Preventing Repeat Compressor Failures Part 2

May 1, 2003
by Harry Hartigan In direct expansion (DX) cooling systems, the load on the cooling coil doesnt match the compressors full-load capacity except on a design

by Harry Hartigan

In direct expansion (DX) cooling systems, the load on the cooling coil doesn’t match the compressor’s full-load capacity except on a design day. Most of the time, capacity exceeds required cooling and requires some control to keep the coil from freezing.

Let’s pick up where we left off last month (CB, April 2003, p. 90) in our discussion of capacity control:

Capacity Control

The Brainerd Compressor website (, states that compressors aren’t designed to run unloaded for extended periods. It also states that, in general, the more capacity control a compressor has and the more it runs unloaded, the shorter the compressor’s life will be.

This statement is likely not true in all cases, but I'm sure it was made to make the point that as a compressor unloads, you have to pay attention to oil return and to the motor winding temperature.

The longer it remains unloaded, the more of a problem oil return may be. Depending on the degree of unloading, the motor may run in excess of its maximum recommended operating temperature.

Always consult with the manufacturer if additional unloading is considered when that falls outside the manufacturer's published data.

All compressors continuously pump a small amount of oil out to the system during normal operation. Any oil pumped out must return to the compressor, or the oil level in the crankcase will fluctuate.

If the oil level drops too far, the oil pump will lose suction and the compressor will operate without lubrication until the oil failure switch trips. Do this a few times and you'll shorten the life of the compressor significantly.

Of equal importance is the rate at which the oil returns — too much oil returning at one time (commonly referred to as an "oil slug") results in broken valves and pistons.

Other concerns as the compressor unloads are as follows:

  • Loss of control over
    superheat as the TXV attempts to accurately control at less than full load. This is especially true if the TXV was selected based on nominal tonnage rather than using the extended capacity tables available from the manufacturer.
  • Liquid return to the compressor crankcase. When refrigerant liquid flows into the oil in the crankcase, it dilutes the oil and reduces its lubricating ability.

In addition, the liquid refrigerant cools the oil, which makes it more susceptible to refrigerant migration on the off cycle. The colder the oil, the more refrigerant will migrate to the crankcase given the chance.

Depending on the quantity of refrigerant in the crankcase, damage on startup can range from premature bearing wear to broken pistons and rods caused by oil slugging.

  • Inadequate motor cooling.
  • High discharge temperature that leads to a loss of the lubricating qualities of the oil within the cylinders.

Piping Design, Gas Velocity, and Oil Return

Some general guidelines for suction and discharge gas piping are as follows:

  • Refrigerant piping must be designed and installed so that it promotes the return of oil to the compressor.
  • The piping must be designed to ensure oil flows up vertical risers.
  • Oil cannot be permitted to accumulate anywhere except in the compressor crankcase.
  • The piping system must be analyzed under all expected operating conditions.

Each system is unique, (although possibly similar to other systems) and therefore must be consciously analyzed and approved by whoever is
responsible for performance and
compressor life.

Additional information on current line sizing procedures is available from the following sources:

  • The Trane Co. Refrigeration Manual
  • The ASHRAE Refrigeration Handbook (1998 & 2002 editions)
  • Carrier Corp.’s E-20 refrigerant line sizing software.

Compressor Condemning Procedure

If you suspect that a compressor has failed, it’s imperative to follow a methodical diagnosis procedure.
Following is the procedure we require our technicians to use:

  1. Examine the system overall. Don’t assume it’s installed correctly, wired correctly, or controlled correctly.
  2. Speak with the owner and find out if there have been any unusual events in the preceding days, such as power failures, brownouts, replacement of other system components, etc.
  3. Remove the refrigerant from the compressor (or the whole system if the service valves won’t hold).
  4. Open the compressor for inspection as far as you can while it’s in place. You’ll see for sure if it’s damaged internally.You may see some indication of what may have caused or contributed to the failure. If the crankcase isn’t accessible in the field, open it at the shop to ensure you have the full story.
  5. Provide a fixed price to the customer for the replacement of the compressor and associated components. Make it clear in the proposal that you’ve included time (4, 8, 12 hours, etc.) to investigate the cause of failure.

