SECRETARY OF LABOR, Complainant, v. CONTINENTAL OIL COMPANY, (DENVER REFINERY), a corporation, Respondent, OSHRC DOCKET NO. 79-0570-E OIL, CHEMICAL, AND ATOMIC WORKERS INTERNATIONAL UNION, AFL-CIO, Authorized Employee Representative. Appearances: Phyllis K. Caldwell, Esq. and Donald R. McCoy, Esq., of Denver, Colorado, for Complainant, Eiberger, Stacy & Smith, of Denver, Colorado for Respondent, Robert E. Wages, Esq.. of Denver, Colorado for the Authorized Employee Representative. DECISION AND ORDER Cronin, Judge, OSHRC: On October 3, 1978, a major fire and explosion occurred at Continental Oil's refinery (CONOCO) killing three employees, hospitalizing twelve persons and causing extensive damage to the Conoco plant and surrounding area. Representatives of the Secretary of Labor commenced their investigation into this matter S:\1984\Processed\79-0570-E\00000000\04.tif " on October 3rd and concluded it on January 12, 1979. As a result of this investigation, the Secretary issued three citations to Conoco under the Occupational Safety and Health Act of 1970 (29 U.S.C. §651 et seq.; hereafter called the Act) alleging 13 serious violations, two other than serious violations, and one willful violation of the Act. Penalties in the amount of $16,480.00 were proposed. On February 1, 1979, the Secretary issued an amended citation amending one serious violation to willful and proposing an additional penalty of $10,000.00. On February 2, 1979, respondent filed its notice of contest, conceding three violations, paying proposed penalties in the amount of $360.00, and contesting all other violations and penalties. Prior to the hearing, Conoco agreed to withdraw its notice of contest to three serious violations; the Secretary agreed to dismiss four other items of the citations. The provisions of this agreement will be incorporated into this order (See Judge's Exhibit J-D). The hearing in this case commenced on October 22, 1979 and closed on November 16, 1979. On January 7, 1980, the Commission accepted the Secretary's interlocutory appeal of this judge's ruling of November 16, 1979. The Commission, on April 27, 1981, upheld the November 16th ruling and remanded this case for further proceedings. Extensive briefs and reply briefs were filed by the parties September 21 and October 15, 1981. Issues The alleged violations remaining at issue are Items 2, 3, 5 and 11 of Citation No. 1. and Items 1 and 2 of Citation No. 2. Each of these alleged violations will be addressed in order. Citation No. 1, Item 2 As amended, this item charges Conoco with failing to provide emergency escape respirators on its sour water stripper tower and hydrogen sulfide absorber tower in S:\1984\Processed\79-0570-E\00000000\05.tif " violation of 29 C.F.R. §1910.134(a)(2), which provides in pertinent part: (2) Respirators shall be provided by the employer when such such equipment is necessary to protect the health of the employee. The employers shall provide the respirators where applicable and suitable for the purpose intended... Summary of Evidence The record reflects that both towers contained hazardous concentrations of hydrogen sulfide (H2S) in a closed system (Tr. 802-805, 813). It also is undisputed that employees were required to go onto the platforms of these towers in the course of their duties. According to the Secretary's compliance officer, however, these employees were not actually exposed to a H2S contaminated atmosphere under normal working conditions (Tr. 49-50, 294, 2264, 2278). Conoco did not locate emergency respiratora on either tower or require employees to carry such respirators on their person. Conoco, however, positioned an H2S monitor on the first level of the sour water tower due to the possible presence of H2S at that location (Tr. 2286). Other uncontradicted evidence establishes that no Conoco employee has ever been exposed to H2S when working on either of the t~ in question. Further, there is no indication in the record that there has ever a release of H 2S from these towers (Tr. 2270). H2S is a highly toxic gas which can cause immediate unconsciousness and death, even at relatively low concentrations (Tr. 40; Ex. C-1, p.7). Exposure to H 2S is a universally recognized hazard in the petroleum industry, and its potentially fatal effects are recognized by Conoco in its safety manual (Tr. 47-48; Ex. C-7, p.14). The Secretary's witness, Mr. Quinlivan, testified that the accidental release of H2S in a petroleum refinery is a "possibility" and a "fairly common occurrence" (Tr. 799-800). In his opinion, the sour water stripper tower and the H2S absorber tower are sources of H2S releases (Tr. 801). Mr. Quinlivan testified that the flanged joints, control instruments, and the reflux drum on the sour water stripper tower S:\1984\Processed\79-0570-E\00000000\06.tif " and the flanged joints on the reboiler, and the vapor line and process control instruments on the H2S absorber tower, might be susceptible to accidental releases Of H2S due to the corrosion (Tr. 805, F16). On direct examination Mr. Quinlivan was asked to assume that proper materials were used to inhibit corrosion and regular inspections made to detect the possibility of corrosion on these towers. He then was asked whether any possibility of an H2S release existed. He answered "Yes." Although asked for his opinion on how this possibility would arise, he never gave it (Tr. 808-809). Later, on cross-examination, Mr. Quinlivan conceded that the possibility of an H2S release resulting from corrosion would be eliminated by the use of materials to inhibit corrosion, regular inspections to detect corrosion, and proper record-keeping (Tr. 943). Although the record is silent on whether materials to inhibit corrosion were used, Conoco does inspect the towers for corrosion "onstream" on an annual basis and internally every other year, and keeps records of these inspections (Tr. 2255-2257). No evidence was introduced to establish that Conoco's practices in this regard were inadequate. Conoco employees were informed about the dangers of exposure to H2S, and Instructed to warn the control room and other employees when H2S is detected. Signs warning of the possible presence of H2S and H2S alarm monitors were located throughout the refinery. Employees were instructed to immediately vacate an area when an alarm is sounded (Tr. 2262-2263). When required "to enter the system" on these towers or expose the "internals" of the towers to the atmosphere, employees must obtain an entry permit. When entering the system an employee also must be a member of a 2-man work team and use a supplied air system. A stand-by rescue worker with a separate air source also must accompany the 2-man work team (Tr. 2265, 2301). When first testifying on direct examination, Compliance Officer Bakewell referenced only one emergency escape respirator which he felt suitable and capable of S:\1984\Processed\79-0570-E\00000000\07.tif " eliminating exposure to H2S on the towers - the Robert Shaw air capsule respirator (Tr. 41-43; Ex. C-2). This respirator is approved for respiratory protection during escape by the Natlonal Institute for Occupational Safety and Health (NIOSH.), but only for temperatures above 20*F (Tr. 297-298; Ex. C-2). Both Officer Bakewell and Mr. Quinlivan admitted that the Robert Shaw respirator is unsuitable for year round use at Conoco's Denver refinery because the temperature frequently falls below 20* in Denver (Tr. 297-298, 947-948). Later in the hearing, the Secretary recalled Officer Bakewell to testify on direct that following his initial testimony he called NIOSH and questioned a Mr. Terry, who said he was a chemist with the NIOSH respirator section, concerning the availability of escape respirators that will work below 20*F. In response, Mr. Terry read several inventory numbers from a list; Mr. Bakewell wrote down two of these numbers (Tr. 1180-1181). Officer Bakewell is unacquainted with the referenced respirators and does not know if these respirators have any use limitations imposed by NIOSH (Tr. 1188- 1189). Officer Bakewell testified that emergency escape respirators should be placed on the platforms of the two towers (Tr. 301). He also testified, however, that the preferable method would be for an employee to carry a respirator on his belt (Tr. 240-241, 303-304). He admitted he does not know how long it takes to don the Robert Shaw respirator or whether an employee exposed to an H2S release could put it on before being overcome (Tr. 240-241, 302). In Officer Bakewell's opinion an escape respirator "is no good" for a "massive leak" of H2S. only for a "minimal leak" (Tr. 243). Discussion This standard requires an employer to provide respirators "when such equip ment is necessary to protect the health of the employee." Conoco first argues that to prove that respiratory protection was "necessary" the Secretary must establish S:\1984\Processed\79-0570-E\00000000\08.tif " that the permissible limits for H2S, which are prescribed in 29 C.F.R. §1910.1000 (Table Z-2), were exceeded on the two towers in question Both Conoco's interpretation of the standard and its view of the Secretary's burden of proof under the standard are too narrow and completely ignore that §1910.134 contemplates the use of respirators when "dangerous atmospheres.....might be encountered in normal operation..." (emphasis supplied; See §1910.134(e)(3)). Clearly, the requirements of §1910.134(a)(2) are intended to apply when employee excessive exposure is foreseeable. Therefore, proof of actual exposure is unnecessary to establish a violation of this standard. The standard at issue is read to require employers to provide respirators whenever employees may be foreseeably exposed to a contaminant in excess of the limits provided in §1910.1000 (Table Z-2) or foreseeably exposed to an oxygen-deficient atmosphere. The Secretary must prove potential excessive exposure to H2S because actual exposure to H2S is permitted within certain prescribed limits. The evidence is insufficient to prove that Conoco employees, foreseeably exposed to an impermissible level of H2S on the two towers, were not provided with respirators by Conoco. The record does establish that Conoco employees entering the tower systems and working on lines and flanges containing H2S might encounter an H2S atmosphere in excess of permissible limits. But those employees were provided by Conoco with a supplied air system and stand-by rescue worker which eliminated the possibility of an exposure to an excessive hazardous concentration of H2S (Tr. 2264-2265, 2301). Conoco also recognized that accidental releases Of H2S on these towers were possible in the event of corrosion. This possibility of exposure, however, apparently I/ Section 1910.1000 (Table Z-2) provides for a maximum ceiling concentration of 20p.p.m. and a maximum exposure of 50p.p.m. in a single 10 minute period, provided no other exposure occurs. S:\1984\Processed\79-0570-E\00000000\09.tif " was eliminated by Conoco's inspection procedures to detect corrosion and leaks due to the corrosion (Tr. 300-301, 2255-2257, 2269). There was no evidence to the contrary. The Secretary's expert, Mr. Quinlivan, suggested that accidental releases of H2S in the towers could occur for "other reasons" besides corrosion, but he failed to either identify or substantiate these other possibilities. It also is unrefuted that no employee has ever been exposed to H2S on the two towers and that neither tower has ever experienced an accidental release of H2S. Conoco's safety supervisor, Mr. Bradley, did admit that the presence of H2S was possible on the first landing of the sour water stripper structure. No explanation of what would cause the presence of H2S at that location, however, was ever given, or whether employees working on the two towers might encounter an excessive concentration of H2S coming from that location which would require the use of emergency respirators. Unable to find that unprotected employees were foreseeably exposed to excessive concentrations of H2S during normal operations, no violation of this standard can be affirmed. An additional reason requires vacation of this item. The standard also is read to require an employer not only to provide respirators but to provide respi rators capable of protecting employees from dangerous atmospheres. For a violation of the standard to be established, therefore, it is necessary to prove that there are respirators "applicable and suitable for the purpose intended ....," and the burden of proving this element is on the Secretary. The Secretary's witnesses conceded that the referenced Robert Shaw respirator was unsuitable for year-round use in Denver, and Officer Bakewell's hearsay testimony that respirators are available and suitable for use below 20* temperatures, standing alone, is considered insufficient to establish this critical element. S:\1984\Processed\79-0570-E\00000000\10.tif " Citation No.1 Item 3 Introduction As amended by the complaint, Item 3 of Citation No. 1 charges Conoco with violation of Section 5(a)(1) of the Act "in that the employer did not provide adequate training and instruction to employees who were employed as operators of the new catalytic, polymerization unit in that during the 8-hour training session given, much of the valvework and piping was not in place on the unit causing employees to be unfamiliar with the full unit in case of an emergency." Several months prior to the anticipated start-up date of Conoco's new catpoly fractionation section. members of Conoco's management, including Phil Nelson, process superintendent, Greg Lorimar, operations superintendent, Leonard Brandt, heavy oil division supervisor, and Glenn Lepard, training supervisor, met for the purpose of designing a training course for operators of the new unit (Tr. 2057-2058). After reviewing the prior experience and training of the operators involved and the matters to be covered, the group jointly decided upon a one-day training course, consisting of a classroom session and "hands-on" field work (Tr. 