Document jdzRnoEbYb7EBeKR4vDvzmB9

Monsanto row same A LOCAT J->. R. Pogue - St. Louis General Offices DATE ftUBjCCT CrENCt TO August 31, 1970 TURBINOL BULLETIN ^Cv"L." Curtis ?!.`..ccujrt.J""': " L. R. Stark - LSTAR NTJ/UFS/DAH Attached is the final copy for the Turbinol bulletin. The physical properties section has been changed as suggested by Texas Eastern to include: 1) Specific gravities and viscosities with variation in tempe rature. 2) Heat transfer data and electrical properties data are reported on a separate page. They should be incorporated in the bulletin under separate headings as originally planned. Hold printing until we get approval from Texas Eastern and Andy Smith, ASLE, to use Earl Farmer's comments. Thanks. D. R. Pogue lb P.S. TITLE: FACTS ABOUT FIRE RESISTANT LUBRICANTS SCAL GOO 15 5 HARTOLDMON0042303 Uj^-04 4* I) ^ cw> ;duX|^ ~^t> tjK i5_____fsi****- *- tJtc_______ kld^L. -fv t/< ^ ^ i^w.-r (d<Kvf_______ */v 71 f"6 ' TC-t^-- L,r< . fie J^)_______ (? `wAT^JLa ,^<. t-fr g o 1 0, . .V.^ tk'-fjn's-<,__ -sLi- C2-g<xx,^ CL^xJLlscs c^Aje^JuSi Mi X ^Sna cX/l G y -&lls:*c ^ 1-34 C IteH I 0 ) Iol ( Aaaw^) ' 0 '0 2:'- O. / O. is.._%_ ^/'SC.o^)TW^ ^A^LttLtS (.C-ULVU-C ('Xt-lejj') 0 2 ioF 0> looaF 4-o F \/Cx \i - Ji^. 4-^ o<SOF ' Ptou.a Po.'ot 3.7,j__________________ | ra <$_________________ S> y / <= O o "1 C"___________ 0 *F ' -Sfi-CANjt /U_Lk /UocLjii? (VsiL) (> /Oqf 4 12 o o V y -- (i<j. >*. c~ i/i --------------------------------------------------------------------------------- 4--wi-.vv^ C. C 1C C | v ^T------------------------------f--1--------------------------- _ ---------------------------- --------------------------------------- / /fc-^ O j i <_ eu./dt O'-- -----------------------------------------------------.---------k^-------------------------------------------------------1-------------- r ^ 77 T O.ooo 3T c ,5o SCAL COO 156 HARTOLDMON0042304 Tf-j."" <1?.- dr. iItu.( \rjcp o.o SI uL$z__ fid ^~tS 5C\JL*^ojlJL /u + ktie-u nfarr `Jll jrAviaJil__ La^-_______ SCAL 000157 ------ HARTOLDMON0042305 FIRE RESISTANT LUBRICANTS IN GAS TURBINES By Earl P. Farmer, Jr. Texas Eastern Transmission Corporation P. O. Box 1612 Shreveport, Louisiana 71102 Presented during annual meeting ofAmerican Society of Lubrication Engineers, Power Generation Council Technical Session, Chicago, Illinois on May 4-7, 1970. / Paper submitted for publication in future issue of Lubrication Engineer, official technical magazine of A. S. L. E. SCAL 000158 HARTOLDMON0042306 ABSTRACT The following paper is a brief resume of our operating experience using fire resistant lubricants in industrial type gas turbines. Basically, this article presents the reasons behind our decision to use the fire-resistant lubricants, a brief outline of the physical properties of the lubricants, and a general resume of our twelve years of operating experience while using the fire resistant lubri cant. ' SCAL 000159 HARTOLDMON0042307 FIRE RESISTANT LUBRICANTS IN GAS TURBINES The potential hazards inherent in using petroleum lubricating oils in high speed, high temperature gas turbine installations prompted Texas Eastern to investigate the applications of fire-resistant lubri cants in our gas turbine installations. From our experiences, several potentially disastrous lube oil fires were prevented by alert operators who happened to be in the right place when turbine failures occurred. In another case, a turbine failure caused a lube oil fire, resulting in a complete station outage. A load compressor failure was responsible for another lube oil fire, which resulted in damages far more critical in the station operation than the compressor failure. Still another hazard exists in the use of centrifugal pipe line compressors. These units require high-pressure seal oil systems with considerable external piping and controls. Even minor control and piping failures initiate emergency shut-down conditions, which, when using petroleum oil, necessitate the immediate shut-down of oil pumps tn prevent feeding of possible fires. This means the high speed rotating elements must coast down with no lubrication, resulting in the