH (mm) 153.5 la --t 133.5 WIDTH (mm) LA LA CD LA ro ^X 2 i--l --1 1 125.5 2 42 115,5 107,5 1 21 Figure 7 Anthropometric Panel (Full-face) HL 573-83 - 40 - SAL. 000 0463 9'; j