1959-1961, 2057- 2058, 2074-2075). As an aid to the training course, a cat-poly operating manual, Exhibit C- 8, was prepared by Leonard Brandt and Jimmie Rhodes, senior design engineer (Tr. 1957). This manual was distributed to all participants approximately two weeks to a month prior to the scheduled training in July, 1978, and they were instructed to study and become familiar with its contents (Tr. 2064-2065, 2112). The Secretary's employee witnesses testified they read the manual prior to the training class. At the beginning of each class each employee was given a 10-question written quiz to test his familiarity with the operating manual (Tr. 2067). S:\1984\Processed\79-0570-E\00000000\11.tif " There were 4 seperate training sessions, July 10, 12, 14, and 17, 1978 (Tr. 2059). There were approximately 5 to 7 employees in each training session, and instruction was provided by Mr. Lepard, Conoco's training supervisor, and Mr. Rhodes, a senior process engineer and designer of the new cat-poly unit. The classroom training lasted approximately three hours and covered the subjects listed in the agenda of June 28, 1978 (Exhibit R-11): sections of the operating manual (Tr. 2071-2072), differences between the old and new cat-poly fractionation sections (Tr. 676) and the process flow and instrumentation diagrams in the manual (Tr. 2067). The training program did not take up the specific characteristics of the "cascade control system" (Tr. 2090-2091). After rest breaks, Mr. Rhodes would answer questions (Tr. 2068). In the afternoon session, the class participants were taken to the new unit for the purpose of familiarizing themselves with the equipment and to "field crace" the flow scheme of the unit (Tr. 2068). After three hours in the field the group would reconvene in the classroom and each employee was given a test (Tr. 2067, 2077- 2078). All operators passed the test with a mark of 70% or more except Ronny DeHerra (Tr. 2079-2080). Each class was told that the operators should continue studying the manual and field tracing the unit (Tr. 2084). Testifying in support of the Secretary's charge under this item was Officer Bakewell and Mr. HcAninch, along with five Conoco employees who attended one of the training sessions in July, 1978. Mr. Donald Kapus, a chief operator at the time of the start-up, had previously been a helper and operator in the old cat-poly unit (Tr. 511-512). In his opinion he did not feel that the training was adequate "in terms of the practical aspects" (Tr. 513). Mr. Kapus testified that the cascade control system was covered in the training session; according to him they were told "how it worked." He does not recall S:\1984\Processed\79-0570-E\00000000\12.tif " being told what to do if the cascade control system did not work (Tr. 556). Another chief operator on the heavy oil side, Mr. George Martel, testified that prior to the training class the students were given the cat-poly training manual and requested "to absorb the information." They also were given flow sheets and told to review them before the unit start-up (Tr. 374). In his view the field training was "difficult" because "a lot" of the piping and control systems were missing. He could not estimate "percentage wise" how much of the system was missing (Tr. 378). In his view the field training was "worthwhile" to "low experienced people" (Tr. 378). He believes it important for an operator to be familiar enough with the unit so that he can "react in emergencies" (Tr. 381). According to Mr. Martel, an operator must have a "pretty good idea of the flow and location of the specific control systems, in order "to handle emergencies" (Tr. 381-382). In order to control or eliminate a pin hole leak in vessel W-55 an operator would have to know the function and position of perhaps a dozen valves (Tr. 384). After he completed his eight hour training session, Mr. Martel did not "feel competent" to handle the new cat-poly unit (Tr. 385). He did not feel his training was adequate because he did not "have" the physical location of all of the piping and control systems involved and the basic knowledge of the systems internal action "in my own mind" (Tr. 385). Mr. Martel also testified that the cascading control system is more sophisticated than a "straight system" and was not adequately covered in the training system (Tr. 387). On cross-examination, Mr. Martel, testified that he had a chance to trace the pipes in the new cat-poly unit prior to start up (Tr. 458-459). He indicated that "management" told the operators to try to get familiar with the piping "as time allows" (Tr. 459). At another point, however, he could not recall whether they were told by a supervisor or whether it was simply a discussion among the operators (Tr. 466). S:\1984\Processed\79-0570-E\00000000\13.tif " Mr. David Judish, another participant in a July, 1978 training class, testified that after the training class, he did not feel that he was qualified to handle an emergency on the cat-poly unit (Tr. 602). He stated that he had let his supervisor, Leonard Brandt, know that he "wasn't really sure of myself" (Tr. 603-604). According to Mr. Judish, Mr. Brandt indicated that there would be no more training, that Mr. Judish had passed the test, and that Mr. Judish should be able to handle the situation (Tr. 604). Prior to start-up Mr. Judish went out to the unit and traced the lines of the new unit (Tr. 629-630). By the time of start-up Mr. Judish had "no problem" understanding how the unit operated (Tr. 630). On redirect, Mr. Judish testified that Mr. Lepard, the class instructor, was unable to answer certain questions. He does not remember what these questions were but he obtained satisfactory answers to those questions prior to start-up (Tr. 633). Mr. Michael Carey testified that he felt the training was adequate for him (Tr. 641-642, 664-666). He did not know, however, the location of all the lines at the time of the class (Tr. 641-642). According to Mr. Carey, approximately 90% of the lines and equipment was in place at the time of the training session (Tr. 666). Mr. Ronald D. Branson, an FCC operator, testified that he was present for only 4 hours of his training class in July, 1978 (Tr. 678). He does not "feel" his training was adequate (Tr. 678). He asked Leonard Brandt, the heavy oils supervisor, and Mr. Glenn Lepard for more training. They responded that they would "catch me later." But they never did (Tr. 680). He testified that he asked Mr. Brandt twice for extra training, once before the unit started up and once after it was started up (Tr. 680). According to Mr. Branson. Mr. Brandt responded the "same" both times (Tr. 680). After the start-up. he was left alone in charge of the control board of the new cat-poly unit on several occasions (Tr. 681, 694, 704-705, 707). S:\1984\Processed\79-0570-E\00000000\14.tif " he indicated that he told Mr. Brandt that he had been left alone on the control board when he asked for additional training the second time (Tr. 701, 707-708). He testi fied he did not "call" Mr. Brandt, however to tell him this or request additional train ing (Tr. 701). On cross-examination Mr. Branson testified that he could operate "the instruments inside" because "that is all I was trained to do" (Tr. 698). He further testified that he did not know enough to shut the unit down in the event of an emergency because he had "no idea" of the location of "a lot of the valves" (Tr. 690, 699". After the training class he went out to the new unit on a single occasion to try and trace the lines but he was immediately called back to the control room (Tr. 687). Mr. Brandt testified the major part of a shut down of the fractionation unit is done outside on the unit - shutting down valves, pumps - getting the gas out of the unit, depressurizing and steaming it out (Tr. 711). Mr. Brandt also testified that he told Mr. Lepard on the day of the class that he had not had enough training time. According to Mr. Brandt, Mr. Lepard responded not to worry about it or something to that effect, "[W]e are going to take a test" (Tr. 710). Officer Bakewell knows of no standard practice in the petroleum refinery industry for training operators for catalytic polymerization units (Tr. 309-311). His opinion that the operators training was inadequate was based solely on the opinions of the operators that their training was inadequate because they did not know the "physical characteristics" of the system (Tr. 250-251, 312). He conceded he really didn't analyze the training program" (Tr. 315). Officer Bakewell had no personal knowledge concerning the operation of the cat-poly unit nor any knowledge concerning the prior training experience of the employees taking their training course (Tr. 314-318). In Mr. McAninch's opinion Conoco could not have adequately trained the operators in "eight hours" to operate the cat-poly unit because the contents of the operations S:\1984\Processed\79-0570-E\00000000\15.tif " manual could not have been covered in eight hours (Tr. 1090, 1137). He retreated from his opinion concerning the inadequacy of this part of the program, however. when informed that the operators were provided with the training manuals 2 to 6 weeks before the class and then tested to determine whether they had read the manual (Tr. 1143-1144). In Mr. McAninch's view there are two general purposes to an operators' training program, safety and efficiency (Tr. 1136). He was unable to conclude that Conoco's training was inadequate from the standpoint of safety because "I don't know what the Conoco program consisted of" (Tr. 1136-1137). He further agreed that he was not in a position to say whether the Conoco training of the operators created a hazard of physical injury or death (Tr. 1137). Mr. Brandt, Conoco's heavy oil division supervisor, testified that as of July 7, 1978, 95% of the fractionation system of the cat-poly unit had been completed and installed. According to Mr. Brandt, the reactor's back pressure control valve, the reactor line to this valve, the compressuring line, the line from the splitter to the de-ethanizer, the line from butane to storage, and the lines from propane to storage and poly- gasoline to storage, were missing (Tr. 1965). In his view the only missing piece of equipment that an operator should know the location of was the reactor back pressure controller (Tr. 1966, 1969). According to Mr. Brandt the operators were assigned to the cat-poly unit approximately 6 days prior to the start-up and were involved in "steaming out" the lines. This process requires that the lines be traced out, bleeders closed, and the entire system be checked for leaks (Tr. 1971). Mr. Brandt stated that Mr. Branson complained in July, 1978 that he had not been present for the entire training session. Mr. Brandt told Mr. Branson at that time "we'd catch up with that training" when Mr. Branson was required to assume the position of a chief operator (Tr. 1973). S:\1984\Processed\79-0570-E\00000000\16.tif " Mr. Brandt agrees that one man should not be responsible for all the control panels (Tr. 1974). Mr. Brandt does not recall Mr. Branson making a second complaint about his training or informing him that Mr. Branson had been left in the control room alone (Tr. 1977). According to Mr. Lepard, the class instructor, part of the test given to the class at the end of the session was to fill in the process flow lines and the process control points on a diagram with the new equipment drawn in (Tr. 2077- 2078). According to Mr. Lepard, a control valve was not in place when the operators went into the field to trace the lines (Tr. 2107). Mr. Lepard testified that Mr. Branson never complained to him about the training course or claim that he did not know how to operate the unit (Tr. 2083). According to Mr. Lepard, Mr. Rhodes discussed the emergency procedure section of the operations manual (Tr. 2096-2097). Mr. Lepard testified that the manual did not explain the operation of cascade pressure control system and he does not recall whether Mr. Rhodes went into detail about this system (Tr. 2101). Mr. Lepard previously had not seen such a system (Tr. 2114). Mr. Kenard Taylor, Conoco's expert on training, described the old fractionation section of the cat-poly unit as follows: the product of the reactor went to the deethanizer, then to the de-propanizer, and then to the stabilizer. In the new fractionation section the product went to the stabilizer, then to the de- propanizer, and finally to the de-ethanizer (Tr. 2163). From the standpoint of the individual operator, the operating conditions, tower pressures and temperatures, and the process flow, were different between the old and the new fractionation sections (Tr. 2163). In his view, the training program on the new unit should cover the differences between the operating conditions and the physical equipment differences (Tr. 2164). S:\1984\Processed\79-0570-E\00000000\17.tif " Mr. Taylor was of the opinion that sufficient equipment was in place in July, 1978, to provide operators with a "meaningful and sufficient" physical inspection and tracing of the equipment (Tr. 2180-2181). In his view, approximately one third of the information in the manual was new to the operators, and all sections except section one should have been covered "in order to permit a safe operation of the new fractionating section" (Tr. 2189). Mr. Taylor testified that in shutting down the new fractionating section, the procedures would be the same; the valves, however, would be different (Tr. 2194). Mr. Taylor would have made two changes in the existing training course. He would have tried to have the training closer to the start-up line as an aid to retention, and he would have inserted a "process control section" which he considers "independently significant" to the training of an operator (Tr. 2202, 2208). Or, cross-examination Mr. Taylor testified that if any of the operators with operating responsibilities did not have exposure or experience with a cascade pressure control system, the cascade system should have been covered, and that if it was not covered "it could have created a problem within the unit" (Tr. 2210). When major oil companies require employees do their own field training, there's a "follow-up" to that training by "supervision" to determine if the employees have done their field training according to Mr. Taylor (Tr. 2212). It was his understanding that Conoco did not formally follow up to determine after the training session whether the operators had traced the lines (Tr. 2213). In his view, however, observations by supervisors would have teen sufficient to determine whether operators were capable (Tr. 2213-2214). According to Mr. Taylor, it is customary for industry to take further action to determine whether additional training is necessary when an operator expresses a lack of confidence in his training (Tr. 2214-2215). Mr. Taylor is of the opinion that Conoco's training program was adequate, did not create a recognized hazard of physical injury or death, and that it was safe for S:\1984\Processed\79-0570-E\00000000\18.tif " Conoco te start the new fractionation section and operate it with the operators. (Tr. 2200-2201). Discussion By a preponderance of the evidence the record establishes a violation of Section 5(a)(1) of the Act resulting from Conoco's inadequate training of Mr. Ronald Branson. Although not a model of clarity, the complaint identifies the alleged recognized hazard as Conoco's failure during the 8-hour training session to adequately familiarize operators of the new cat-poly fractionation unit with the valvework and piping on the unit so as to enable the operators to deal with an "emergency." The hearing record, however, reflects that the parties tried a much broader theory of recognized hazard: whether any of the operators assigned operating duties on the cat-poly unit were inadequately trained by Conoco so as to create a recognized hazard likely to cause death or serious physical injury. Because both parties plainly tried this alternative theory of recognized hazard, it has been treated in all respects as if it had been raised by the pleadings. 