strong possibility of wiped bearings. If the bearings are wiped badly, additional internal damage to the rotating elements and seals can result. Thus a minor failure can lead to a major station outage. Advancing technology combined with rising operating costs led to the development and rapidly increasing use of automatic, remotely controlled, unmanned pipeline compressor stations. Obviously, this type of installation becomes most vulnerable to the potential hazards of lube oil fires. The obvious solution to these potential hazards is the use of fire resistant lubricants. Early in 1958, Texas Eastern initiated a program to investigate the possible application of fire-resistant lubricants in our large gas turbines. Many articles had been published covering years of development and laboratory testing of fire resistant fluids. However, these fluids had not been applied in lubricating systems in any major turbine, gas or steam. Fire resistant fluids were developed initially for hydraulic systems. As the problems of air compressor fouling and resulting fires became more prominent, fluids were developed that successfully eliminated this SCAL CC016G HARTOLDMON0042308 2 problem. With the progress and acceptance of the hydraulic and air compressor applications, the fluid manufacturers began to investigate potential applications in the steam turbine lubricating field. At this time, fire resistant fluids offered for turbine application were divided, on the basis of formulation, into five general groups: 1. Pure Phosphate Esters. 2. Phosphate Ester Based. 3. Halogenated Mixtures. 4. Water and Glycol Mixtures. 5. Water and Oil Emulsions. Of the five groups, the pure esters and the ester based fluids had shown the most promise, both in the laboratory and on paper; therefore, most of the small scale field tests recorded had been limited to these two groups. Our investigations were limited primarily to these two groups. For evaluating the various fluids for our specific application, the following items were considered most important: 1. Viscosity. 2. Fire Resistance. 3. Lubricity. 4. Oxidation Resistance. 5. Field Performance Tests. As to viscosity, the essential criteria was that the absolute viscosity of the fluid should match that of a 150 SSU @ 100F. petroleum turbine oil at the maximum operating oil temperature of 180F. Also, it is desirable for the viscosity curve to approximate the petroleum oil curve down to the lower operating temperature of 110F. Finally, the fluid must exhibit comparable viscosity stability after many hours of severe se rvice. The fire resistant and lubricity properties of both the esters and ester based compounds are satisfactory. The fire resistant properties of these fluids have been published numerous times and are readily available from any of the fluid manufacturers. Among all fluid properties, fire resistance is the most difficult to measure and the most controversial. There have been many attempts to standardize fire test methods, and, although several similar test types have emerged from these attempts, there is still essentially no standardization. The fire resistant rating of a fluid can not be established by one or two tests alone, but must be evaluated from all the possible tests that are usually run.' For example, the AIT (Autogenous Ignition Temperture) for our fire resistant fluid is 1130F. , while the AIT for Benzene is 1200F. Obviously, Benzene is not a fire resistant fluid, but actually a specialized fuel. The major desirable fire resistant prope rties are the high Autogenous Ignition Temperature SCAL 000161 HARTOLDMON0042309 3 in excess of 1100*F. and, because of their endothermic nature (verified by the Hot Manifold Test), the ability of the fluids to retard flame pro pagation. Laboratory testing for lubricity, wear, and sheer stability indicated that the lubricating properties of these fluids were adequate. These tests and tests on the fluid stability, foaming, oxidation, and corrosive characteristics of the pure esters and ester based fluids established that these fluids closely matched the desired qualities of a top grade petroleum