Document J6QkV2gGQdOpq5LDe43pKvbX

800 CHAPTER 50 Rg. 37 .... Viscosity of Refrigerant 13-Oil Solutions at Low-Side Conditions* parts approaches zero. Such low speeds may occur during start-up, or upon reversal of reciprocating motion. Bearing pressures and oil viscosities are also involved, and the boundary condition may prevail at relatively high speeds when bearings carry heavy loads and/or viscosities are so low that the lubricating fluid tends to be squeezed out of the bearing clearance.- It is generally accepted that adequate lubrication under these circumstances depends on existence of a nonfluid film which adheres strongly and permanently to bearing surface despite squeezing-out of the oil. Boundary lubrication has not been a serious problem in most refrigerant systems. As a rule, fluid lubrication can be depended, on during`operation, except under some starting conditions.14 Commercial`brands of- refrigeration oils appar ently possess a natural film-forming ability which is adequate to meet the service conditions. The natural film-forming property is believed to depend on slightly reactive constituents left in the oil after refining. These reactive substances supposedly act on bearing metal surfaces to form tenacious films which, although invisible to the eye, are sufficient to prevent metal to metal contact. The oils which are most stable against attack by hot re frigerant gaaww are not necessarily the best lubricants. Tests have shown that removal of the colored, more reactive con stituents from a pale yellow refrigeration oil leaves a white oil which then resists attack for longer periods of time when heated with metals and Refrigerant 12. However, the .reac tive constituents, by chemically changing and combining with tiie metal in the bearing surfaces, may actually provide a good'boundary lubrication film. The experiments of Thorpe and Larsen,1* for example, have shown that with a white'oil, boundary lubrication is readily driven to the dry friction state, but not with a less refined 10 W mineral oiL< The refrigerant itself may sometimes aid in boundary lubri cation as indicated by recent studies with gaseous refrigerants at high temperatures.** When run very hot in the absence of any other lubricant. Refrigerant 12 apparently reacts with steel surfaces and forms a lubricating film/ However, Re frigerant 22 does not form such a film. Some compressor designers use special processes and oil additwes, or both, as a supplement to the natural film-form ing property of the oiL Steel pistons, shafts and wrist pins have been treated by phosphating .processes (Parkerizing or Granodizing) which form iron phosphate films on the metallic surfaces, and thus provide an' initial base for a boundary lubri cation film. In some instances a; phosphate ester (tricresyl . 1965 Guide.Arid Data.Boolc phosphate)' which has been found to improve boundary lubrication of heavily loaded hearings in'an air environment, has also been injected into the bearings of refrigeration com pressors and added to the refrigeration oil itself. At present, there is no simple test for boundary lubrication in the presence of refrigerants. Divers** has reported the use of a Falez machine to evaluate refrigeration oils by comparing wear rates, but this machine operates in air and,.as Divers points out, the data obtained may not be too significant when applied to reciprocating refrigeration compressors. Elaborate tests mang radioactive,tracers in refrigerant systems have been reported as giving useful results.11'11 -- Oxidation Resistance In the early days of refrigeration oil development, the chemical stability of lubricating oils was usually judged by an oxidation resistance test. In addition, it was assumed that an oil which resisted air oxidation would show less tendency toward copper plating in a refrigerating system.** In recent years the value of oxidation tests as a measure of the chemical stability of refrigeration oils has been seriously questioned.4* Oxygen, at least so far as significant quantities are concerned, is normally excluded-'from sealed refrigerant systems. Refrigerant systems must be rid of air and all other noncondensable gy* if they are to function properly (see Chapter 47). None of the standard oxidation resistance tests have won universal approval for qualifying refrigeration oils. However, tests of this type are still included in many refrigeration ofl specifications.** Oxidation resistance provides another due to oil quality although first consideration' in this respect is gener ally directed toward specific procedures (refrigerant resistance tests) which more closely duplicate-the service environment' Moreover,. air oxidizing conditions may actually be .en countered .during some manufacturing- operations. .A small amount of oil must be used during assembly and test of pumps and compressors, and-it is not always practicable to com pletely.remove .this oil before the system is dehydrated, i. < j If--the, subsequent dehydration process is one. in which compressors or systems are heated in a stream of hot, dry air, as is frequently the case, the hot oxidizing conditions may change the residual oil to a gummy, varnish-like coating on bearings and valves. This gummy film may then lead to stuck pumps, overloaded motors, and other.operating difficulties. For this reason, a high resistance