2/ The Secretary's conclusion that Conoco's training was inadequate was based in part on Officer Bakewell's and Mr. McAninch's opinions that the training manual could not be learned in a one-day training course. After learning, however, that the training manuals were distributed and read by the operators and then tested to determine whether they had read the manuals, Officer Bakewell admitted that the length of the classroom training was not the basis for the citation (Tr. 251). Mr. McAninch indicated that if the operators were provided with the manuals prior to the class, tested to determine their knowledge of their contents, and achieved at least a passing grade 2/ Both parties introduced evidence on this issue, and Conoco's brief, at pp.64 and 65 acknowledged that the question presented is far broader than the issue presented by the complaint. S:\1984\Processed\79-0570-E\00000000\19.tif " of 70 percent, a one-day training course was "adequate" (Tr. 1143-1144). Conoco's expert witness, Mr. Kenard Taylor, also concluded that Conoco's one-day training format was adequate under the circumstances (Tr. 2149- 2150, -2152, 2190, 2195), and above average based on his knowledge of industry practices and customs (Tr. 2195, 2203-2204). on this record, it cannot be found that the eight hour training session was inadequate due to its length. Another of the Secretary's apparent objections to Conoco's training program involves Conoco's failure to adequately cover the subject of the "cascade pressure control instrumentation." It was established that Conoco's training course barely touched on the subject of the cascade pressure control system on the new cat-poly unit. Further, Mr. Taylor concedes that this subject is "independently significant to the training of an operator" (Tr. 2202, 2208). In his view the training should have covered this subject unless those operators with operating responsibilities on the cat-poly unit had previous exposure or experience with a cascade pressure control system. He expressed the opinion that lack of training in this area "could have created a problem within the unit" (Tr. 2210). The evidence, however, failed to show that any of the operator-students (with the exception of the trainees who would not be operating the units) were unfamiliar with or unable to operate the cat-poly cascade pressure control system. Indeed, the record establishes that the operators who would be assigned operating responsibilities had either worked closely with or directly on the vapor recovery unit (VRU) of the FCC unit which used a similar cascade pressure control system (Tr. 2126). None of the Secretary's operator- witnesses stated that he was unfamiliar with the operation of the cascade system at the time of start-up. Even Mr. Aranson testified that he could operate the control room instruments of the cat-poly unit because he had been "trained to do it" (Tr. 698). S:\1984\Processed\79-0570-E\00000000\20.tif " On this record, this aspect of Conoco's training program cannot serve as a reason for concluding that the training program was inadequate. Both in the pleadings and at the hearing, the Secretary's primary complaint about Conoco's training program was its failure to adequately instruct the operators in the location of the process lines and equipment. In July of 1978, all of the equipment and lines for the new cat-poly unit were not in place. Witnesses variously estimated that the cat-poly unit was 90 to 992 completed. Approximately three hours of the training session was devoted to field work for the purpose of familiarizing the students with the equipment and to permit them to trace the lines (Tr. 2068). The first training class took place on July 10, 1978 and Mr. Branson was a member of this class. According to the instructor, Mr. Lepard, the students in that class were impeded in their examination of the unit because "construction was going on" (Tr. 2084). Due to a power interruption which occurred during the field portion of the July 10th training session, the class members assisted in securing the process units and restarting them (Tr. 677-678, 2081). Mr. Lepard estimates that training session was cut short by "somewhat less than two hours" (Tr. 2095-2096); Mr. Branson estimates that he missed four hours from class (Tr. 686). During the period between the one-day training course and the start-up, Mr. Branson did not trace the lines (Tr. 687). At the time of the start-up of the new cat-poly unit, Mr. Branson believed that he did not know enough to be able to shut the unit down in event of "an emergency." He did nit know the location of the valves necessary to shut down the unit (Tr. 690). Conoco's expert, Mr. Taylor, recognizes that field work is "very important" (Tr. 2159). He also recognizes that training of these operators should have covered the "physical equipment differences" between the old and new cat-poly unit (Tr. 2164). S:\1984\Processed\79-0570-E\00000000\21.tif " Mr. Taylor states that an operator has to spend time in the field reviewing the flow, "looking at it, reviewing it again, finding the actual pieces out there" (Tr. 2191-2192). He indicated that the operators in this case would need to know where the valves, equipment and instruments necessary to shut down the unit were located (Tr. 2194-2195). Mr. Taylor suggests that Mr. Branson had the "availability" on the unit to follow that field work (Tr. 2182). Unfortunately, however, he never completed this necessary field work. Although Mr. Branson apparently was capable of operating the board to initiate an emergency shut-down, he felt unable to locate the valves necessary to complete the shut down procedure (Tr. 2186). Mr. Taylor concedes it would be imprudent to ignore an employee claim that he has not been adequately trained, but Mr. Taylor believes that "an unsafe condition" was not created in the instant case because Mr. Branson was not directly involved (Tr. 2184). This belief clearly implies that if Mr. Branson was directly involved, Mr. Taylor's opinion concerning the creation of an unsafe condition would be different. The evidence, of course, shows that Mr. Branson was directly involved with the cat-poly unit on a number of occasions. According to Mr. Taylor, if an operator did not know "the new unit versus the old unit," he would need additional training (Tr. 2216). Also, Mr. Taylor agrees that a critical factor in how an operator reacts in an emergency situation is the confidence an operator has in his own ability to operate (Tr. 2215). Mr. Branson lacked this confidence and Conoco was aware of this situation. In sum, it is found that Mr. Branson was inadequately instructed by Conoco in the location of valves and equipment necessary to accomplish an emergency shutdown of the fractionation unit. Conoco was aware that Mr. Branson lacked confidence in his ability to operate the unit, yet took no further action to provide additional training or take steps to assure that Mr. Branson would not be assigned operational S:\1984\Processed\79-0570-E\00000000\22.tif " responsibilities for the cat-poly unit before receiving this additional training. The record further establishes that Mr. Branson's lack of training created a life-threatening hazard. Mr. Taylor testified that an operator's field training and his knowlege of the location of valves to accomplish the shut-down procedures were "very important." Mr. Taylor also clearly indicated that an operator should know all sections of the training manual (including the emergency shut-down procedures section) "in order to permit a safe operation of the new fractionation section (emphasis supplied; Tr. 2189). The emergency procedures section of the training manual provides for emergency shut-down procedures in case of fire and explosion and requires, among other things, isolation or bypass of the unit section in which the emergency has occurred, and then a shut-down in accordance with normal prescribed procedures (Ex. C-8, p.4.). From these requirements, it is reasonable to infer that the inability of an operator to follow shut-down procedures during a fire and explosion could exacerbate the fire and explosion conditions, which in turn would likeIy lead to death or serious injury. Furthermore, Mr. Branson's opinion that his not knowing which valves to shut off "could cost lives" went unchallenged and unrefuted (Tr. 699). Another question is whether Mr. Branson's inability to accomplish emergency shut-down procedures due to his lack of field training was a "recognized" hazard. Conoco's training manual and Mr. Taylor's testimony establishes that both the petroleum refinery industry and Conoco recognize the need for adequately training operators in emergency shut-down procedures. Further, the record establishes recognition by the petroleum refinery industry of the need to determine whether an operator has completed his assigned field work. That was not done by Conoco in this case even though Conoco was on notice that Mr. Branson believed he needed additional field training. The feasible abatement methods for this violation are obvious. If Conoco had followed industry practice and provided Mr. Branson with the field training S:\1984\Processed\79-0570-E\00000000\23.tif " necessary to enable him to accomplish the emergency shut- down procedures, the "recognized" hazard would have been eliminated. The Secretary proved, however, that only one Conoco employee was inadequately trained at the time of start-up. A substantial penalty therefore, is unwarranted. A penalty of $100.00 will be assessed. Citation No. 1,Item 5 Item 5 of Citation No. 1 alleges a violation of section 5(a)(1) of the Act "in that" certain designated heaters and boilers with "open flames and/or hot surfaces were not protected in a manner to prevent the ignition of light hydrocarbon vapors and/or gases in the event of accidental rupture or breakdown of the system." This charge is similar, but not identical, to the Secretary's charge in Asamera, Oil Co., Inc., 9 BNA OSHC 1426, 1981 CCH OSHD (p) 25,131 (No. 79-0949 and 79-1756, 1980)..3/ In Asamera, the complaint identified the existing recognized hazard as Asamera's failure to separate or protect in any manner certain heaters and boilers to prevent ignition of hydrocarbons. In this case, separation is not mentioned. In both cases, however, the Secretary's theory of liability was identical; the employers' failure to separate heaters at least 50 feet and boilers at least 100 feet from process equipment containing hydrocarbons, or. protect these heaters and/or boilers so as to prevent ignition of hydrocarbon releases. Summary of Evidence To establish that Conoco's separation and protection practices constituted a recognized hazard, the Secretary introduced the testimony of Compliance Officer Bakewell and two expert witnesses, Mr. Quinlivan and Mr. McAninch. Officer Bakewell testified about his various distance measurements taken from the cited heaters and Boiler B-4 to valves, flanges or pipes of other pieces of process equipment. All of these measurements were less than 50 feet (Ex. C-9). Specific __________________ 3/ Asamera was heard after the hearing in this case. The decision in Asamera, however, was issued while this case was pending before the Commission on interlocutory appeal. S:\1984\Processed\79-0570-E\00000000\24.tif " measurements were not taken with respect to Boilers B-8 and B-6 because in Officer Bakewell's words, "there were lines everywhere in the area" (Tr. 269). In this regard, he also testified that "many" of the lines in the "Sulfur plant" contained hydrocarbon vapors and hydrogen sulfide "which is also flammable" (Tr. 102, Ex. C-12). Officer Bakewell was unable to identify W-38 and conceded he did not know what materials were contained in W-55 and W-45. He also admitted that he did not determine the materials, the temperature, or pressure of these materials behind the various valves or flanges which he used as measuring