turbine lubricant. Besides meeting all requirements for satisfactory lubrication of the gas turbine and natural gas compressor rotating elements, those synthetics must also function as a seal oil in the gas compressor. Additional tests were conducted to determine the solubility characteristics of the synthetics when exposed to 1, OOOpsi natural gas. Both esters reacted favorably in these tests. Because of incompatibility with some types of materials normally used in gaskets, packings, seals, O-ring, paints, and electrical insulations, all five groups of the fire resistant fluids presented problems. The natural rubbers, Buna N, and Neoprene materials are generally unsuitable and should be replaced with compatible materials such as butyl, Viton A, nylon, silicone and Teflon. Most paints are not compatible with these fluids, therefore, surfaces normally in contact with the fluid must remain unpainted. Considering the reluctance of turbine manufacturers to use unpainted oil reservoirs, the oxidation resistance of the fire resistant fluids becomes most important. It was decided that the ASTM oxidation test would give a good indication of what could be expected in an unpainted oil reservoir. Laboratory tests determined that some of the ester based fluids could pass the required 1, 000 hour ASTM oxidation test required of the premium petroleum turbine oils. One specific fluid was on specification after 7, 000 hours. After reviewing all the literature and data available on the fire resistant fluids, it was decided that one of the phosphate ester based fluids met all our requirements. The physical properties of this fluid are tabulated in Table I, along with the properties of a typical petroleum oil and another fire resistant fluid, which will be discussed later. SCAL 000162 HARTOLDMON0042310 4 TABLE I PHYSICAL PROPERTIES OF TURBINE LUBRICANTS Prope rty Petroleum Oil Phosphate Este r Base Chlorinated Polyphenyl Ester Base Specific Gravity @ 77*F. @ 210*F. Flash Point, COC, (F. ) Fire Point, (*F. ) Autogenous Ignition Temp., Neut. No. (mgKOH/gr) Pour Point, (*F. ) Specific Heat, (BTU/#-*F. ) ' @ 100-F. @ 150*F. @ 200F. . 871 . 830 340 455 (*F. ) 700 0.14 0 . 47 . 50 . 53 1. 40 1. 33 385 660 1160 0. 07 15 . 319 . 340 . 360 1. 335 1. 268 360 560 1130 0. 05 0 . 304 . 318 . 327 Viscosity, SSU @ 100F. @ 130F. @ 210F. 155 86 43 166 70. 5 37 108 65 37.1 Absolute Viscosity Centipoee @ 100F. @ 130F. @ 210F. Oxidation Test 28. 9 14. 6 4. 34 2000 HR 49.1 18. 2 4. 27 3000 HR 30 14 4. 3 3000 HR In 1958, Texas Eastern pioneered the use of fire-resistant lubricants in large, industrial gas turbines when we installed four 13,400 HP @ 80F. units with the phosphate ester based fluid as the only lubricant. These units were installed in four isolated, unattended, remotely controlled stations that are used primarily for increasing gas deliveries during the winter months. These simple cycle, single shaft, two stage gas turbines operate at 4860 RPM and drive a two stage 30 inch by 30 inch pipe line booster compressor. Both the turbine and the load compressor are served by a common lube oil system built around the 2000 gallon turbine reservoir. The oil is pumped from the reservoir through a cooler and through a series filter system consisting of standard magnetic, knife edge, and full flow filters. The oil then flows from the filters into the various bearing systems and is returned to the reservoir. Basically, the oil systems operate with a normal bearing oil supply temperature between 110*F. and 130F. and a bearing drain temperature of175*F. SCAL 00.0163 HARTOLDMON0042311 5 The load compressor seal oil system is supplied by a positive dis placement pump taking suction from the turbine lube oil header and discharging through a small filter bank to the seal oil control system. The seal oil is delivered at about 1000 psi to the compressor seals and the emergency seal oil tank. The return seal oil from the compressor is passed through a separator-accumulator to remove the entrained natural gas before returning to the reservoir. Low pressure control oil is taken from the main lube oil header ahead of the