“Secretary of Labor,\t Complainant\t v.\t OSHRC Docket No. 04-2179Loren Cook Company,\t Respondent.\t Appearances: Oscar Hampton, Esquire and Leigh Burleson, Esquire, Officeof the Solicitor, U.S. Department of Labor, Kansas City, Missouri For Complainant Douglas B. M. Ehlke, Esquire, Ehlke Law Offices, FederalWay, WashingtonFor RespondentBefore: Administrative Law Judge Ken S. WelschDECISION AND ORDER Loren Cook Company (Cook), manufactures metal parts forcommercial and industrial fans at its facility in Springfield, Missouri.Occupational Safety and Health Administration (OSHA) compliance officerRobert Vezeau conducted a complaint inspection of Cook\u2019s facility onOctober 4, 2004. Although he found no basis for the complaint thatprompted the inspection, Vezeau observed what he considered a violationof 29 C.F.R. \u00a7 1910.212, the machine guarding standard. On November 4, 2004, the Secretary issued Cook a one-itemcitation alleging a repeat violation of \u00a7 1910.212(a)(1), for failing toprovide machine guarding on certain semi-automatic spinning machines.Cook contested the citation and the proposed penalty of $17,500.00. Therepeat classification is based on a citation issued in February 2004,for a violation of the same standard. The parties resolved that citationby an informal settlement agreement. The court held a 13-day hearing in this matter during Mayand July, 2005, in Republic and Springfield, Missouri. The partiesstipulated jurisdiction and coverage. They filed post-hearing briefs.Cook opposes the citation on several grounds, but its central argumentis that guarding the cited spinning machines is both technologically andeconomically infeasible. For the reasons discussed below, the court determines Cookestablished guarding the spinning machines was infeasible. Item 1 of thecitation is vacated and no penalty is assessed. FootnoteFacts Cook manufactures fans and blowers for commercial andindustrial use at its plant in Springfield, Missouri. Cook employsapproximately 400 employees, who work in three shifts, five days a week.The plant comprises two buildings. Building 2 hold six semi-automaticspinning machines and approximately fifteen manual spinning machines.The six semi-automatic spinning machines are located in a line anddesignated as machines #1 through #6. Cook purchased the machines in thelate 1980s or the early 1990s from a company called Leifeld. Themachines come in various models; Cook owns the models PNC 150, PNC 350,and PNC 750 (Exhs. C-5, C-6, C-7, R-20). Footnote Cook producesapproximately 70% of its product parts on the semi-automatic spinningmachines. In order to manufacture a part on a semi-automatic spinningmachine, an employee places a disc of metal (a blank) against a mandrel(also referred to as a chuck). A roller forces the blank against themandrel with hydraulic pressure as the machine spins at a controlledrpm. The spinning process causes the blank to flow and change shape. Atrimming tool trims excess material from the blank as it spins,producing metal chips and shavings. Cook employs two programmers who have specialized trainingto program the spinning machines. They develop programs using joysticks.Programming, which is performed at a slower speed (200 to 250 rpm) thanthe operating phase, requires a great deal of interaction between theprogrammer and the machine. The machine records the individual motionsthe programmer performs. Once the programmer completes the programming,an operator (whose training is less specialized than a programmer\u2019s)places and removes each newly formed part. The largest part produced onthe #3 and #4 machines is 60 inches in diameter. Most of the metal blanks processed by Cook are aluminum.Generally, the aluminum used is .05 gauge which is very thin. A criticalstep in the spinning process is lubricating the blank. Lubricationprevents overheating of the blank and galling on the surface where theroller presses on the blank. If the roller grabs the blank, it couldcause the blank to rip. Based on its years of experience, Cook usesmutton, a paste-like substance made from sheep, as lubricant for theblanks. As it spins, the operator applies mutton to the blanks using apiece of canvas rolled up into a 12-inch tube and taped with duct tape(Exh. R-14). The operator applies the mutton to the front and back ofthe spinning blank. OSHA compliance officer Robert Vezeau watched one operationof the #4 machine. He observed a Cook employee standing next to thenon-rotating shaft, approximately 2 feet from the spinning blank. Subsequently, the Secretary issued the instant citation onNovember 4, 2004. During an informal conference with the Secretary, Cookfurnished the Secretary with a videotape showing a programmerprogramming the #3 machine (Exh. C-5). FootnoteThe CitationTo prove a violation of an OSHA standard, the Secretary must show by apreponderance of the evidence that (1) the cited standard applies,(2) there was noncompliance with its terms, (3) employees had access tothe violative conditions and (4) the cited employer had actual orconstructive knowledge of those conditions. \/Southwestern Bell Telephone Co.