points. He assumed that hydrocarbone were present, but agreed that the material behind the measuring points could have been steam (Tr. 267-269). Officer Bakewell also testified that he did not determine the temperatures of the various heater or boiler surfaces (Tr. 270). Mr. Quinlivan testified that the petroleum refinery industry generally recognizes that a "safe" distance for separating boilers, heaters, and furnaces from process units is 100 feet (Tr. 869). He then identified potential sources of hydrocarbon vapor. For example, he testified that vacuum tower W-36 was a source for a hydrocarbon vapor release (Tr. 870-871). He identified W-45 as "a fractionator or a catalyst storage or some such hydrocarbon-containing vessel" (Tr. 874). According to Mr. Quinlivan, any fractionating unit at Conoco's refinery or its associated equipment, exchangers, pumps, and piping were all potential sources of vapor release (Tr. 871). Mr. Quinlivan also identified the "Sulfur Plant" as a potential source for the release of "flammable vapors - Hydrocarbon, H2S would be one" (Tr. 875). Testifying more generally, Mr. Quinlivan testified that the proximity of H-27 to W-36 (21 feet. 9 inches) and the exchangers X-176 and X-139 were "definite hazards." He also contended that the "same unsafe condition; too close: existed with respect to the proximity of H-4 and H-22 to the FCC converter (Tr. 873). The import of Mr. Quinlivan's entire testimony was that the Iocations of Conoco's S:\1984\Processed\79-0570-E\00000000\25.tif " cited heaters and boilers with respect to the process equipment were "hazardous... unsafe .... too close" (Tr. 870-875, 877-878). In his view, sufficient seperation of heaters and boilers from process equipment is necessary to "give the vapor cloud time to disperse and dilute below the combustible range." He provided no support, however, for his thesis that greater distance will result in the dispersal of a vapor cloud. In this regard, he testified there wasn't any empirical data or learned treatise support for his thesis (Tr. 924). The Secretary's other expert, Hr. McAninch, based his opinion concerning the separation of heaters and boilers upon his 28 years of employment and experience with American Oil Company (Amoco). He testified that Conoco's practice when building new refinery units was to locate furnaces on the upwind side of process equipment and no closer than 50 feet from the "first major piece of equipment" (Tr. 1058). According to Mr. McAninch, Amoco treated boilers as a separate installation and they were "segregated" from process units (7:r- 1059). There is no indication whether or not Amoco separated boilers from process units by a prescribed distance. Mr. McAninch also indicated that two Amoco refineries had boilers and heaters within 50 feet of process units (Tr. 1115-1116). In Mr. HcAninch's view, the purpose of Amoco's spacing practice was to give operators more time to take corrective action to disperse hydrocarbon releases (Tr. 1113-1114). At another point in his testimony, Mr. McAninch indicated that the purpose of locating heaters at least 50 feet upwind from a process unit was to separate the sources of ignition from the sources of hydrocarbons (Tr. 2526). Mr. McAninch identified potential sources of hydrocarbon vapor and testified that the heaters were "too close" (Tr. 1062-1068). In his opinion none of the cited boilers and heaters were "properly located in terms of safe spacing standards" (Tr. 1068). S:\1984\Processed\79-0570-E\00000000\26.tif " Both the Secretary's expert witnesses referred in their testimony to the document published by Industrial Risk Insurers entitled "General Recommendations for Spacing in Refineries, Petrochemical Plants, Terminals, Oil Pump Stations and Offshore Properties" (hereinafter called "IRI Recommendations"; Ex. C-16). Mr. Quinlivan testified that the IRI Recommendations are considered "authoritative in the industry" (Tr. 868). Hr. McAninch stated that Amoco's design philosophy with respect to the separation of boilers and heaters was "consistent" with the IRI Recommendations (Tr. 1061-1062). Neither witness stated, however, that the purpose of the spacing recommendations of the IRI document was to prevent the ignition of hydrocarbon vapors by heaters and boilers (Tr. 925, 1116). The recommended minimum separation distance of boilers from process units in IRI Recommendations is 100 feet; the recommended minimum separation distance of heaters from process units is 50 feet (Ex. C-16, p.4). Discussion The record clearly establishes that the potential for fire and explosion resulting from the ignition of hydrocarbon vapor releases by heaters and boilers existed at Conoco's refinery. That the petroleum refinery industry, including Conoco, recognizes the existence of this hazard, which is likely to cause death or serious physical harm, also was thoroughly established. Even the testimony of Conoco's expert witness, Mr. Doyle, compels this finding (Tr. 2321, 2364-2366; Ex. R-19). Insufficient evidence, however, was introduced to establish that the petroleum refinery industry or Conoco recognizes that the failure to adhere to a 50foot separation rule for heaters and process units and a 100-foot separation rule for boilers and process units is hazardous. Mr. McAninch's opinions are based on the practices of a single company, Amoco. Certainly the separation policy of a single company does not establish the practice of an entire industry. Moreover, Amoco applies its 50-foot spacing policy for heaters S:\1984\Processed\79-0570-E\00000000\27.tif " when constructing new units and has not applied a mininum 50-foot spacing policy to all heaters or boilers in existing refineries. Mr. Quinlivan, on the other hand, testified on direct examination that the general practice of the petroleum refining industry is to apply a 100-foot spacing rule for both heaters and boilers. Yet, when testifying about the separation practices of the Amoco refinery in Casper, Wyoming and the Gary operating refinery at Fruita, Colorado, the only refineries specifically named by him, he confirmed that both refineries followed a 50-foot separation rule for heaters, not a 100-foot rule (Tr. 925).4/ Clearly, the testimony of Mr. McAninch and Mr. Quinlivan is insufficient to permit a finding that the petroleum refinery adheres to, or recognizes the need to follow, either Mr. Quinlivan's 100-foot rule for separating heaters and boilers from process units or Amoco's minimum 50-foot rule for separating heaters. Both of the Secretary's expert witnesses referred to the IRI Recommendations in their testimony, and the Secretary relies heavily on this document to establish the spacing practices of the petroleum refinery industry. Mr. Quinlivan testified that the petroleum refinery industry recognizes the IRI Recommendations as "authoritative." Mr. McAninch simply indicated that Conoco's policy is "consistent" with the IRI Recommendations with respect to new installations. Neither witness named any refineries that have specifically recognized, adopted, and adhered to IRI Recommendations. Moreover, the preface of Exhibit C-16 indicates that these general recommendations for spacing are presented as guidance for "new construction projects" and "additions" to existing facilities. On this record the IRI 4/ Conoco's brief references the finding in Asamera Oil Co., Inc., supra, that none of the heaters at Amoco's Casper refinery or Gary's Fruita refinery adhered to a 50-foot or 100-foot spacing requirement. Although tempted, this judge has given no consideration to the findings or expert testimony in Asamera in deciding this case. S:\1984\Processed\79-0570-E\00000000\28.tif " Recommendations cannot be found to represent the spacing practice of the petroleum refinery Industry with respect to existing refineries. Because the record fails to prove that the petroleum refinery industry 'has a consensus spacing practice for existing refineries, the Secretary cannot rely upon industry practice to establish that Conoco's manner of separation constitutes a recognized hazard. The Secretary also did not introduce any evidence to establish that Conoco actually recognizes its separation practice as hazardous. However, the decision in National Realty and Construction Co., Inc. v. O.S.H.R.C., 489 F.2d 1257, 1265 n.32 (1973) indicates that the existence of a recognized hazard also can be shown by proof that an employer's practice is unacceptable when judged by a common standard recognized by safety experts familiar with the industry. The Secretary's two expert witnesses agree that the proximity of the cited Conoco heaters and boilers to potential sources of hydrocarbon releases was hazardous. They disagree, however, on what constitutes a safe separation distance. Although his testimony is far from clear on this point, Mr. Quinlivan apparently adheres to a personal 100-foot separation rule for both heaters and boilers. Mr. McAninch, in contract to Mr. Quinlivan, adopts Amoco's 50-foot rule for the separation of heaters from process units. 5/ Mr. Doyle, Conoco's expert, has still another opinion: that the 50-foot and 100-foot spacing recommendations contained in the IRI Recommendations for heaters and boilers are arbitrary and not an effective means of preventing ignition of hydrocarbon vapor releases (Tr. 2333-2334, 2337-2338). Certainly, these three divergent opinions on this critical issue provide no basis for finding that safety experts agree on what separation distance is necessary to prevent boilers and heaters from igniting hydrocarbon vapors coming from process _____________________________________________ 5/ This record fails to indicate whether or not Mr. McAninch has a prescribed distance rule for the separation of boilers from process units. S:\1984\Processed\79-0570-E\00000000\29.tif " units, or that they even agree that separation is an effective means of preventing such ignition. Instead, these opinions support a finding that safety experts familiar with the petroleum refinery industry do not hold a standard of common knowledge concerning the separation of heaters and boilers from process units to prevent ignition. Unless a common standard is established by the record, and the evidence proves that Conoco's separation practice is contrary to that standard, it is impossible to find that Conoco's separation practice is a recognized hazard under section 5(a) (1) of the Act. The Secretary also unsuccessfully attempted to prove that Conoco's failure to use certain protection methods or devices to prevent boiler and heater ignition of hydrocarbon vapors from process units constituted a recognized hazard. Mr. Quinlivan testified that "there are many schemes around" the petroleum refinery industry to prevent the possibility of heaters and boilers igniting hydrocarbon vapor releases. He specified fire walls, air curtains, water curtains, flame and combustible gas detectors, and fire protection systems which would blanket areas and isolate ignition sources from process vapor sources (Tr. 881). In connection with the water curtain system he also recommends an automated shut-down system (Tr. 884). He testified that his preferred method for abating "the hazard of the proximity of process units to heaters and boilers" in Conoco's refinery would be to " redesign or properly space" (Tr. 884-885). In his view, the fire wall and air curtain would not have been practical for installation at Conoco's refinery due to space limitations (Tr. 929). Mr. Quinlivan also agrees that it would be impractical to adhere to his spacing recommendations for heaters and boilers at Conoco's refineries (Tr. 931). Without explaining why a C02 or water curtain system also would not have been practical due to space limitations, Mr. Quinlivan recommended the installation of a water curtain or a CO2 curtain system at Conoco's refinery. S:\1984\Processed\79-0570-E\00000000\30.tif " Mr. Quinlivan estimated, "off the top of my head," that the "Iikely cost" of a water curtain system combined with flame and combustible gas detectors to protect heaters from. igniting, vapors as between $100,000 and $150,000 (Tr. 586, 933-934). He testified that he has seen a water curtain and combustible gas detector system at Amoco's refinery in Casper and Gary's refinery in Fruita, Colorado, but never answered the question of "how they worked" or indicated whether they were used around heaters and boilers (Tr. 886-887). Mr. Quinlivan has never seen a CO2 curtain system installed in a refinery (Tr. 931). The Secretary's evidence falls short of establishing that the petroleum refinery industry generally uses water or CO2 curtains, or any other of the Secretary's recommended protective devices, to protect heaters or boilers from ignition of hydrocarbon vapors, or that the industry or Conoco recognize that the failure to use such systems or devices is hazardous. Also, the same evidence is insufficient to establish that safety experts familiar with the petroleum refinery industry recognize that Conoco's failure to use the Secretary's recommended devices is hazardous. On the entire record, the existence of a recognized hazard with respect to Conoco's manner of separating or protecting its heaters and boilers to prevent ignition of hydrocarbons from process units was not proven. The Secretary's charges, both here and in Asamera, are apparent attempts by the Secretary to establish a separation performance standard for the entire petroleum refinery industry through selective use of the general duty clause and ad hoc adjudication rather than employing the standard-making machinery provided in the Act. It is suggested that if the Secretary is persuaded that 50-foot and 100-foot spacing rules, or use of certain protection devises, are essential for employee safety, his proper course is to use the standard-making procedures of the Act. As was pointed out by the Court in B&B Insulation, Inc. v. OSAHRC and Marshall, 583 F.2d S:\1984\Processed\79-0570-E\00000000\31.tif " 1364 (5th Cir. 1978): "The use of standard-making procedures assures that not only would employers be apprised of the conduct required of them and responsibility for upgrading the safety of the industry would be borne equally by all its members, but the resulting standard would benefit from import of the industry's experts, both employer and employee, cost and technology obstacles could be weighed and more interested parties can articulate in the process." Citation No. 1, Items 11(b-d) Citation No._2,_Item 2( a) As amended, Citation 1. Items 11(b-d) charges Conoco with violating several sections of the 1971 National Electrical Code, NFPA No. 70-1971 (NEC), which have been incorporated by reference by standard 29 C.F.R. {sec}1910.309(a). The cited sections of 1971 NEC Article 501 all require that "approved" or "explosion-proof" electrical equipment be provided in "hazardous locations," as defined by Article 500 of the 1971 NEC. The Sscretary alleges that Conoco failed to provide "approved" electrical equipment, or in lieu thereof, positive pressure ventilation, at six locations in its Denver refinery. 