system filters and passed through a small bank of micronic filters before being fed to the control oil system. Because of the high pour point and high viscosity at low temperatures, it is necessary to insulate outside storage tanks and related oil transfer piping. An immersion he ate r in conjunction with a small circulating transfer pump and a circulating loop maintains the storage system oil at a pumping te'mperature. These precautions are essential in our systems because make-up oil is automatically added. At some of our more recent stations, which are remotely controlled outdoors installations, reservoir he ate rs and heat tracing of oil line s was required to maintain the oil system at pumping conditions during down time. Initial start-up of the first of these four units was a time anxiety over numerous unknowns. The turbine and gas compressor were the largest ever applie d to pipe line operation and the compressor was a new de sign. The use of the fire resistant fluid as the only lubricant was a radical departure. Considering the lubricant as an unknown element, certain functions were observed closely to determine the action of the fire resistant lube oil in the system. Clear plastic cover plates were installed in the re se rvoir taps to obse rve any foaming tendencie s. Further, oil temperatures and pressures were recorded frequently; bearing temperature diffe rentials and vibrations we re recorded at two to three minute inte rvals. Filter pressure drops were frequently checked for signs of filter clogging. Incidents noted during start-up were control malfunctions, slight scoring of a major bearing, several worm gear failures, and hot bearings first at one location, then another. Each time, the initial blame was placed on the lubricant. Howeve r, in each case, other causes we re established for each incident and when time came for start-up of the third and fourth units, the fire resistant fluid was no longer considered the unknown. In reviewing the start-up data, it was noted that all the lube oil systems operated as calculated. Oil foaming in the reservoirs was found to be less than normally experienced with petroleum oils. Filters reacted normally with a 4 to 6 psi pressure drop and normal build up rates. Seal and control oil systems, pumps and heat exchange equipment all reacted as designed. SCAL COO 164 HARTOLDMON0042312 6 The temperature rises across individual bearings were observed closely as an indication of bearing action with the synthetic oil. Data indicated that the bearing temperature rises were some 2*F. to 7'F. higher using the synthetic fluid as compared to a petroleum oil. Vibration levels appeared to be lower with the synthetic fluid. The quality of the synthetic oil was checked carefully twice per week during start-up operations. These quality control tests indicated that the oil was completely stable and remained on specifications. After the fourth station was completed and placed into service, all indications were that the fire resistant fluid had not all criteria for use in large industrial gas turbines. To date, these four stations, with a total of 53, 600 HP installed have operated for 86, 000 hours without any problems associated with the fire resistant fluid. One of these units has operated in excess of 31, 000 hours. Normally, these units are operated only during the severe winter months when our system is at its peak loads and are idle during the off load seasons. This intermittent service with long downtimes can be more severe on the oil than continual operation. This fire resistant fluid has proven excellent oxidation and thermal stability. To date, we have not experienced a single oil change caused by excessive oxidation in these four original installations. Today, we have twenty industrial type gas turbines in service using the fire resistant fluid in foe oil systems. These twenty units, with a total of 323, 800 HP have operated in excess of 410, 000 hours without any failures attributed to the fire resistant fluid. These units which have been installed during the past twelve years, consist of nine 19, 800 HP units, three 18, 