The Heart of any Industrial Refrigeration System is the Compressor!
A major manufacturer of compressors conducted a yearlong tear-down study and analyzed their findings. One third of the compressors returned were so badly damaged that a cause of failure could not be determined, one third were found to have failed due to system problems, and one third were perfectly good compressors — there was nothing wrong with them!
Concentrating on the third where the cause of failure could be identified revealed that the reasons for failure had not changed from other year’s tear-down studies. Compressors were still being “killed” by system problems. System problems, if corrected, would have saved the compressor.
Broken valves, scored shafts, bent rods, overheating, single phasing, misapplication, improper operation, and not properly cleaning up a system when replacing the original compressor are evidence of system problems. Replacement compressors failed at a rate four times greater than that of original compressors.
The ability to disassemble a semi-hermetic compressor in the field allows the service technician to inspect the internal parts of the compressor and helps to determine the cause of failure. With a full hermetic compressor, identifying a cause of failure by inspecting internal parts is not possible; since the shell is not usually cut, open in the field.
In any case, when replacing a compressor, the technician should go over the entire system to locate the reason or reasons the compressor failed, keeping in mind the compressor was “killed.”
Compressors fail because of:
Slugging: a short-term return of a mass of liquid, consisting of refrigerant or oil, or as a mixture of both. The slug enters the cylinders of the compressor instead of super-heated vapor. Slugging almost always occurs on startup, but a very rapid change in system operating conditions can also cause slugging. A loud knocking noise heard at the compressor is evidence of slugging. The noise is produced by hydraulic compression – the compressor is attempting to compress a liquid - something it wasn’t designed to do. Extremely high pressure will form in a cylinder, a hole may be punched in the top of a piston, and more often, the suction, discharge, or both valves will be bent or broken. Cylinder head or valve plate gaskets will blow out at the internal partition between the high and low side.
If a compressor continues to run with this damage, the head where the failure occurred will run very hot compared to any other cylinders. Slugs can break connecting rods, even crankshafts. If a technician finds a compressor with broken or severely damaged internal parts, he or she should suspect slugging.
Refrigerant will condense in any cold part of a system during the off cycle. The coldest parts of the system are the evaporator and suction line. If allowed, they will collect the refrigerant and oil mixture. On the next start-up, this liquid mixture will return to the compressor as a slug. Using a “pump down” system of control can prevent this condition. Pump down control stops off cycle migration of refrigerant to any part of the system, including the compressor’s crankcase (which we will see later) can cause lack of proper lubrication. An over sized TXV will hunt under a light load and can cause a slug. It is best to slightly undersize rather than oversize a TXV. Poor suction line sizing or poor installation of the suction line also causes slugging. The design and installation of a suction line is one of the most critical parts of a whole refrigeration system.
Flood back/Flooded Starts: the continuous return of liquid refrigerant as droplets in the suction vapor instead of all super-heated vapor. This flood back washes oil off of bearing surfaces. Refrigerant is a lousy lubricator. All bearing surfaces will prematurely wear and overheating will result. On a compressor with a crankcase sight glass, flooding can be observed as constant foaming of the oil during run. Flood back is always due to a problem with the expansion device. If the expansion device is a TXV, check its installation. Is the bulb properly located and insulated? Is it the correct tonnage? Poor installation of the wrong TXV cannot be compensated for by valve adjustment.
Check super-heat. Too low of a super-heat allows more refrigerant to the low side of the system than the load will require. Saturated, not super-heated refrigerant will have liquid droplets in the vapor, gradually washing the oil off lubricated surfaces. If a cap tube or other fixed orifice expansion device is used, the system is critically charged. Those devices do not react much to load changes. Overcharging will increase the head pressure, which increases the flow rate through the expansion device until there is too much flow available for the heat transfer in the evaporator to boil the entire refrigerant off. Result – flood back.
Low load is a major problem of most refrigeration systems. The most carefully designed and installed system will, at some time, suffer from low load running conditions. The single best device to protect the compressor from slugging and flood back is the suction line accumulator. Accumulators are most often found on low temperature systems, but all refrigeration systems can benefit from this relatively low-cost protective device.