6/ 1. The control room and fire building complex; 2. The water pumphouse; 3. Electrical equipment west of the old cat-poly unit (designated at the hearing as "west electrical disconnects"); 4. Electrical equipment south of the old cat-poly unit (designated at the hearing as "P+M 322-11"); 5. Electrical substation No. 2, and; 6. Electrical substation No. 1. _____________________________________________ 6/ Conoco withdrew its Notice of Contest with respect to a seventh location, Item 110(b) of Citation No. 1, in a stipulated Settlement Proposal, Ex. J-1 (Tr. 4-5). S:\1984\Processed\79-0570-E\00000000\32.tif " Item. 2a of Citation 2, as amended, additionally alleges that Conoco "willfully" violated Article 501 of the 1971 NEC with respect to Conoco's control room. The issue is whether the cited locations are "hazardous locations," and more particularly, "Class 1, Division 2" locations within the meaning of Article 500-4(b) of the 1971 NEC. Article 500-4(b) provides: "500-4 Class I Locations. Class 1 locations are those in which flammable gases may be present in the air in quantities sufficient to produce explosive or ignitable mixtures. Class I locations shall include the following. (b) Class I, Division 2. Locations (1) in which volatile flammable liquids or flammable gases are handled, processed or used, but in which the hazardous liquids, vapors or gases will normally be confined within closed containers or closed systems from which they can escape only in cases of accidental rupture or breakdown of such containers or systems or in case of abnormal operation of equipment..." With the exception of the location of substation No. 1, it is Conoco's position that none of the cited locations were Class I, Division 2 locations, and as a result, "approved" electrical equipment was not required. [[7/]] Summary of Evidence In support of these charges of violation of the 1971 NEC, the Secretary introduced the testimony of Officer Bakewell, and his expert, Mr. McAninch. Officer Bakewell's testimony that electrical equipment "not approved" for Class I, Division 2 was installed in the six cited locations went unrefuted. The accuracy of his measurements from the unapproved electrical equipment to certain other pieces of equipment also went unchallenged, except for his measurement from the P/B dryer S:\1984\Processed\79-0570-E\00000000\33.tif " to P-4 332-11. Officer Bakewell measured this distance as 45 feet (Ex. C-9). Mr. Brooks testified that the distance was 78 feet 7 inches (Tr. 1254). This measurement , however, was made by Mr. Greg Ricks,. not Mr. Brooks. According to Officer Bakewell, every measurement was taken from a "potential source of hydrocarbons" such as flanges, valves, pumps and pipes (Tr. 128-129, 131). On rebuttal he also testified that he "took measurements from areas where flammable vapors were housed" in a closed system (Tr. 146). All of the measurements were less than 100 feet (Tr. 106). In his opinion all of these 6 locations were in Class I, Division 2 areas because "under abnormal circumstances" hazardous vapors could be released (Tr. 146-147). In addition to relying on the NEC when issuing the citation, Officer Bakewell relied on document AP1 RP500A entitled "Recommended Practice for Classification of Areas for Electrical Installation in Petroleum Refineries" (Ex. C-32; hereinafter called "RP 500A"). Officer Bakewell interprets RP 500A as extending Class I, Division 2 locations out 50 feet from a potential vapor source; "Where large releases of volatile products may occur," however, Class 1. Division 2 locations extend 100 feet from the potential vapor source (Tr. 151, 157). According to Officer Bakewell, the NEC provides an employer with the option of either installing "approved" electrical equipment in a Class 1, Division 2 area or enclosing the electrical equipment and positively pressurizing the enclosure area (Tr. 160). On cross-examination, Officer Bakewell stated he was unable to testify as to the pressure, temperature, or quantity of the material behind the points to which he measured (Tr. 277). His measurement of 27 feet from the No. 2 transformer electrical substation to the corner of the "FCC Unit" was not taken to a valve or flange but rather to "a group of overhead lines" for the 48-foot measurement (Tr. 278). On S:\1984\Processed\79-0570-E\00000000\34.tif " rebuttal, he testified that with respect to these locations he measured to horizon tal overhead lines, not to a vertical pipe, and to a "pump" (Tr. 2535). Mr. McAninch testified that in classifying Class I, Division 1 and 2 areas he relied on the National Electrical Code "Chapter 5 Special Occupancies, ARTICLE 500 - HAZARDOUS LOCATIONS" (Ex. C-51). According ti Mr. McAninch, Amoco also used RP 500A (Ex. C-32) as "minimum guidelines." He considers that RP 500A is an interpretation of the NEC by the API and serves as a "guide" to the industry (Tr. 1073, 1121). Amoco as a "general rule" used a 50-foot standard unless the electrical equipment was close to "huge quantities" of hydrocarbons (Tr. 1119). In his experience, Amoco never used a 100-foot area classification for Class I, Division 2 locations (Tr. 1123-1124). In his view, all control rooms should be pressurized. He concedes, however, that the NEC does not require control rooms to be pressurized (Tr. 1075-1077). Mr. McAninch testified that Conoco's cat-poly unit and "some" of the recovery section of the crude unit contained "copious quantities of heavier than air hydrocarbons" that would go directly on release to the control room due to the prevailing wind (Tr. 1077-1078). His opinion that the control room at Conoco is in a Class I, Division 2 location, as those terms are used in the NEC, is based on the fact that API 500A requires a 100 foot extended area "where hydrocarbons can be available at the two foot level" (Tr. 1081). He believes the below grade pumphouse is in a Division 2 area because the "separator" is immediately east of the pumphouse (Tr. 1082). With respect to P+M 32-11, Mr. McAninch believes it should have been classified Division 2 because of its 45-foot distance from the PB dryer (Tr. 1083-1084). He indicated, however, that if the PB dryer was located "down wind" he would be "more lenient" (Tr. 1080). S:\1984\Processed\79-0570-E\00000000\35.tif " If the No. 2 Transformer Electrical Substation was located only some 27 feet west of the FCC, he would consider the substation as a Class I, Division 2 location (Tr. 1084). He further indicated that the "west electrical disconnects" also were in a Division 2 area (Tr. 1085). Conoco's expert, Mr. Robert P. Howell, was employed by Standard Oil of California for 40 years (Tr. 1364). A member of the API subcommittee on electrical equipment from 1953 to 1969, he considers himself "a principal author" of RP 500A (Tr. 1367-1368). He also was a panel member on the National Fire Protection Association (NFPA) "code making" panel at the time the NFPA issued the 1971 NEC (Tr. 1368). According to Mr. Howell. the NEC first defined Division 1 and Division 2 areas for electrical classification purposes in 1947. Due to the "vagueness" of those definitions, however, the petroleum industry found it impossible to classify areas in a consistent manner (Tr. 1375-1376). In response to this vagueness, the NFPA developed a number of "special occupancy" rules, setting forth "rigid, or mandatory" distances or dimensions for Division 1 and Division 2 area classifications for certain industries, such as service stations (Tr. 1376-1380; See e.g., NFPA, Flammable and Combustible Liquids Code, Exhibit R-4; see also, 1971 NEC, Articles 510-555). According to Mr. Howell, the NFPA never developed a special occupancy rule or a rigid definition of classified areas for petroleum refineries due to its recognition that each petroleum refinery is different in terms of process, size, terrain, and other factors, that go into determining how the NEC requirements for area classification should be applied (Tr. 1380-1381). Mr. Howell testified that RP 500A was prepared to serve as an interpretative guide of the NEC for the petroleum refinery industry (Tr. 1382, 1386). Mr. Howell stated that the recommended distances of 50 and 100 feet in Figure 1 of RP 500A are S:\1984\Processed\79-0570-E\00000000\36.tif " "extremely conservative" (Tr. 1390). He indicated that "many" of the major oil refineries do not automatically go to 50 feet in classifying a Division 2 area - they use "judgment" and do not use Figure 1 as a "standard" (Tr. 1389). He named Atlantic Richfield, Union Oil, Texaco, Chevron, and Amoco as companies using less than 50 feet in defining a Division 2 area (Tr. 1402). According to Mr. Howell, Gulf, Atlantic Richfield, Union, Soho, Texaco, Shell, Chevron, and Amoco, do not classify a Division 2 area beyond 50 feet (Tr. 1415). Also, the term "abnormal" in RP 500A is not to be equated with "catastrophic" (Tr. 1392). If the material is a volatile flammable liquid with a flash point of under a hundred, and if the pressure in the process equipment is over 500 psi, Mr. Howell, personally would classify the area extending out 50 feet from the process equipment as a Division 2 location (Tr. 1399-1400). If the pressure in the equipment. however, is less than 500 psi he would extend the classified area for this volatile flammable liquid out only 25 feet from the edge of the process equipment (Tr. 1400-1401). Moreover, if the process fluids have flash points considerably above 100*, he "might" suggest that the classification be limited to the "boundry limits" of the process area (Tr. 1400). His theory regarding temperature and psi in classifying a Division 2 location, however, is not mentioned in RP 500A. Mr. Howell considers pumps. closures on heat exchangers, flanges on vessels, and exchangers, as sources of vapor (Tr. 1404). According to Mr. Howell, the nearest source of vapor to the control room was X-302 and X-304, feed preheaters in the catpoly plant. The material contained in these preheaters were various propanes and butanes operating at 600 psi (Tr. 1420). For this equipment, he would go out 50 feet from the preheaters and designate everything within this area as Class I, Division 2. Because the electrical equipment in the control room was more than 50 feet from these preheaters, he believes the electrical equipment was not in a Class I, Division 2 location (Tr. 1420-1421). S:\1984\Processed\79-0570-E\00000000\37.tif " With respect to the general purpose disconnect switches and circuit breakers and the exhaust for the diesel engine in the below grade water purnphouse, Mr. Howell testified that the nearest source of vapor vas vessel the propane dryer (Tr. 1423). This dryer was operating at a pressure of "230, 250 pounds" (Tr. 1424). In his view the two pools directly adjacent to the pumphouse were not sources of vapor for classification purposes (Tr. 1424-1425). Mr. Howell further testified that the exhaust of the diesel engine is not covered by the NEC, which limits itself to electrical equipment (Tr. 1423). According to Mr. Howell, the NEC makes no mention of below-grade areas. However, API - RP 500A does suggest that, if a below-grade area is within a Class 1, Division 2 location, the below-grade area should be classified as Division I (Tr. 1425-1426). Mr. Howell believes that the nearest source of vapor to the electrical equipment in the area designated P+M 322-11 is the stripper reboiler, X-239, or the fractionating tower W-54. They both contain light hydrocarbons at moderate pressure, 200 to 300 pounds (Tr. 1426-1427). Mr. Howell, therefore, believes there was no need to classify P+M 322-11. The pump which he measured 53 feet from P+M 322-11 was a butane reflux pump and had not been in service for "some time" (Tr. 1435). He also would not have classified the #2 process electrical substation. According to Mr. Howell, the MEA unit contained no flammable liquids or volatile hydrocarbons, and the FCC unit nearest to the substation was handling "heavy hot pitchlike materials." Mr. Howell testified that the closest source of vapor were a pair of pumps, P-554 and 555, that handled naptha, via very high vapor pressure gasoline" (Tr. 1428). Pumps 558 and 559 are closer, according to Mr. Howell, but they contain a very hot slurry and definitely are not a source of vapor. Mr. McAninch disagrees with Mr. Howell's opinion in this regard, however, and testified the slurry would vaporize (Tr. 2532). Mr. Howell further testified that pumps P-556 and P-5571 handle S:\1984\Processed\79-0570-E\00000000\38.tif " a heavy cycle oil that would not be a source of volatile vapors (Tr. 1428-1429). Mr. Howell testified that W-17 is the main fractionating tower and was within 50 feet of # 2 transformer substation (Tr. 1428- 1429). Mr. Howell testified that the Secretary's 48-foot measurement on Exhibit C-9 was a flange in the MEA unit. Behind the flange, however, was "steam" (Tr. 1434). The nearest vapor source to the west electrical disconnects and circuit