500 HP units, four 13, 400 HP units, three 8, 000 HP units, and one 12, 500 HP unit. Two of these units have logged over 67, 000 hours of service each and two other units have operated in excess of 50, 000 hours each. As noted in Table I, the physical properties of a second fire resistant fluid were given. This fluid, our present gas turbine lubricant standard, is formulated with a chlorinated polyphenyl and a phosphate ester. Phosphate ester exhibit excellent lubricating properties in film lubrication, good viscosity-temperature characteristics and excellent fire-resistivity, the chlorinated polyphenyls exhibit poor viscosity-temperature character istics, excellent fire resistivity, and excellent EPL or boundry lubrication. Therefore, the proper formulation of these two fluids has created a superior lubricant exhibiting the desired properties of both fluids. Since this formulation was introduced in the early sixties, we have phased out the original phosphate ester based fluid and now use this formulation exclusively. SCAL 000165 HARTOLDMON0042313 7 Our operating experience over the past twelve years has proven the fire resistant fluid used in our gi.s turbines to be extremely stable, both thermally and oxidatively. None of the units using this lubricant have experienced an oil change because of excessive oxidation. Periodic tests have indicated little change in the neutralization number. A coupling guard leakage at one station provided a practical test of the oil's fire resistance by spraying oil against a 700eF. exhaust housing with no fire resulting. Modifications and repair.s by welding on oil piping within the oil tanks further demonstrated fire resistance. Because of the added safety factor and the proven reliability and stability of the fire resistant fluid, it has become the standard lubri cant in all our industrial type gas turbine installations placed in service since 1958. SCAL 000166 HARTOLDMON0042314 FIRE RESISTANCE PROPERTIES High-Pressure Spray Test (AMS 3150-C anti Federal Test Method Std. 791) Low Pressure Spray Test Hot Manifold Test (Federal Test Method Std. 791) 130OF. Tube 950f. Tube Molten Metal Ignition (Aluminum alloy @ 1200F.) Localized flashing In con tact with flame but self extinguishing. Slight Increase in flame intensity. Intermittent flashing on tube. Does not propogate flame to pan below. Does not flash. Does not ignite spontaneously. Flashes intermittently with ignition source, but selfextingulshes. oooifa'3 SCA'- HARTOLDMON0042315 LUBRICATION PROPERTIES Four-Ball Wear Test (mm scar diameter) 620 rpm, 167 F., 1 hr. Steel/steel 10 kg 40 kg Steel/bronze 10 kg Timken Test Passing Load, lbs. Falex Load Jaw Load, lbs. Torque* lb. inches Turblnol 153 0.23 0.53 0.92 40 Typical Premium R&O Turbine Oil -0.6 to 1.0 2.0 to 3.0 15 SCAl CCClfcS HARTOLDMON0042316 . ALCOR DEPOSITION TEST 48 hours, 250F. fluid, 1000 ml/min. air flow @ 400-475F tube temperature. Turblnol 153 Petroleum Oil Total Deposit Rating 10.5 10.5 Filter, mg. 10 13 Acidity, TAN Change +0.03 + 0.02 Viscosity % Change, 100F. 15.6 5.8 SCAL 000169 HARTOLDMON0042317 .STABILITY AND CORROSION PROPERTIES Turbinol 153 ____ Typical Premium RScO Turbine Oil ASTM Rust Te^t D-665 l40uF. DistilledWater Pass ASTM Emulsion Test P-l401 ml Emulsion after 30 minutes 0 ASTM Foam Test D-892 Sequence 1 Sequence 2 Sequence 3 Pass Pass Pass ASTM Turbine Oil Stability Test P-9^3 (Modlfleda) Hrs. to reach 2.0 mgKOH/g >2000 Hydrolytic Stability Tes (MILH 19^57 - Beverage B ottle Test) 200F. water, 48 hrs. Cu strip Pass Oxidation/Corrosion Test lbfci hrs., 250OF., air Metal Attack, mg/cm2 Mg Cu Fe Cd -0.01 -0.01 -0.04 0.0 -0.01 Viscosity % Change, 100F. % Change, 210F. ++08.. 56 Acidity, TAN Change +0.02 Pass 0 Pass Pass Pass 2000 Pass aBecause- of the higher specific gravity of these fluids, water used in the test would not settle to the bottom of the test cylinder. Hence, oxygen stream was saturated with water before being bubbled through the test fluid. 