\/, 19 BNA OSHC 1097, 1098 (No. 98-1748,2000).Item 1: Alleged Serious Violation of \u00a7 1910.212(a)(1) Section 1910.212(a)(1) provides: One or more methods of machine guarding shall be provided to protect theoperator and other employees in the machine area from hazards such asthose created by point of operation, ingoing nip points, rotating parts,flying chips and sparks. Examples of guarding methods are-barrierguards, two-hand tripping devices, electronic safety devices, etc. Although Cook claims the Secretary failed to meet her burdenon each of the four elements, a fair reading of the record shows theseelements are not in serious dispute.Applicability of Cited Standard Subpart O addresses \u201cMachinery and Machine Guarding.\u201dSection 1910.212(a)(1) addresses \u201cGeneral requirements for allmachines.\u201d The machines cited in amended item 1, spinning machines #3and #4, are covered by the cited standard. Cook argues \u00a7 1910.212(a)(1) does not apply to the citedmachines because the only hazard identified at the hearing is created bythe spinning blank as it is formed into the finished part. The part isthe product sold by Cook to its customers. Cook cites \/Allis-ChalmersCorporation, \/4 BNA OSHC 1876, 1877 (No. 8274, 1976), in support of itsargument that \u00a7 1910.212(a)(1) does not apply to the product beingmanufactured, only to the machine itself. In\/Allis-Chalmers, \/theSecretary cited the employer for failing to guard the rotating reartires of assembled tractors being tested. The Commission reversed theadministrative law judge, who affirmed the citation, holding:The standard at 1910.212(a)(1) is clearly directed towards machineswhich are or can be used in the manufacturing process. . . . Moreover,each of the sections of subpart O specify safety requirements formachines used in manufacturing processes but none specify suchrequirements for products as a result of the processes. The tractors inquestion are not machines which are used in the manufacturing processbut rather are the products of the process. \/ Allis-Chalmers \/is easily distinguishable from the instantcase. In \/Allis-Chalmers, \/the cited hazard was unguarded rotatingparts, and the parts cited were the rotating tires of a newly-assembledtractor. The tractor was not being used to manufacture a product; it wasthe product itself. In the instant case, the blank is clamped between two partsof the spinning machine, the mandrel and the roller. The hazard resultsfrom the action of the machine as it shapes the blank. Cook\u2019sinterpretation is at odds with \u00a7 1910.212(a)(3) (Point of operationguarding). The section provides:Point of operation is the area on a machine where work is actually performed upon the material being processed. The material being processed in this case is the blank. Theplain meaning of the standard renders \u00a7 1910.212(a)(1) applicable to thecited machines and conditions here.Noncompliance with the Terms of the Standard The standard requires guarding to protect employees \u201cfromhazards such as those created by point of operation, ingoing nip points,rotating parts, flying chips, and sparks.\u201d The only hazard establishedby the Secretary is the point of operation hazard created by thespinning blank. The Secretary failed to show any other part of thespinning machines presented a hazard requiring guarding. The Secretarytacitly acknowledges the only hazard at issue is the spinning blank.\u201cThe rotating blank is a point of operation, \u2018material being processed\u2019,anticipated by the standard . . . .\u201d (Secretary\u2019s Brief, p.8.) The record establishes the blank is unguarded as it spinsbetween the mandrel and the roller.Employee ExposureSection 1910.212(a)(1) requires the Secretary to prove the existence ofa hazard. She \u201cmustshow that employees are in fact exposed to a hazard as a result of themanner in which the machine functions and is operated.