breakers, according to Hr. Howell, was D-66; the propane dryer. The distance was "such" that there would be no reason to classify the west electrical area (Tr. 1432). Mr. Lynen Brooks, a Conoco senior electrical engineer, testified concerning Conoco's classification procedures. According to Mr. Brooks, a Class I, Division 2 location is an area where the process fluids would be contained within closed systems and when releases would occur under an abnormal situation such as a "breakdown or failure of a piece of equipment (Tr. 1226). Conoco uses Conoco's Central Engineering Department Standard 12 as the basis for its classification of Division 1 and Division 2 areas (Tr. 1230; Ex. R-1). Conoco also recognizes that RP 500A is a reliable authority for the purpose of electrical classification in the petroleum refinery industry (Tr. 1234-1235). Mr. Brooks believes the primary purpose of RP 500A was to permit the petroleum refinery industry to determine the extent of Division 1 and Division 2 areas (Tr. 1236). In defining and classifying a Class I, Division 2 location, Conoco measures out 50 feet from sources "that are most likely to cause a problem in a refinery." According to Mr. Brooks, these sources are "rotating pumps and control valves with stems" (Tr. 1205). Conoco relies on Conoco's standard 12 and API RP 500A for distances and upon the NEC "in terms of the classification of the areas" (Tr. 1246). Conoco considers the quantity of the hazardous material that might escape in case of an accident, the total area involved, and the record of the industry and Conoco's with respect S:\1984\Processed\79-0570-E\00000000\39.tif " to explosion or fires (Tr. 1246-1247). To Mr. Brooks' knowledge, no fire or explosion at Conoco's refinery was due to electrical ignition (Tr. 1247). On cross-examination, however, he conceded, that he did not know the cause of some of the fires that had occurred at Conoco's refinery (Tr. 1268-1270). Mr. Brooks indicated that he would classify a Division 2 location less than 50 feet if he wanted to classify around a piece of equipment that was handling "fairly heavy substances" which would give off "very little vapor" if released. He also might classify beyond 50 feet "in a case of a large storage area" where "large quantities" might be released (Tr. 1248). Mr. Brooks does not believe that the control room was in a Class 1, Division 2 location (Tr. 1250). He testified that the closest sources of vapor to the control room were more than 50 feet distant (Tr. 1251). He believes that the material in the MEA unit, was "only water or some other types of material that would not constitute a hazard or would not constitute a source of vapor" (Tr. 1251-1252). He measured 76 feet 6 inches from the window of below-grade pumphouse to D-66, the propane dryer (Tr. 1254). The nearest source of vapor, the exchanger X-130, was farther away from the pumphouse than D-66 (Tr. 1255). Mr. Brooks does not consider D-66 to be a source of flammable vapor (Tr. 1284). The nearest source of vapor to P+M 322-11, according to Mr. Brooks, was the control valve near P-496, and it is over 50 feet away (Tr. 1260). Mr. Brooks believes that the nearest source of vapor to No. 2 Transformer substation is the reflux pump P-554 and further than 50 feet away (Tr. 1260-1261). Mr. Brooks does not consider pumps P-558 and P-559 as sources of vapor because they handle a hot slurry. Mr. Brooks believes that the material in P-558 (within 50 feet of the control room) would auto-ignite if leaked to the atmosphere (Tr. 1301-1303). P-558 was 27 feet from #2 substation (Tr. 1304). According to Mr. Brooks, pumps P- 556 andd P-557 are not sources of vapor because they handle heavy fuel oil, a Class III liquid (Tr. .1261). S:\1984\Processed\79-0570-E\00000000\40.tif " The closest source of vapor to the west electrical disconnects was the control valve on X-130 and further than 50 feet away (Tr. 1263). He measured the distance from the west disconnects to D-66, however, as 39 feet (Tr. 1328). According to Mr. Brooks, the No. I electrical substation was in a Class I. Division 2 area because the control valve associated with the desalter was within 50 feet of the substation (Tr. 1265). Mr. Brooks testified that the distance from D-66 to P+M 32-11 was 78 feet 7 inches. This measurement was made by Mr. Greg Hicks, not Mr. Brooks. If the distance from D-66 to P+M 32-11 was actually 45 feet (as measured by Officer Bakewell), Mr. Brooks still does not consider D-66 a source of vapor because D-66 does not have a control valve, and Mr. Brooks does not consider an "ordinary valve" as a source of leak (Tr. 1297). Based on Exhibit R-3, the 78-foot, 7 inch measurement is preferred over the 45-foot measurement of Officer Bakewell. Mr. Brooks' measurement from X-239 to P+H 322-11 was 63 feet 6 inches and 51 feet 6 inches from X-239 to transformer C-17 (Tr. 1299-1361). Mr. Brooks measured the control valve below exchanger 130 to be 51 feet from the west electrical disconnects (Tr. 1328). D-66 was "39 feet exactly" from the west electrical disconnects (Tr. 1238). Discussion As Conoco's brief points out, the Secretary's interpretation and application of Article 500-4(b) of the NEC to the unapproved electrical equipment at issue can be "simply" stated: a Class 1, Division 2 area in an oil refinery must extend a minimum of 50 feet from the nearest source of hydrocarbon, in all cases. But, in this case. the alleged Division 2 areas must extend 100 feet from the neatest sources of hydrocarbons. Because the six electrical installations were located less than 100 feet from a source of hydrocarbon, the Secretary maintains that Conoco violated the S:\1984\Processed\79-0570-E\00000000\41.tif " sections of Article 501 of the NEC by not installing "approved" electrical equipment or, in lieu thereof, positive pressure ventilation. Article 500-4 of the NEC is couched in imprecise and general terms and provides little or no guidance on how to determine the existence and extent of a Class 1, Division 2 location. Article 500-4(b) simply states that a Class 1, Division 2 location is an area "in which volatile flammable liquids or flammable gases are handled, processed, or used, but in which the hazardous liquids, vapors or gases will normally be confined within closed containers or closed systems from which they can escape only in case of accidental rupture or breakdown in such containers or systems or in case of abnormal operation of equipment..." Although Article 500-4 of the NEC does not define a Class I, Division 2 location in terms of a specific distance from a potential source of flammable material, both the Secretary's witnesses, Officer Bakewell and Mr. McAninch used specific distances as the critical factor in determining the existence and extent of the Class 1, Division 2 locations at Conoco's refinery. Conoco's witnesses Mr. Howell and Mr. Brooks, also used specific distances as benchmarks for determining the extent of Class I, Division 2 areas as defined by the NEC. In Asamera Oil Co._, Inc., 9 BNA OSHC 1426, 1981 CCH OSHD (P)25,131, No. 79-0949 and 79-1756, 1980), this judge held that Article 500-4 of the NEC was imprecise on its face and that its provisions standing alone failed to provide employers with an appropriate basis or standard capable of determining the existence and extent of Class I, Division 2 areas. This decision also held that Article 500-4 was unconstitutionally vague as applied to Asamera. At the hearing in this case, however, Conoco abandoned its affirmative defense that Article 500-4 of the NEC is unenforceably vague (Tr. 157). S:\1984\Processed\79-0570-E\00000000\42.tif " Conoco's abandonment of this defense, however, does not relieve this Commission of its duty to interpret and apply Article 500-4 in a manner which provides employers with an objective standard of violation. In interpreting and applying broadly worded standards of the Act, the courts and the Commission have focused on how an employer, his industry, or a "reasonable person" interprets the requirements imposed by the standard. E.g., B&B Insulating, Inc., OSHRC 583 F.2d 1364 (5th Cir. 1978); Cotter & Co. v. OSHRC, 598 F.2d 912 (5th Cir. 1979); S&H Riggers & Erectors, Inc., 79 OSAHRC 23/A2, 7 BNA OSHC 1260, 1979 CCH OSHD [[P]]23,480 (No. 15855, 1979). Therefore, the critical inquiry here, is whether Conoco or the petroleum refinery industry, or a reasonable person familiar with the petroleum refinery industry, would recognize that the NEC requires Conoco's six cited locations to be classified as Class I, Division 2 locations. The Secretary claims that RP 500A constitutes the petroleum refinery industry's interpretation of Article 500-4 and the general practice of the petroleum refinery industry with respect to classifying Class I, Division 2 locations. Conoco agrees, but disagrees that RP 500A requires that a Class I, Division 2 area must extend a minimum of 50 feet, or that RP 500A requires a 100 foot extension with respect to the six cited locations. RP 500A, Exhibit C-32, is not an occupational safety and health standard under the act, but is an industry document - an attempt by the API to present to the petroleum refinery industry a guide for classifying electrical installation areas consistent with the NEC. But there is no attempt to denote its recommendations as mandatory; clearly, the recommendations are intended to be advisory only. The foreword of Exhibit C-32 contains the following statements and disclaimers: "This publication includes generalized statements and recom mendations on matters on which there are diverse opinions. It is important, therefore, that Judgment take precedence over a literal interpretation of the text... The information contained in this publication does not constitute, and should not be construed to be a code of rules or regulations." S:\1984\Processed\79-0570-E\00000000\43.tif " The Secretary contends that the recommended 50-foot distance in Figure I at p.10 of RP 500A must be construed as a "minimum" when determining the extent of a Class I. Division 2 area. Mr. McAninch supported this contention and testified that Amoco considered the 50-foot distance as a minimum in determining the extent of a Division 2 area (Tr. 1073). At another point in his testimony, however, he suggested that he might not always require a minimum of 50 feet (Tr. 1084). Mr. Howell contradicted Mr. McAninch's testimony in regard to Amoco's practice and testified that unidentified Amoco representatives had informed him that Amoco in some cases extended Class 1, Division 2 areas less than 50 feet (Tr. 1848-1485). He further testified that the refineries of Atlantic Richfield, Union, Ohio, Texaco, and Chevron also used less than a 50-foot distance in defining a Division 2 area (Tr. 1402). Mr. Brooks, of course, testified that Conoco does not interpret Figure 1 of RP 500A as establishing a minimum 50-foot distance (Tr. 1248). Contrary to Conoco's contentions,Section 11 and Figure 1 of RP 500A must be read as recommending a 50-foot minimum distance in Division 2 locations in petroleum refineries. But a recommendation, advisory in nature. cannot be the basis for finding a violation of the Act even if the Secretary had adopted RP 500A pursuant to Section 6(a) of the Act. See A. Prokosch and Sons Sheet Metal, Inc., 80 OSAHRC 96/A2, 8 BNA OSHC 2077, 1980 CCH OSHD (P)21,275 (No. 76-576, 76-406, 1980) and the cases cited thereunder. Furthermore, it cannot be concluded that the petroleum refinery industry generally or Conoco specifically recognizes that the NEC mandates the use of a minimum 50-foot distance in all situations when determining the existence and extent of Class 1, Division 2 locations. All three experts interpreted the recommended additional 50-foot Division 2 area in Figure 1 of RP 500A as applicable only "where large releases of volatile products may occur." RP 500A, however, does not specifically define "large releases . S:\1984\Processed\79-0570-E\00000000\44.tif " Although Mr. McAninch testified on direct examination that all sources of vapor referenced bv the Secretary required an additional 50 foot extension, he acknowledged on cross-examination that a 100-foot area classification rule was not customarily employed by the petroleum refinery industry with respect to small process units such as those at Conoco's Denver refinery (Tr. 1119). He also testified at his deposition that he was unaware of any instance where Amoco used a 100-foot radius for a Division 2 location (Tr. 1123-1124). He also indicated that he would define "huge quantities" of vapor in terms of storage-type units such as propane storage units containing 10,000 barrels (Tr. 1119). Mr. Howell testified that judgment must be used in defining "large volumes of hydrocarbons" (Tr. 1167). Mr. Howell further stated that there was no consistent application of the 100-foot recommendation within the petroleum refinery industry. According to Mr. Howell, a major segment of the petroleum refinery industry, including Gulf. Amoco, Union, Texaco, Shell, Union, Arco, and Standard of California did not classify refining process units beyond 50 feet (Tr. 1415). He conceded however,that other companies do go beyond 50 feet (Tr. 1536- 1537). Conoco similarly interprets the additional 50-foot area as applicable only to storage areas containing extremely large quantities of hydrocarbons (Tr. 1248-1249; See also, Tr. 1310, 1335). As previously noted, RP 500A does not specifically define "large releases," and there is no indication that the petroleum refinery industry has arrived at a consensus definition of that term through use of the 100-foot standard. Therefore,it is impossible to ascertain under what circumstances the petroleum refinery industry recommends that the 100-foot distance should be used when determining the existence and extent of Class 1, Division 2 locations. S:\1984\Processed\79-0570-E\00000000\45.tif " In summary, Article 500-4 of the NEC provider no ascertainable objective standard of violation. There is no way of determining when an employer