0001"*0 SCAL HARTOLDMON0042318 PHYSICAL PROPERTIES Appearance Clear, amber colored oily liquid Specific Gravity 25/25C ' 1.34 Neutralization Number (Acidity)(Mg KOH/gm) 0.02-0.1 Moisture (Max.) 0.15/6 Viscosity, Saybolts @ 210F. @ 100F. @ 40F. (see p.2) 37.1 108 2,160 Vapor Pressure (mm Hg @ 200F'. ) 0.025 Pour Point 0F. Boiling Point 625F. Secant Bulk Modulus (psi) @ 100F. 412,000 Coefficient of Therroal Expansion cc/cc/F. 0.00037 Specific Heat BTU/lb./F. @77F. 0.30 Thermal Conductivity BTU/hr./ft./ O.O58 F. Solubility In Water Negligible SCAL 000171 HARTOLDMON0042319 o<iiL.OZjuj a. T t^*rC-____ Qtfu o r^-c*. I?h < < ^-k' LA. ^ Aaj t} Jp (Ls\ ^J~r\ e - ?~W *^-0 Iju. L ig -fo /5 irv 'K't,sy\4 Pjgjf-ro Uu^vv. r U*. tf--LX--< C-'QP _; Vf U^v. t>iJ, c c^> v ecU*. u_ vj,.->~> rJ>~t ~f^\ ^v\ fu SCAL CC0172 HARTOLDMON0042320 f FIRE RESISTANCE PROPERTIES Monsanto Screening Tests . jsssnsx. furU^I 165 Flash i Firei I 9 Pressure Spray Test (AMS 3150-C and Federal Test Method Std. 791) r /rf Localized! flashing in contact with flame UUL1 Intenrti flasht hg, but self-e ictinguishing. s^c4 i Js Hot Manifold Test (Federal Test Method Std.791) Q 1300F Tube [ 950eF Tube Intermittent flashing on tube. Does not flame to pan below. Does not flash. e Molten Metal Ignition (Aluminum alloy @ 1200F) .spatdt . I .w, I f .v . tt^nt shing ittent |ig, but does ppagate to ?eiLow. Does flash tube. Does not lignite spon taneously. Flashes intermijttJently with ignitioi source. Flashes^ self extinguishing until irost of fluid \laporized; then bujrns to completion -continued- SCAL 000173 HARTOLDMON0042321 FIRE RESISTANCE PROPERTIES -continued" 10 (*) Pydraul\L35 meets the U.S. Bureau of Mines' fire-test requirements. It dopfinot have official approval since required field tests in mining machinery have not been performed. Pydraul 135 ijff^iisted with Underwriters' Laboratories, Inc. SCAL COOHA HARTOLDMON0042322 f' 7, ------- ----------- - ----- f - ~U^M r --j^<4Q~__v J~ fg f ft 1 e 0 _) A^b COMPATIBILITY CHARACTERISTICS OF PPRMftL 135 11 Metals Pydraul l^sis compatible wi>-'efll common metals of construction. See Corrosionsand Oxidatj^ffcest results for aluminum, magnesium, copper^, iron, ar^cj^wlurn/iron specimens, page 8. Packinc Seals, etc. /"A Viton o-ring's u-cups, chevron packings, etc. ^re recommended where dynamic sealir is required. Silicone, nylon and Teflonelastomers are equally com^iatible with Pydraul 135. for dynamic sdals but these types may exhibit, inadequate mechanical properti'ba^^For static seals and gaskets,\butyl, Viton, asbestos, and Leather are generally satisfactory. Oil-type seals (Buna N and neoprene) are sometimes used satisfactorily ak^ static seals, but th^s practice is not re commended . Hoses Butyl-lined hoses are widely iteed/a'nd are recommendedNyl/on and reinforced nylon_hose has proye)/hig'bly satisfactory wit^f/ydraul 135 and MCS 153. CSertain ethylene/propylene/terpolymers (EPT or EPR types) are also\ satis factory l^se-Ilnep/materia Is. Viton- and Teflon-lined hoses aVe completely compatible but may be deficient in flexibility for some applicationsA\Oi1-type Buna N hoses are sometimes used but are\j\6t/ recommended^ Filters Pydraul 135 can baXfiltered thYough^ to 10 micron paper-type return-line filterA without separatioiy or other adverse \effects. For fluid cleanup and/or ref lama tion, Puller 1 s earoh or other i active type ^ttlapulgus earth filters are\recommended\ Thr^e to five micron .alze filoration/is recommendefv^Binders \^ed wTth oil-type paper filters, may be dissolved, however, and ccyred phenolic ay epoxy binders are recommended, Paint Pydraul/T35 is most commonly used in systems unpa inted^or/ > inferior. \ Epoxy/paints ake compatriBTev^here pointing is requir-e^. L -continued000 SC*L HARTOLDMON0042323 COMPATIBILITY' yC-POtJMV&C /V ^^Solubility with Other Fluids) b i s~\ y\ /> s\ ^ M ^ <V>V v\_o\ Fydrau-1 135 12 f I I fresh py^uiyi35 I ^ /, SCAL 000176 HARTOLDMON0042324 HEAT TRANSFER PROPERTIES HARTOLDMON0042325 flEEl E53UU P--* EXES m CTTI B153 ccHrAi^pvi nomnifl or oowmcuL toibihi tupiw iA P P E N D IX n > 0 Data via takan (roa fubllihid literature except r Four-Bill Wear data, which vaa obtained at Honaaato Labontorlia. o o (E) Leaded industrial sear oil, 4450 lbat/Lnv o 0 03 HARTOLDMON0042326