Flooded starts are the result of the oil in the crankcase-absorbing refrigerant during the off cycle. The amount absorbed will vary according to the temperature and pressure in the crankcase. The refrigerant and oil mixture will stratify, with the refrigerant at the bottom of the crankcase, where a bearing and/or oil pump intake is located. On startup, bearing lubrication will be marginal at best, and as the crankcase pressure drops, the refrigerant will boil, flashing to a gas, causing “foaming.” This foam clogs oil passages and may even enter the cylinders, resulting in slugging. All semi-hermetics use an oil pump to force oil to bearing surfaces. These oil pumps will not pump foam. Full hermetic compressors rely on splash lubrication and semi-hermetics also utilize some splash lubrication. Foam will not “splash.”
As previously noted, pump down helps keep excess refrigerant out of the crankcase. Crankcase heaters should be used, and if present, should be checked that they are not burned-out and “on” at the proper time. Flooded starts can be minimized on cap tube systems by having the proper charge in the system.
Loss or Lack of Lubrication: often the major cause of compressor failure, especially if we include flood back and flooded starts. Other causes of lack of lubrication could be as simple as not enough oil in the crankcase to begin with, to as complex as needing a double riser suction line.
During operation, oil should leave the crankcase, travel throughout the system, and return to the compressor at the same rate that it left. Oil will leave the crankcase of the compressor at an excessive rate because of foaming and flooding. Reasons for oil not returning include low refrigerant velocity, low loads, traps in the suction line, suction line piping errors, and short cycling. All can cause a lack of lubrication. The results of a lack of oil are overheating because of excess friction, and finally, a “seized-up” compressor. The internal parts of the compressor will show uniformly worn and scored bearing surfaces.
Contamination: found to be the major cause of replacement compressor failure. It bears repeating that only two things should be inside any refrigeration system: refrigerant and oil. Anything else is a contaminant! Air, moisture, non-condensable, chips, oxides, scale, brazing, or soldering flux, any “dirt” that gets into a system accidentally or through poor service techniques is a contaminant that will lead to sure and swift failure!
Air in a system will displace refrigerant in a condenser resulting in high head pressure and higher than normal temperatures. The temperature at the valves of a compressor is about 50°F hotter than the discharge line temperature. This high temperature will cause carbonization of oil on the discharge valves. A build up will occur, and in a short time, the valves will leak. Leaking valves cause higher temperatures, causing more build up – finally, total failure.
Moisture in a system will react with refrigerant to form acids. Acids can cause electrical failures by eating at the wires on electrical terminals, can etch parts, and can travel throughout the system eroding other components.
All contaminants must be eliminated. A thorough evacuation will eliminate air and moisture. Solid contaminants, chips and dirt, can be trapped in strainers, filters, or filter driers.
Most modern day filter driers have acid removal capability. Replace the filter drier, and if the compressor is semi-hermetic, change the oil, until an acid test proves the system is free from acid.
Make good installation and service practices a habit. Keeping contaminants out of a system in the first place is better than any procedure to remove them. Replacing a “burned-out” compressor presents the ultimate challenge in eliminating contaminants. It is best to assume that all burnouts are bad burnouts and proceed accordingly. The chemical reactions that take place when a compressor burns out will create a lot of acid, moisture, and solids in the form of carbonized oil and “soot.” The soot will coat the inside of the discharge line at least to the first elbow, and depending on the size of the system, some distance beyond the elbow. The suction line may even be coated six inches to a few feet from the compressor. It is best to cut out, remove, and replace these coated sections of line with clean, new tubing (Government regulations no longer allow cleaning up by pumping R-11 through systems). An over sized liquid line filter drier should be installed along with a suction line filter designed for high acid removal and having access fittings for checking pressure drop through the suction line filter. They should be installed to facilitate easy replacement. Deep or triple evacuation of the system is an absolute necessity to remove air and moisture before putting the system back into operation.
Of course, the technician has found the reason for the burnout, repaired the malfunction, and can now operate the system. After 48 to 72 hours of operation, the technician should return to check the pressure drop across the suction line filter and take an acid test. Depending on what is found, replace the suction line and liquid line filter driers until an acid test and pressure drop test shows no more contamination. The extra expense of proper cleanup of a system will be more than paid for by the long life of the much more expensive compressor.