has failed to comply with Article 500-4's requirements concerning classification of Class I, Division 2 areas. Based on the provisions of Article 500-4. who can say with any degree of accuracy what is the extent of a Class I, Division 2 area? Nor can the existence and extent of a Class I, Division 2 location be resolved by resort to RP 500A. This document is nothing more than an attempt by the petroleum refinery industry to interpret the vague classification provisions of the NEC; it does not purport to establish an objective standard of violation for the NEC. As the Secre tary's counsel conceded, a "violation" of the RP 500A does not constitute a violation of the NEC (Tr. 152). Moreover, RP 500A, also is vague and imprecise as to when an additional 50-foot distance should be used in determining the extent of Class 1, Division 2 locations. Even assuming that the RP 500A 100-foot separation requirement was mandatory and noncompliance with that requirement a violation of the NEC and the Act, no violation of that requirement could be found in this case because the record fails to establish that the petroleum refinery industry would have used the 100-foot rule at Conoco's refinery, or that Conoco actually recognized the need to extend its Division 2 areas 100 feet. For the foregoing reasons, no violation of {sec}1910.309(a) will be found with respect to the first five locations previously referenced. In view of Conoco's admission or concession that substation No. 1 is located within a Class I, Division 2 location, a violation of {sec}1910.309(a) and Article 501 of the NEC with respect to that location is affirmed. Conoco's argument that the requirements of Articles 500 and 501 do not apply to this installation because it was installed before March 15, 1972 must be rejected on the authority of Deleware & Hudson Railway, 1980 CCH OSHD {sec}24,422; 8 BNA OSHC 1252 S:\1984\Processed\79-0570-E\00000000\46.tif " (no. 76-787, 1980). In that decision the Commission held that §1910.309(a), incorporating Articles 500 and 501 by reference, contains no clause excluding electrical equipment installed before 1972 and that the pre- 1972 limitation in §1910.309(b) does not apply to the NEC Articles incorporated in former §1910.309(a). A penalty of $100.00 for this violation is consistent with the agreed upon penalty for Item 11C(b) of Citation 1 (See Ex. J-1), and will be assessed. Unable to find that Conoco's control room is a Class I. Division 2 location, the alleged "willful" charge under Item 2a of Citation No. 2 with respect to the control room also must be vacated. To obviate the need for a remand should a reviewing authority find that the NEC can be applied to the facts of this case and that the NEC mandates the use of a 50-foot minimum distance in classifying a Class I, Division 2 location at Conoco's refinery, special findings will be made with respect to the 5 Conoco locations at issue.8/ 1. Control Room and Office Building Complex ("A" on Ex. R-3) No potential sources of volatile flammable vapor were 50 feet or less from the control room (Tr. 1314-1315, 142). Although the control room was equipped with positive pressurization equipment, adequate positive pressurization was not maintained in the control room. Conoco intentionally deceived the representatives of its insurance carrier and led them to believe that the control room was adequately pressurized (Tr. 1025- 1026). The motive for this deception however, and the terms of Conoco's insurance agreement are unknown. Without this knowledge, it is impossible to conclude whether or not this 8/ No findings with respect to the alleged 100-foot classification rule for Class 1, Division 2 locations are necessary because sources of vapor were within 100 feet of the 5 locations. S:\1984\Processed\79-0570-E\00000000\47.tif " deception constitutes recognition of the need to classify the control room as a Class I, Division 2 location. 2. Water Pumphouse. The nearest source of flammable vapor to the water pumphouse was D-66, and was more than 50 feet from the pumphouse. 3. West Electrical Disconnects The nearest source of vapor to the west electrical disconnects, D-66, was within 50 feet of the west electrical disconnects (Tr. 1328). 4. P+M 322-11 This location was probably more than 50 feet of the nearest source of vapor, W-54 (Tr. 1299, 1426; Ex. R-3). 5. Electrical Substation 2 This location was within 50 feet of the nearest source of flammable vapor, W-17 (Tr. 1469-1470, 1472). Citation No. 2, Item I This item charges Conoco with willfully violating Section 5(a)(1) of the Act "in that" Conoco "allowed and/or required relief valves on pressure vessels D-241 (stabilizer for reflux drum) and D-48 (P.B splitter reflux drum) to be turned down so that the relieving pressure would be increased above the normal and recommended safe pressure setting, which created the hazard of increasing the internal relieving pressure on the pressure vessels and their associated equipment." A. Background The function of a safety relief valve is to prevent overpressure in process equipment. Failure to relieve overpressure can result in leaks and rupture of the equipment. Whenever the pressure in the process system exceeds the "set point" or "release point" of the safety relief valve, the spring-loaded disk of the valve opens, discharging the overpressure into a closed flare system and onto a remote S:\1984\Processed\79-0570-E\00000000\48.tif " flare where the released hydrocarbons can be safely burned. Once the pressure in the process system is reduced to the pressure of the set point of the relief valve the relief valve is designed to automatically close (Tr. 15, 73; Ex. R-7). The "set" or "release" point of a relief valve necessarily must exceed the normal operating pressure of the process, but it is set lower than the "maximum allowable working pressure" of the equipment protected by the relief valve. "Maximum allowable working pressure," as defined in Section 8 of the Pressure Vessel Code of the American Society of Mechanical Engineers (ASME) is the calculated maximum pressure at which the equipment can safely operate. A publication of the American Petroleum Institute (API), the Guide for Inspection of Refinery Equipment, Chapter XVI, Pressure Relieving Devices, provides that the pressure at which a relief valve releases should be within + 1/2 percent of the prescribed set pressure before the relief valve is approved for service (Section 16.8.2.7; Ex. C-50). The same API Guide also contains detailed inspection and testing procedures to be followed to assure that the proper pressure setting and release of relief valves is maintained. (For example, see Sections 16.5.4, 16.6, 16.7 and 16.8, Ex. c-50). The two relief valves in question were located in Conoco's Catalytic Polymerization Unit (CPU). The CPU included a "reaction" process and a "fractionation" process. The fractionation process had three separate sections containing a "stabilizer" tower, W-68, a propane - butane "splitter" tower, W-69, and a "le-ethanizer drier." The stabilizer tower, W-68. ("A" on Ex. R-6) first separated the product or "feed" coming from the reaction process into a "poly-gasoline" and a propane mixture. The gasoline flowed out of the bottom of tower W-68 to storage and the propane - butane mixture flowed out the top of W-68, down through four heat exchangers, X-307, X-308, X-309 and X- 310, and into the reflux drum, D-241 ("B" on S:\1984\Processed\79-0570-E\00000000\49.tif " Ex. R-6). A safety relief valve was located on top of D-241. A safety relief valve also was located on top of tower W-68 ("PSU 431" on Ex. R-6). The propane/butane mixture in D-241 was them pumped to splitter tower W-69 ("C" on Ex. R-6). W-69 separated or "split" the propane from the butane. Butane flowed out of the bottom of W-69 and was sent for further treating and storage. Propane flowed out the top of W-69, down through two heat exchangers, X-315 and X-316, and into a reflux drum, D-48 ("D" on Ex. R- 6), and then into further treating and storage. A safety relief valve was located on the top of reflux drum D-48. The facts surrounding Conoco's admitted changes of the pressure set or release points on the safety relief valves on D-48 and D-241 will be set forth below. B. Summary of the Evidence 1. D-48 Relief Valve On the morning of September 23, 1978, the relief valve on reflux drum D-48, opened, relieving hydrocarbons into the flare system. The splitter system was operating at 265 pounds per square inch guage (psig) at the time of release according to the pressure indicator in the control room; the normal operating pressure on D-48 is "270, 275 psig." This relief valve had been initially set to release according After the release, the pressure indicator registered no pressure above 250 psig, the lowest recording point on the pressure indicator. The valve in D-48 continued to relieve (Tr. 1569-1570). Mr. Greg Lorimer, Conoco's operations superintendent and acting shift supervisor, then went onto the second level of the cat-poly unit to inspect the relief valve on D-48. According to his testimony the relief valve in D-48 was vibrating and relieving, and the protective cap of the valve was laying on the deck. The lock nut, ("N" on Ex. R-7), which keeps the adjustment screw ("S" on Ex. R-7) in place, was separated from the adjustment screw by several turns. The adjust S:\1984\Processed\79-0570-E\00000000\50.tif " ment screw sets the tension on the spring of the relief valve, and when the set tension pressure is exceeded, the seat opens and the valve relieves (Tr. 1571- 1574). When the adjustment screw is turned down or tightened the spring tension is supposed to increase and raise the pressure point at which the relief valve will open or relieve (Tr. 1574). According to Mr. Lorimer, he saw the adjustment screw turn counterclockwise or upward, two turns. When an adjustment screw turns counterclockwise, the tension on the spring is supposed to decrease and lower the pressure point at which the valve relieves (Tr. 1574- 1575). The pressure guage on drum D-48 then read 200 psig; the relief valve was vibrating, and a small amount of hydrocarbons was seeping from the stem of the valve (Tr. 1576). Mr. Lorimer ordered the employee accompanying him to turn down on the adjustment screw until the valve on D-48 stopped relieving but "in no case more than two turns." The employee turned down the adjustment screw "approximately one-half to one turn down" and the D-48 relief valve stopped relieving (Tr. 1576). After the adjustment screw was turned down, Mr. Lorimer, again looked at the pressure guage on D-48. "[I]t was still 200 psig, but I could see that the pressure was beginning to increase" (Tr. 1578). Mr. Lorimer than went back to the control room and gave the operators instructions to watch the pressure on tower W-69 and take immediate steps to relieve the pressure if it increased above 280 psig (Tr. 1579). Mr. Lorimer then went to his office, about 40 to 50 feet from the control room, to make a phone call. He claims he wanted to ascertain whether there was a possibility of making an "on-stream" adjustment of the relief valve setting so that "we could continue to operate the vessel" or whether it would be necessary to shut the vessel down, remove the relief valve and "reset it" (Tr. 1579). He tried unsuccessfully to telephone Conoco's mechanical inspector, mechanical craft superintendent, and the refinery manager, and then returned to the control room (Tr. 1580-1581). S:\1984\Processed\79-0570-E\00000000\51.tif " The relief valve on D-48 again relieved, about "an hour or an hour and a half" after the employee turned down the adjustment screw. According to the pressure indicator, the pressure in the vessel at this time of relieving was less than 250 psig (Tr. 1581). Mr. Lorimer then instructed the operators to depressure W-69 and shut it down. This process normally takes approximately an hour (Tr. 1582). On the evening of September 23rd, the relief valve was removed and taken to a shop, "not Conoco's shop," where the pressure was reset at 310 psig. On Conoco's instruction, the shop inserted a set screw in the lock nut to prevent it from turning (Tr. 1583). On September 23rd, Mr. Lorimer was unaware of Conoco's existing policy at the Denver refinery which "requires permission of the refinery manager before any 'hot setting' or on-line adjusment to a relief valve" can be done (Tr. 1584). 