\u201d \/JeffersonSmurfit Corp.,\/ 15 BNA OSHC 1419, 1421 (No. 89-0553, 1991). In order tomeet this burden, the Secretary must do more than show that it may bephysically possible for an employee to come into contact with theunguarded machinery. Cook requires its operators and programmers to wear personalprotective equipment (PPE) in the form of safety glasses, safety gloves,steel-toed shoes, and hearing protection. The spinning machine must berun during the programming and operation processes. The blank rotates atspeeds of up to 900 rpm, the equivalent of 150 mph. The programmerstands on a platform behind a non-rotating shaft. The rotating shaftmoves into place to hold the blank against the mandrel. The programmersets the tracings and places mutton paste on the blank to keep the metalfrom overheating. The programmer must maintain a sight line to the areawhere the roller follows the contour of the mandrel. Other than whileapplying the mutton paste, the programmer remains approximately 2 feetfrom the rotating blanks. Operators also must be close enough to the spinning machineto see, hear, and feel the vibration as the machine forms the blank todetect any irregularities in the metal as it changes molecularstructure. The operator needs to be close to the point of operation whenlubricating the blank with the mutton paste. Cook argues its programmers and operators are not exposed toa hazard. The company points out that no Cook employee has ever beeninjured by the spinning blank point of operation. The occurrence or absence of injuries caused by a machine isprobative evidence of whether the machine presents a hazard. If,however, the objective facts concerning the operation of the machineshow the presence of a hazard, then the existence of the hazard is notnegated by a favorable safety record which an individual employer mayhave experienced. \/A. E. Burgess Leather Company, Inc., \/5 BNA OSHC 1096, 1097 (No. 12501,1977), \/aff\u2019d,\/ 576 F2d 948 (1st Cir. 1978). The objective facts here show the programmer and operatorapply the mutton paste directly to the front and back of a piece ofmetal spinning at 250 rpm and 900 rpm, using a piece of rolled upcanvas, approximately 12 inches long. The potential hazard is at the endof the rotating blank when the employee reaches behind the blank toapply the mutton paste (Exh. C-5). The Secretary has established exposure during those timesthe employees apply the mutton paste to the spinning blank.Employer Knowledge The management personnel at Cook were well aware programmersand operators of the semi-automatic spinning lathe machines were exposedto the point of operation of the spinning blank Senior operators hadbeen training new operators to apply mutton paste in the prescribedmanner for years. The Secretary has established Cook had actualknowledge of the unguarded blanks.Fair Notice Cook argues it did not have fair notice the Secretary couldcite semi-automatic spinning machines under \u00a7 1910.212(a)(1). Cook alsoargues the citation lacked particularity and so failed to provide itwith fair notice of the facts which formed the basis for the allegedviolation. The court rejects both arguments. The lack of prior citations for violating \u00a7 1910.212(a)(1)or the withdrawal of a citation for that standard during an informalsettlement agreement does not establish the Secretary has a policy ofnot citing semi-automatic spinning machines for guarding violations.\u201cThe Secretary\u2019s failure to issue a citation for a violation of astandard does not immunize an employer from future enforcement of thatstandard.\u201d \/Cardinal Industries, Inc., \/14 BNA OSHC 1009, (No. 82-427,1989). Cook\u2019s argument concerning the lack of particularity is alsorejected, although the company is correct in pointing out the citationwas inartfully drafted, requiring two amendments during the hearing. Itwas clear, however, Cook understood the point of operation of thesemi-automatic spinning machines was the focus of the Secretary\u2019s case.Cook objects to the Secretary\u2019s \u201cmoving target\u201d suggestions forabatement but, as discussed below, it was not the Secretary\u2019s burden toestablish a feasible abatement method.Burden of Proof for Feasibility or InfeasibilityA point of vociferous debate between the two parties was the allocationof the burden ofproof regarding the feasibility of guarding the blanks. Cook contends \u00a71910.212(a)(1) is a general standard, requiring the Secretary toidentify and prove the existence of a feasible compliance method, as ina \u00a7 5(a)(1) violation. The Secretary argues \u00a7 1910.212(a)(1) providesthree suggested abatement methods, which makes it specific enough toshift the burden of proof to the employer, as an affirmative defense.Footnote Cook relies heavily on a Sixth Circuit Court of Appealscase, \/Diebold, Inc. v. Marshall,\/ 585 F.2d 1327, 1333 (6^th Cir. 1978).In \/Diebold\/, the court reversed the Commission\u2019s finding of a violationof \u00a7 1910.212(a)(1), holding:Where a standard imposes a duty without specifying the means ofcompliance, the Secretary has the burden of establishing the specificand technologically feasible means of compliance as an element of hisshowing that a violation has