to oxidation may.be a valu- Rg. 38 Viscosity of-Refrigerant 13B1 -OH Solutions at Low-Side Conditions* ,k Lubriairifc in RefrigerantSystems 80.1; able attribute of any oil which must pass through an air de hydration oven: '' Much work has been done in refining and treating.base oils as well -as in attempting to develop chemical additives (oxidation inhibitors) which will minimize or possibly prevent gum and varnish formation. Refrigeration oils containing oxidation inhibitors, Lei, antioxidants, are commercially available. Under the extreme conditions' imposed by high tem perature dehydration processes; the usual oxidation inhibitor may not be wholly effective and the compressor manufacturer can discover only by trial if benefit is obtained: Once a refrigerant system is sealed and oxygen excluded, no further beneficial effect of oxidation inhibitors is to be ex pected. Refrigerant resistance, rather than oxidation resist ance, is then' important. Some commercially used antioxidants have bee" found to decrease the chemical stability- in a refrigerant environment, while others seem to have no effect one way- or the other. Rizzuti, Staffing and Jenkins4* have dpyptwt a series of tests in which the refrigerant resistance of four naphthene base oils,, with and without' a certain unspedfied additive, was measured. According to theirlabora- tory data, the use of the additive improved the chemical stability, of the oil-refrigerant'mixture, particularly if the oil had been severely refined. A number of claims for chemical additives which are said-to improve the stability of. oil- refrigerant' systems in service have appeared in patents from titnp to time.41 However, the extent to winch such inhibitors may be in actual -use is unknown. Refrigerant Resistance - : Refrigerant resistance tests are sometimes referred to as oU stability tests.'AH'published tests have been relatively simple/ at least in a physical sense. In a typical test0 a small predetermined quantity of oil is put into a strong glass tube or a pressure vessel (bomb) which is then dehydrated thoroughly, charged with a specified amount'of refrigerant, and sealed. The ratio of oil and liquid refrigerant used is usually 1:1, but this ratio does not appear to be critical. Quite often, the tubes or bombs .will also contain one or more of the common metals, iron, copper and aluminum, usually in the form of small strips of metal or wire coils which are so placed as to bepartially or completely covered by the liquid during the test. After sealing, the tubes or bombs are heated at some specified temperature for a time long enough to produce observable results. tcaaucTvw.r Rg. 39 ..; ..Viscosity of Refrigerant 22-Naphthenk Oil Solutions at Low-Side Conditions1 Rg. 40 .... Viscosity of Refrigerant 502-Naphthenic-Oil Solutions at Low-Side Conditions* In making such a test the observer usually looks for symp toms of the amft type found in operating systems which have given trouble in toe field. Various developments, such as sludge formation, carbon deposits on valves and pistons, gumming and copper plating of bearing surfaces, acid forma tion, corrosion and damage to electrical insulation have been well publicized in the literature; and at one time or another have been attributed to oil decomposition (see Chapter 48). For this reason, refrigerant resistance tests are frequently, evaluated by noting such effects as the degree of discoloration or of the oil; the time required for acid decom position products to appear; and if metals have been included in toe tests, to observe the extent'of copper and steel corrosion and whether copper plating is seen on the stceL - Chemical Reactions in Oil-Refrigerant Systems . It has been' known for many years that'destructive chemical changes can occur inside refrigerating systems if the refriger ant, the'lubricating oil, and all other' materials used in con structing the systems are not carefully selected and controlled. Visible evidence of such chemical reactivity has appeared in several forms, one early wgn being the so-called copper plating that is spmt.im observed on iron or steel parte inside com pressors when a tear-down examination is made after labora tory or field service testa. These copper films, even though light and without apparent-effect on the operation of the compressor at toe time of examination, are symptoms of chem ical changes that have taken place within the system and should not be ignored. Severe cases of copper plating can fre quently be traced to contaminated systems (see Chapter 48); however, toe lubricating oil may sometimes be involved, as was first recognized by McGovern1* in 1939. The mafthiwi-wn through which copper is dissolved from parts vntAining that metal and transferred to steel surfaces in refrigerant-oil systems has been studied -by Shaw and Brandon, Steinle and Seeman,44' and Spauschua. According to Steinle and Seeman, copper plating should not be produced with Refrigerant 12-oil systems at test temperatures up to about 212 F if the lubricating oil satisfies the following re quirements: (a) oil resin content, maximum 0.3 percent, (b) sulfur content, maximum 0.2 percent, (c) aniline point, minimum 212 F; (d) refrigerant resistance test (in low pres sure vapor at 482 F), minimum 96 hr. Spauschus measured the copper plating that occurred when a highly refined water-