2. D-241 Relief Valve On either September 30 or October 1, 1978. the relief valve on D-241 was releasing product to the flare system. The relief valve was leaking and frost had formed on the outlet piping of the relief valve (Tr. 1657-1658). Normal operating pressure of D-241 was "120 pounds" and the relief valve was set to release at "150 pounds" (Tr. 1660). Mr. Rhodes, Conoco's Senior Process Engineer, who was acting as shift supervisor at the time, ordered the operators to apply steam to melt any ice or dislodge any material that might have been preventing the relief valve from resetting. After about 15 or 20 minutes there was still ice on the outlet piping and the relief valve had not reset (Tr. 1658-1659). Mr. Rhodes then had the operators lower the pressure on the unit to 90 or 95 pounds. After 25 or 30 minutes there was still frost on the outlet piping of the relief valve indicating that it had not reset (Tr. 1660). Five minutes after this phone call to Mr. Lyles, the relief valve reseated at an operating pressure of 90 to 95 pounds (Tr. 1661). The operating pressure S:\1984\Processed\79-0570-E\00000000\52.tif " then was raised to 120 pounds (Tr. 649-659). After Mr. Lyles arrived, Mr. Rhodes ordered him to tighten the adjustment screw on the D-241 relief valve. Mr. Lyles testified that Mr. Rhodes ordered him to take down the relief valve "two rounds" or two turns in the adjusting screw. In response Mr. Lyles told Mr. Rhodes, "I will give you one" (Tr. 346). According to Mr. Carey, who was present, the adjustment screw was turned down "five flats" (Tr. 651). Mr. Rhodes testified that the screw was tightened "about a turn, a turn and a half" (Tr. 1664). Mr. Rhodes contends that his sole purpose in adjusting the valve was to prevent the relief valve from again leaking (Tr. 1664). Based on Mr. Rhodes' "gut feeling," he believes that turning down the D-241 relief valve would increase the pressure at which it would relieve "about 7 pounds or so" (Tr. 1691). The relief valve on D-241 is designed to protect D-241 from overpressure that might occur on that vessel when (1) the four "block" or "gate" valves to the heat exchangers, X-207, X-308, X-300 and X-307 are closed, and (2) D-241, is full of liquid and (3) D-41 is exposed to external fire (Tr. 1665-1668, 1817, 1828). The relief valve on W-68 is designed to relieve any overpressure on D-241 that might occur when those three conditions are not present. J. Discussion To establish that an employer has violated Section 5(a)(1) of the Act, also known as the "general duty clause," the Secretary must prove by a preponderance of the evidence: (1) that a hazard likely to cause death or serious bodily harm existed at the employer's workplace; (2) that the hazard was recognized as such either by that employer or generally within the industry, and (3) that there was a feasible method by which the employer could have abated the "recognized hazard. National Realty and Construction Co., Inc. v. O.S.H.R.C., 489 F.2d 1257 (D.C. Cir. 1973); Usery v. Marquette Cement Manufacturing Co., 568 F.2d 902 (2d Cir. 1977); S:\1984\Processed\79-0570-E\00000000\53.tif " Brennan v. O.S.H.R.C.and VY Lactos Laboratories, 494 F.2d 460 (8th Cir. 1974); Baroid Division of NL Industries, Inc., No. 79-1775 (10th Cir. September 21, 1981). The language of the citation identifies the recognized hazard as increasing the relieving pressure of the relief valves on D-48 and D-241 above the "normal and recommended safe pressure" settings for these particular valves. The hearing record, however, shows that the parties also tried an alternative theory of recognized hazard, namely, Conoco's "on-stream" or "on-line" pressure setting adjustments on valves D-48 and D-241. This alternative theory of recognized hazard, therefore, shall be treated in all respects as if it had been raised by the pleadings. See Baroid Division of NL Industries, supra; See also, Rule 15(b) of the Federal Rules of Civil Procedure. The petroleum refinery industry generally, and Conoco specifically, recognize that properly designed, applied, installed, and maintained, pressure releasing devices, are essential to the safety of refinery employees. Without the protection of such devices the pressure within the process equipment could foreseeably exceed the maximum allowable working pressure of that equipment and result in rupture of the equipment. The likely consequence of any such rupture, of course, would be serious physical harm or death to employees. Conoco also readily concedes that failing to set a relief valve to release below the maximum allowable pressure working pressure of the vessel it protects is hazardous. Conoco, however, denies that the relief valves in question were ever set to release above the maximum allowable working pressure on D-48 and D-241. Conoco further contends that so long as the set pressure point on a relief valve does not exceed the maximum allowable working pressure no hazard exists. The record contains no evidence that the maximum allowable working pressures on D-48 and D-241 were exceeded on the dates in question. In point of fact, the maximum allowable working pressures on these vessels were never established. The S:\1984\Processed\79-0570-E\00000000\54.tif " record also fails to establish, as alleged by the Secretary in his citation, that the initially set pressure of 300 psig on the D-48 relief valve was ever exceeded or that the initial set pressure on D-241 of 150 psig was exceeded as the result of the on-line pressure adjustments. In the absence of this proof, the Secretary failed to sustain his first theory of recognized hazard as alleged in his citation. The Secretary's second theory of recognized hazard, however, that Conoco's on-line adjustments on the D-43 and D-241 relief valves constituted recognized hazards, finds ample evidentiary support in thie record. A "safety hazard" (recognized hazard) has been defined as a condition (or practice) at the worksite that creates or contributes to an increased risk that an event likely to cause death or serious physical harm to employees will occur. See Baroid Division of NL Industries, supra. A critical inquiry in this case, therefore, is whether Conoco's on-line adjustments of the relief valves on D-48 and D-241 increased the risk of overpressure on these vessels. If they did, recognized hazards existed. Numerous sections of the API Guide (Ex. C-50) support the conclusion that the failure to properly set a relief valve to release at a prescribed safe level Is recog nized as hazardous by the petroleum refinery industry (See Sections 16.5.4, 16.7.2, 16.8.2.7, 16.8.3.1, Ex. C-50). These sections emphasize the importance of accurately testing to determine the set pressure at which a valve will relieve. Even when experience has indicated that the repair of valves on- line is a safe and suitable practice, the API Guide recognizes the importance of testing the valve for its release set point by an approved procedure. Furthermore. Conoco, itself, has previously declared that "a good, safe operating" practice is to "verify safety valve set points" when "hot" setting a relief valve to stop leaks (Ex. C-35). It logically follows, therefore, that any on-line change in the relieving pressure point of a malfunctioning relief valve, without determining its new pressure S:\1984\Processed\79-0570-E\00000000\55.tif " relief point, also is recognized as hazardous by both the petroleum refinery industry and Conoco. On this record, Conoco's adjustments to the relief valves on D-48 and D-241 are found to be recognized safety hazards. Conoco's argument that the adjustments in question were not hazardous because there was no proof that the set pressures after the adjustments exceeded the maximum allowable working pressure levels of the equipment is unpersuasive. It is similar to claiming that spinning a revolver barrel and playing a game of Russian roulette without knowing the location of the bullet is not hazardous so long as the hammer strikes an empty chamber. Conoco additionally contends that adjusting the pressure an the safety valve on D-241 was not hazardous because the safety valve on tower W-68 eliminated any possibility of overpressure on D-241. Not so. The failure to determine the set pressure point on the malfunctioning safety valve on D-241 after attempting to increase the pressure above the relieving pressure point clearly added to the risk of overpressure in the event the gate valves became blocked and D-241 became exposed to an external fire. . Conoco focuses on the lack of proof that the set pressure on the D-241 relief valve exceeded the maximum allowable working pressure of the drum, that the gate valves ever became blocked, and that D- 241 was exposed to a fire. But as previously stated, the recognized hazard with respect to both relief valves is changing the relieving pressure point on a malfunctioning relief valve without determining the set point at which it will relieve. The Secretary also had the burden of showing that Conoco knew, or with the exercise of reasonable diligence could have known, of the likelihood of the hazardous practice in this case. See Mountain States Telephone & Telegraph Co. v. OSHRC, 623 F.2d 155 (10th Cir. 1980); National Realty and Construction Co., supra. The Secretary sustained this burden. The evidence shows that Conoco's policy which for S:\1984\Processed\79-0570-E\00000000\56.tif " bids turning down safety valves to stop them from leaking without the express permission of the refinery manager or, in his absence, the process superintendent, was not effectively communicated and implemented by Conoco. Mr. Lorimer never heard of the prohibition; and Mr. Rhodes deliberately ignored it. As was pointed out in the National Realty case: "the fact that a foreman would feel free to break a company policy is strong evidence that implementation was lax." Because Conoco's safety rule in this regard was not effectively communicated and implemented, it is found that Conoco failed to exercise reasonable diligence in discovering these instances of violation. See H.B. Zachry Co. v. OSHRC, No. 80-1357 (5th Cir. Mar. 2, 1981). Having found a violation of the general duty clause, the final inquiry is whether the violation was willful. A willful violation of the general duty clause is established when an employer makes a conscious and deliberate decision not to comply with its requirements or when an employer's conduct is properly characterized as showing a careless disregard of the requirements or an indifference to employee safety. Tri-City Construction Company, 8 BNA OSHC 1567 (No. 77-3668, 1980). After Conoco was placed on notice of the hazardous practice of turning down safety valves to stop leaks, Conoco took action to remedy this practice by issuing a safety directive. Although inadequate, the action taken exhibited an intent to eliminate the known and recognized hazardous practice, not to permit its continued existence. See Mel Jarvis Construction Co., Inc., (No. 77-2100, September 30, 1981). The violation, therefore, will be affirmed as serious, not willful. Penalty The high gravity of this violation dictates a substantial penalty. In addition, the two incidents reveal a glaring deficiency in the communication and implementation of Conoco's safety program. To impress upon Conoco the need to correct this deficiency, the maximum permissible penalty of $1,000.00 will be assessed. S:\1984\Processed\79-0570-E\00000000\57.tif " Findings of Fact All findings of facts relevant and necessary to a determination of the contested issues have been found specially, and appear above in the body of the decision. (See Rule 52(a) of the Federal Rules of Civil Procedure). Conclusions of Law 1. The record fails to establish by a preponderance of the evidence that respondent violated 29 C.F.R. §1910.134(a)(2). 2. The record establishes by a preponderance of the evidence that respondent committed a serious violation of 29 U.S.C. §654(a)(1) with respect to the inadequate training of Mr. Ronald D. Branson. 3. The record fails to establish by a preponderance of the evidence that respondent violated 29 U.S.C. §654(a)(1) with respect to the separation and protection of respondent's heaters and boilers. 4. 29 C.F.R. §1910.309(a), incorporating Articles 500 and 501 of the National Electrical Code (1971 edition), applies to electrical installations installed before March 15, 1972. 5. Respondent judicially admitted that its electrical substation No. 1 was in a Class 1, Division 2 location. 6. The record establishes by a preponderance of the evidence that respondent's electrical substation 1 contained electrical equipment not "approved" or "explosion-proof." 7. Respondent's electrical substation 1 was in violation of §1910.309(a). 8. Standing alone, Article 500-4 of the National Electrical Code, 1971 edition, provides no objective standard to determine the existence and extent of Class 1, Division 2 locations. Unless the record evidence establishes that an employer or his industry recognizes the need to classify a particular location as Class 1, S:\1984\Processed\79-0570-E\00000000\58.tif " Division 2, application of Article 501 of the NEC. and §1910.309(a) of the Act is impossible and no violation can be affirmed. 9. In this case, the record fails to establish that Conoco or the petroleum. refinery industry recognizes the need to classify the five electrical installations at issue as Class 1, Division 2 locations. 10. The record establishes by a preponderance of the evidence that respondent committed a serious violation of 29 U.S.C. §654(a)(1) with respect to the pressure setting adjustments on the relief valves for D-48 and D-241. The record fails to establish by a preponderance of the evidence that respondent willfully committed this violation. ORDER Based on the Findings of Fact, Conclusions of Law and the entire record, its ORDERED: 1. Item Nos. 4, 6, and 11c(b), as amended by the complaint, are AFFIRMED, and penalties of $280.00, $360.00 and $100.00 respectively are ASSESSED. 2. Respondent shall comply with the abatement procedures of paragraph 3 of the Stipulation (See Ex. J-1). 3. Item 8 of Citation No. I is VACATED, but respondent shall implement the measures agreed to in paragraph 4 of the Stipulation (See Ex. J-1). 4. The alleged violation of 29 C.F.R. §1910.134(a)(2) is VACATED. 5. A serious violation of 29 U.S.C. §654(a)(1) with respect to the inadequate training of Ronald D. Branson is AFFIRMED, and a penalty of $100.00 ASSESSED. 6. The alleged violation of 29 U.S.C. §654(a)(1) with respect to the separating and protection of respondent's heaters and boilers is VACATED. 7. The alleged violation of 29 C.F.R. §1910.309(e), (Article 501 of the NEC) with respect to electrical substation No. 1 is AFFIRMED, and a penalty of $100.00 ASSESSED. S:\1984\Processed\79-0570-E\00000000\59.tif " 8. With the exception of Items 11c(b) and 11c(e). of Citation 1, as amended, the other alleged serious violations with respect to Article 501 of the NEC are VACATED. 9. Item 2a of Citation No. 2 is VACATED. 10. Item Nos. 1, 7, and 10 of Citation No. 1, as amended, are VACATED. 11. A serious violation of 29 U.S.C. §654(a)(1) with respect to the adjustments of the relief valves on D-48 and D-241 is AFFIRMED, and a penalty of $1,000.00 ASSESSED. James A. Cronin,Jr. Judge,OSHRC Cror4in,_Jr. A judge, OSHRC Dated: January 6, 1982 FOOTNOTES: [[7/]] Conoco concedes that substation No. 1 is located within a Class I, Division 2 location but argues that the requirements of 29 C.F.R. §1910.309(a) do not apply to electrical equipment, including the equipment in substation No. 1, installed before March 15, 1972.