occurred. The Court of Appeals for the Eighth Circuit, in which theinstant case arises, holds in \/Brock v. Dun-Par\/ \/Engineered FormCompany, \/843 F.2d 1135 (8^th Cir. 1998):Where a specific duty standard contains the method by which the workhazard is to be abated, the burden of proof is on the employer todemonstrate that the remedy is infeasible under the particularcircumstances. It is noted that after the \/Diebold \/decision, \u00a71910.212(a)(1) was amended to identify several means of abatement, i.e.barrier guards, two-handed tripping devices, electronic safety devices,etc. Also since the \/Diebold\/ decision, the Commission has continued tohold that the employer had the burden of proving infeasibility whencited for a \u00a7 1910.212(a)(1) violation. \/Consolidated Aluminum Corp., \/9BNA OSHC 1144, 1156-1158 (No. 77-1091, 1980). The court is required tofollow the precedent established by the Commission and the circuit inwhich the case arises. Accordingly, the burden of proving infeasibilityis on Cook, who asserts infeasibility as an affirmative defense.Infeasibility Defense In order to prove the affirmative defense of infeasibility,the employer must show: (1) the means of compliance prescribed by theapplicable standard would have been infeasible, in that (a) itsimplementation would have been technologically or economicallyinfeasible or (b) necessary work operations would have beentechnologically infeasible after its implementation, and (2) there wouldhave been no feasible alternate means of protection. \/V.I.P. Structures,Inc.\/, 16 BNA OSHC 1873, 1874 (No. 91-1167, 1994). The fact complianceis difficult or expensive is insufficient grounds to excuse compliancefrom the requirement of the standard. \/Hughes Brother, Inc., \/16 BNAOSHC 1830 (No. 12523, 1978). The Commission expects employers toexercise some creativity in seeking to achieve compliance. \/Pitt DesMoines, Inc., \/16 BNA OSHC 1429 (No. 90-1349, 1993). The evidence during the hearing was primarily devoted to thefeasibility of the myriad suggestions proposed by the Secretary\u2019switnesses. Thus, while it is Cook\u2019s burden to establish infeasibility,it was the Secretary who proposed the methods of abatement discussed(and dismissed). The Secretary asked OSHA regional office machine guardingspecialist Brian Drake to make suggestions for guarding spinningmachines #3 and #4. Drake did not testify at the hearing. Hissuggestions were, however, the starting point of much of the hearingtestimony. Specifically, Drake suggested using a pressure sensitive matinterlocked to the machine motor; an extended paste application stick;or an internal Plexiglass shield between the operator and the point ofoperation. The Secretary also hired safety engineer Robert Andres as aconsultant to evaluate and suggest abatement methods for guarding thespinning lathe machines. Andres did not consider himself an expert inmetal spinning operations or safety. At the time of the hearing, he hadnever seen a spinning lathe machine. Andres never visited Cook\u2019sfacility and he never spoke with Cook\u2019s programmers or operators. Andres issued a report on May 20, 2005, in which herecommended abatement methods for the unguarded spinning blanks (Ex.C-15). Andres had considered the methods proposed by Drake. Andresconcluded most of Drake\u2019s suggestions were unworkable (Tr. 553-554): After reviewing all the input documents and looking at thesystematic removal of certain suggestions from consideration, I agreewith the Loren Cook Company that a safety net or light curtain method ofguarding the primary hazard, which is the spinning disc, was notfeasible because the disc would not stop fast enough if someonepenetrated that type of a barrier. Other items that had been considered did not seem feasiblefor various reasons. I ended up with, my conclusion was that the obviousprimary barrier for that hazard should be a mechanical or physical barrier. Later on in his testimony, Andres conceded that the barrierguard he was proposing could not be used while operators were applyingthe mutton paste: \u201cNot as they are currently doing it, no\u201d (Tr. 652). Heconceded as well that the primary hazard is the spinning blank (Tr. 662). Andres reviewed reports issued by Cook\u2019s manufacturingengineers, as well as a videotape of the spinning machines beingoperated. He agreed with Cook that, \u201cIt is not possible to take oneguard and cover 100% of the hazards present\u201d (Tr. 551). Jeff Kallenberger is a manufacturing engineer for Cook. Hesupervises Ian Quinn and Eddie Thompson. Together, these three engineersspent over 700 hours attempting to design a feasible guard for thespinning blank. They spoke with the programmers and the operators andtook measurement of all parts of the machine. They used CAD softwaremodeling to evaluate various proposed abatement methods. Cook\u2019s engineers established the infeasibility of severalabatement methods the Secretary no longer seriously pursued by the closeof the hearing. Footnote The engineers\u2019evaluations of the various other abatement methods are as follows: 1. Pressure Safety Mats The Secretary proposed a 74\” pressure sensitive mat thatcould be placed in front of the spinning machine. If an employeeapproaches the machine within this distance, all power to the machinewill be cut off. Quinn and Thompson, as well as Andres, concluded thesafety mat was infeasible (Exh. R-39; Tr. 932, 936, 2143). 