With few exceptions, all compressors that are lubricated with oil will release oil into the gas stream. The rate of release can be as small as parts of oil per million parts of refrigerant for direct excursion hermetic centrifugal compressors and as much as several percent for screw compressors. Oil release rates are usually expressed in terms of lbm of oil discharged per lbm of refrigerant compressed or in mass percent of oil in the release gas.
Oil in compressor release gas is in two forms: fine oil droplets (mist) in the gas stream; and liquid oil pushed by the gas velocity, crawling along the tube walls. Oil flows from the compressor with the release gas by the oil separator (if equipped and always less than 100% efficient), and into the condenser. The liquid leaving the condenser consists mostly of refrigerant with some amount of dissolved oil (assuming that the oil is miscible in the refrigerant). The oil content in the liquid refrigerant at this point is the same as the oil release rate of the compressor/separator.
The liquid oil-containing refrigerant flows by the expansion valve and into the evaporator. In the evaporator, the refrigerant boils off delivering its refrigerating effect. The oil, however, does not evaporate as its boiling temperature is very high relative to the temperatures existing in the evaporator. In the absence of an oil return system, oil will continue to collect and concentrate in the evaporator which will rule to two negative consequences: heat move in the evaporator will be progressively degraded and the compressor will ultimately run out of oil shutting it down. Hence, an effective oil return system is basic.
Refrigerant and Oil Mass Flow Balance in a Flooded Evaporator
Consider the evaporator of an operating water chiller. Oil is arriving at a certain rate, specifically: the oil release rate of the compressor less the removal rate of the oil separator, if equipped. For illustration purposes, assume the mass arrival rate in the evaporator to be 2 lb of oil along with 1000 lb of refrigerant liquid in one hour. The compressor/separator has an oil release rate of 0.2%, i.e. mass of oil per mass of refrigerant compressed expressed as a percent. This would be a good release rate for a screw compressor/separator.
Oil is also leaving the evaporator via the oil return system. The amount of oil leaving via the oil return system is a function of the liquid removal rate and the concentration of oil in that liquid. Let us assume that the oil return system draws 50 lbs of refrigerant/oil combination from the evaporator per hour. If the concentration of oil in the evaporator liquid is say 2%, then the oil returned is 1 lb per hour. Since this leaving rate is less than the arrival rate, oil will further build up in the evaporator and the oil concentration will rise. Under the conditions stated above the oil concentration in the evaporator will rise to and stabilize at 4%.
Four percent is unacceptably high. There are two things we can do to reduce this concentration. The first is that we can increase the oil return liquid withdrawal rate. If we double the oil return flow rate to 100 lbs/hr and the oil concentration is 2%, the oil arrival and removal rates will be equal at 2 lbs/hr and the concentration will be stable at 2%. Or, we can decline the concentration of oil in the liquid entering the evaporator (perhaps by installing a more efficient oil separator). These two possibilities also suggest the cause of unacceptably high oil concentrations in evaporators and of chiller shutdowns due to loss of oil. The first is a failure of the compressor (leaking o-rings, missing plugs, etc.) and/or of the oil separator that causes unusually and unacceptably high oil release rates. The second is a failure of the oil return system, such as plugged lines, inadequate capacity of a pump, or inadequate driving pressure difference for an eductor. Considering the above, it should be obvious that the more effective improvement to any oil return system is to reduce the oil arrival rate; i.e. reduce the compressor oil release rate and/or enhance the efficiency of the oil separator.
Oil Inventory in the Evaporator
If you were to do an oil mass balance examination on an operating flooded evaporator as described above, by measuring liquid line flow and concentration and oil return line flow and concentration, you might in addition experimentally find more oil in the evaporator than you expect. The discussion which follows offers a possible explanation. The point of the discussion is that the design of the evaporator itself and the location of the oil return pickup can have a major impact on the success or failure of an oil recovery system. This is applicable because it can average that replacing a poorly operating oil return system of one kind with another (e.g. pump with eductor) may not fix the problem, the real problem being that the oil return pickup point is poorly located.
Consider a one pass flooded evaporator. Warm water enters tubes at one end and exits as chilled water at the other end. Refrigerant liquid surrounds the tubes and is introduced by a pipe at the cold water end of the shell. Liquid refrigerant is withdrawn from the shell by the oil return system from the middle of the shell (or worse, from the cold end by the liquid inlet). As above, the refrigerant entering the evaporator contains 0.2% oil, and refrigerant is drawn by the oil return system at a rate of 100 lbs/hr and the concentration at the point of withdrawal is 2%. The arrival and removal rates are identical at 2 lbs per hour. If the evaporator refrigerant charge were 100 lbs, one would be tempted to conclude that the evaporator contained 2 lbs of oil. in addition, if you were to measure the oil concentration at the ends of the shell, you might find that the concentration was 10% at the warm end and 0.2% at the cold end. Why would this be? The answer is that most of the evaporation of liquid refrigerant takes place at the warm end of the shell where the temperature difference between water and refrigerant is the greatest. Gravity will see to it that this liquid is replaced with liquid from a higher elevation: liquid at the cold end of the shell which is evaporating, but slowly. Hence, there will be a slow axial flow of liquid refrigerant from the cold end of the shell to the warm end and it will take oil with it that will not return while the chiller operates. But that oil will not evaporate at the warm end nor will it be picked up by the oil return system which draws from the middle of the shell. Hence, oil will tend to concentrate in a place where the oil return system does not pick it up. And where the oil return system does pick up liquid, that liquid will not contain much oil. This will consequence in a “stored inventory” of oil in the evaporator which can be substantial. So it is important to know where in the evaporator the oil tends to concentrate and to draw return liquid from that point. That location varies by design of the evaporator and any associated internal liquid dispensing system.