2. Interlocked Plexiglass Sliding Barrier Guard Drake recommended using a interlocked plexiglass guardbetween the programmer and the spinning blank. Andres testified theinterlocked barrier guard would interfere with the operator of themachine (Tr. 803). Andres stated the operator could not apply the muttonpaste with the door in place (Tr. 876). Thompson and Quinn also foundthe interlocked door infeasible (Tr. 1945, 2143). 3. Sliding Door A barrier door slides in front of the machine, blockingaccess to the mandrel, blank, and point of operation. When the door isslid out of position, an interlock switch activates and cuts power tothe machine. This would prevent set up and programming duties andlubricating the blanks (Tr. 2568). Andres photographed machines in a plant in North Carolinawhile visiting it on an unrelated case (Exhs. C-20-25). The NorthCarolina machines had a guard for the mandrel. The operator lubricatedthe blanks using a spray applicator (Tr. 2666, 2672). Oncross-examination, Andres admitted he did not know the size and shapesof the parts of the machine, the function of the machines, how theyoperated or how they were programmed. (Tr. 2691). The machines shown inExhibits C-20 through C-25 are not semi-automatic spinning machines,like the ones at issue here. Evidence of the guarding of the NorthCarolina machines has no probative value to the case at hand. 4. Sliding Interlocked Barrier Guard Leifeld, the company that manufactures the spinning machinesat issue, manufactures a sliding door guard for the PNC 150 (Exh. C-9).The door must be closed in order for the spinning machine to beactivated (Tr. 359). With the door closed, operators could not load orunload the blanks, lubricate the blanks, or set up themachine (Tr. 360-361). Winfried Walter, who has worked in the metalspinning industry for over 40 years, testified the spinning machineswould be inoperable with the door closed (Tr. 1065). 5. Lubrication The loading and unloading of blanks are done when the blankis not rotating. Most functions performed by the programmer and operatordo not require them to be within 2 feet of the point of operation.Applying the mutton paste is the one activity that exposes employees tothe hazard of amputation of their fingers or hands while the machine isrunning. The Secretary and Cook\u2019s engineers investigated theconsequences of changing the manner in which the lubricant is applied.Cook vouches strongly for the superiority of mutton as a lubricant,stating repeatedly that mutton has been used as a lubricant in the craftof metal spinning since the time of ancient Egypt. Mutton presents noenvironmental hazards, either in its use or its disposal. Cook believesmutton is the ideal lubricant for aluminum, which comprises 80% to 85%of Cook\u2019s products (Tr. 1848). Cook contends mutton works best toprevent dry spots, which result in damage to the blanks. Andres recommended the use of Gleitmo, a thick sticky pasteused on stainless steel. Cook\u2019s spinning supervisor Jim Pawlikowskitested Gleitmo and several alternative lubricants, including liquidlubricants that can be sprayed onto the blanks (Tr. 2533-2534, 2569).These lubricants did not work as well a mutton, resulting to damage tothe blanks. According to Cook, oils and other lubricants do not have theviscosity to remain on the blanks while rotating at 900 rpm. The most promising method explored by Cook is the use ofextended lubrication sticks. The idea is operators use a longerapplicator, removing their hands from the zone of danger created by thespinning disc. Thompson evaluated this alternative method in his report,concluding an extended lubrication stick is not feasible (Exh. R-39,p.2). Thompson performed several tests using 3D models of the machine toanalyze the effectiveness of an extended stick. Thompson\u2019s evaluationsare as follows:Lubrication Stick Length: Test & Results Test #1: Standard 12\” Lube StickThe operator was observed lubricating the rotating metal blank with thestandard 12\” long lubrication stick. The operator was able to maintaincomplete control over the stick as he ran it back and forth across theface of the blank. By using the 12\” stick, the lubricant was appliedevenly to the blank. Test #2: 30\” Lube StickThe operator was observed performing the same lubrication test as Test#1, but this time a 30\” stick was used. The 30\” stick was not longenough for the operator to use while standing outside of the safety matarea, which does not meet OSHA\u2019s recommendation. This stick was used tocompare any amount of control lost by a stick that was only 2-1\/2 timeslonger than the current [12\”] lubrication stick. The operator did nothave nearly as much control over the stick as he held it against therotating part. Also loss of control could lead to potential strains andinjuries. Test #3: 95\” PVC Lube StickDuring this test, the operator was given a stick that was of adequatelength to keep him out of the simulated safety mat area of 74\”. With the95\” long stick, the operator had no control of where he applied thelube. The stick was extremely hard to control and was thrown off of theface of the rotating blank several times. The long stick was also verydifficult for the operator to hold on to because of the extra forcegenerated by the increased length (leverage), forcing the operator tobrace the stick against his hip. After only lubricating 2 parts theoperator complained of lower back pain and soreness in his wrist. Test #4: 95\” Wood Lube StickThis test was identical to Test #3, but we substitute a wood stick forthe PVC stick. The results were identical to Test #3…no control andincreased forces exerted on the operator causing the before mentionedproblems.Safety Mat and Extended Lubrication Stick Summary:With the operator unable to load the part, program the machine, andproperly lubricate the part blank the safety mats prove to NOT be atechnically feasible solution due to the fact that all three of theseprocesses require the machine to be running to complete these steps.Also, the extended lubrication stick is NOT technically feasible due tothe improper lubrication, lack of control, difficulty to hold on to, andthe exposure to injuries such as muscle, back and wrist strains, cuts,abrasions, and punctures. Other than engaging in speculation, the Secretary failed to offersufficient evidence to rebutCook\u2019s showing of technology infeasibility.Economic Infeasibility Cook also contends it is economically infeasible to guardthe spinning blank. Dennis Blake is the vice president of manufacturingat Loren Cook. Over the past 30 years, Blake has gone to numerousspinning companies to look at their operations, equipment and to discussquality, delivery, and prices (Tr. 1278). Over the past 20 years, Blakehas attended machinery trade shows, including those of the PrecisionMetalforming Association (PMA) to view the newest technology (Tr. 1279).Blake knows the industry practice in supplying spun parts for the fanindustry (Tr. 1279). Cook offers 155 different standard fan and blower productsthat have 10 to 20 different sizes each, and dozens of configurations.They also do custom products. On the two cited semi-automatic machines,the blanks spun range from 20 inches to 60 inches (Tr. 1292-1293, 1298).The six semi-automatic spinning machines are involved in producing 70%of all products shipped by Cook (Tr. 1294). Cook argues if thesemi-automatic machines are down for a period of time, it woulddevastate its business. In addition to the semi-automatics supplying themajority of Cook\u2019s products sold, those products must be spun andshipped expeditiously. Customer orders are time sensitive, one-third ofall orders are shipped next day express (Tr. 1298). Cook experienceddown time before due to a flood. That down time was only two days, butthey lost accounts permanently (Tr. 1347-1348). Cook states it is acomplex process to acquire and keep commercial accounts. Cook operates its six semi-automatic machines approximately33,300 hours a year. Programming is only performed .093% of the time (31hours), while operating production is performed 99.907% of the time (Tr.1339). Cook also argues that during the time for retrofitting, itwould be forced to lay off a number of employees (Exh. C-19). Thespinning operators get parts from the shear and the punch – so all ofthose employees would not be able to produce parts to be run on thesemi-automatic machine. Blake testified, \u201cSo, the impact is significant,not just to the customer, not just to the Company, but also to theemployee.\u201d (Tr. 1349). Although the Secretary inquired about rentingspinning machines, there is no evidence such machines can be rented.Even assuming a semi-automatic machine could be rented, the connectionsfor all the rollers and mandrels would not be the same; every part wouldneed to be reprogrammed. It may also be necessary to retrofit the rentalmachine (Tr. 1465-1466). Blake has experience with retrofitting (Tr. 1340). Hecontacted retrofit companies specifically for adding a sliding orinterlock barrier guard to the semi-automatics (Tr. 1342). He contactedLeifeld, but they do not retrofit. Leifeld recommended a company calledPrism, a specialist in retrofitting (Tr. 1343, 1416). Cook engineer Kallenberger had discussions with Prism aboutadding an interlocked door to the semi-automatic machines. Prism\u2019s JeffRuhl stated that all the controls and electronics must be replaced (Tr.1786-1787). It costs more to replace old electronics than to replace thewhole thing (Tr. 1820). Cook received a quote from Prism in June 2005, for the knowncost of retrofits (Exh. C-18; Tr. 1370-1371, 1410, 1789). According toKallenberger, Ruhl put 14 items in the quote that are all required toput on the door (Tr. 1791). Retrofitting would include installing newmotors for the safety gate to work, changing the scales on the X and Zaxis. Blake testified he was told by Prism, \u201cThis control on yourmachine will not work. You can\u2019t retrofit that control. You have to puta new control. You have to put new software on the machine, you have toinstall new relays. You cannot do it any other way.\u201d (Tr. 1416). Prismtold Blake the machine would need to go back to their facility (Tr.1417). Prism\u2019s estimates to retrofit each of the six semi-automaticspinning machines was in excess of $100,000. The quote does not include(Exh. C-18): a) rebuilding and replacing servo valve systems b) doing any mechanical work other than what isdescribed in the quotation, i.e. slideway grinding c) headstock bearings, etc. d) any spindle drive work e) any seal replacement or cylinder work f) any hydraulic repairs, i.e. pump, motor, valves, etc. Kallenberger estimated the items listed above would cost anadditional $135,000 per machine (Tr. 1793-1794). Thus, the cost ofretrofitting each machine would be in excess of $250,000 and eachmachine would be out of operation while being retrofitted. After allthis expense, the retrofitted doors would still have to be open forloading and unloading and for lubrication. Kallenberger saw the new Leifeld spinning machines at aMarch 2005 trade show (Tr. 1763). A new spinning machine costs $400,000to $2 million, depending on the accessories. He also saw a Leifeldemployee program the machine. The employee did so by bypassing the doorand creating a new program with the door open (Tr. 1771). Kallenbergertestified, the Leifeld door would not allow normal work duties to beperformed (Tr. 1779). He discussed doors with a Leifeld engineer, whosaid the doors still need to be opened to allow the operator interactionduring the process. All the doors have bypasses on them to allow thatoperator interaction (Tr. 1764). Kallenberger stated, \u201cIf the door is inbypass, that door is totally eliminated and it is worthless. It\u2019s as ifthe door is not there at all\u201d (Tr. 1765). The door would have to be open95% of the time in order for Cook to have that operator interaction (Tr.1766). Dennis Blake also attended the March 2005 trade show, atwhich time he had a discussion with Leifeld regrading Leifeld doors andinterlocks (Tr. 1352-1353). He looked at the new model for 2005, andfrom his discussions with the Leifeld individuals, the machines cannotbe programmed with the doors closed (Tr. 1354). Nor can the pastelubrication be applied with the doors closed (Tr. 1355). The machine wasrun for Blake at the trade show (Tr. 1373). Blake calculated a substantial loss of sales for eachmachine being down (Exh. C-19; Tr. 1345). Blake\u2019s calculation includedrunning the other spinning machines on Saturday and Sunday withadditional shifts to make up for retrofitting down time of 16 weeks permachine (Tr. 1375, 1380). His calculations did not include theadditional cost of overtime or maintenance time (Tr. 1381). Cook has established it would be economically infeasible forit to retrofit the spinning machines with door guards. It would also bepointless, because the door guards would not protect employees whilethey performed the one activity identified as hazardous, i.e., applyingmutton paste to the spinning blank. Cook has established its affirmativedefense of infeasibility, both technological and economic. Item 1 of thecitation is vacated.FINDINGS OF FACT AND CONCLUSION OF LAW The foregoing decision constitutes the findings of fact andconclusions of law in accordance with Rule 52(a) of the Federal Rules ofCivil Procedure. ORDER Based upon the foregoing decision, it is ORDERED that: Item 1 of Citation No. 1, alleging a repeat violation of \u00a71910.212(a)(1), is vacated and no penalty is assessed. \/s\/KEN S. WELSCHJudge Date: June 19, 2006 “