    Explain that when the cause(s) of failure are found, it may result in additional work that’s not part of the proposal.
  6. Order the compressor (after the customer has signed off on the scope of work and the price) and any related components you may need. Standard operating procedure should be to replace the following components along with the compressor in every case:
  • Oil failure switch.
  • Compressor contactor.
  • Compressor overload protector (if installed).
  • Compressor internal overload protector module.
  • Liquid line filter drier (preferably a changeable core type).
  • Suction line filter drier (changeable core type).

When compressors fail, debris usually spreads throughout the system regardless of the type of failure. This debris must be removed ahead of the compressor to avoid clogging any internal suction strainers or causing bearing wear.

Startup Procedure For New Compressors

When performing a startup, take your time; it’s not a race. Mediocre technicians rush, good technicians take their time, check their work, and make sure the compressor won’t fail again.

Please note that the following doesn’t include steps relative to such things as leak checking, leak repair, evacuation, charging, etc. It also doesn’t address each step relative to system startup such as opening service valves.

Following is the start-up procedure:

  • Ensure there is adequate oil of the proper type in the compressor.
  • Ensure that all components called for above, and any others found to be faulty, have been replaced.
  • Trace any refrigerant lines and note installation details such as traps before risers (or lack thereof), lines pitched incorrectly, long traps caused by improper installation, etc.
  • Check the size of the refrigeration piping at full load and at fully unloaded conditions to ensure adequate gas velocity for oil return.
  • Inspect hot gas bypass piping. Ensure that the velocity in parts of the system does not decrease to unacceptable levels when at full hot gas bypass (this is typical where hot gas is introduced into the suction line rather than at the coil inlet. Make sure the hot gas valve is located close to the compressor and not at the coil (except in close coupled systems).
  • Ensure there aren’t oil traps created in the hot gas bypass line, as this line experiences variable flow.
  • Test the operation and timing of the oil failure switch by applying and removing pressure from the sensing tubes and witnessing that the switch trips in the prescribed time period.
  • Check all wiring to ensure it complies with the wiring diagram. If not, raise a flag and get more information. If the wiring is messy, clean it up so you can work with it and understand it.
  • Check for any interface from a third party control system.
  • Install at least the following test instruments:
    • Suction pressure service gauge.
    • Suction temperature probe.
    • Discharge pressure service gauge.
    • Discharge temperature probe.
    • Gross oil pressure service gauge.
    • Crankcase pressure service gauge.
    • Liquid line pressure service gauge.
    • Liquid line temperature probe.
    • Voltage per phase.
    • Amperage per phase.
  • Start the compressor.
  • Add refrigerant as required to provide a full operating charge when the building is within its normal temperature range.
  • Record system readings and continue adjusting the system and recording readings until you are satisfied the system is operating properly and you have positively identified the cause of failure for the prior compressor.

Watch the system work over the entire range of operation (loaded/unloaded, hot gas bypass in/out, etc.), and record information at each condition.

There is a reason for every compressor failure, and most of them are attributable to system problems, not compressor problems. Record the following readings as a minimum:

  • Suction pressure (58 psig ¯ 76 psig is the normal range).
  • Suction temperature.
  • Saturated suction temperature (30F to 45F is the normal range).
  • Suction superheat at the compressor (20F including error correction).
  • Compressor crankcase temperature (105F to 125F is normal and in some cases up to 150F). Check with the compressor manufacturer if in doubt.
  • Compressor discharge pressure (up to 300 psig on older systems, 225 psig ¯ 240 psig on newer systems on a 95F day). Another rule of thumb is that saturated condensing temperature on systems manufactured prior to about 1975 will be about 30F to 35F above ambient. On newer systems, it should be about 15F to 20F above ambient.
  • Compressor discharge temperature (160F is the minimum, 190F is ideal. Anything over 200F must be compared against the manufacturer’s recommended maximum temperature. For example, on Copeland semi-hermetics, the maximum is 230F, on Carlyle semi-hermetics, the maximum is 275F, etc. Refrigeration oil begins to break down at 350F. The manufacturer’s maximum limits give an indication of the actual cylinder temperature based on their test).
  • Gross oil pressure (varies by manufacturer).
  • Crankcase pressure (should be within a pound or two of the suction pressure at the service valve).
  • Net oil pressure (varies by manufacturer).
  • Oil level (varies by manufacturer and compressor model).
  • Liquid line pressure.
  • Liquid line temperature.
  • Saturated liquid line temperature.
  • Liquid line filter drier inlet temperature minus outlet temperature (normal is 2F or less).
  • Sub-cooling (10F to 15F is normal).
  • Sight glass condition (normal is clear when compressor is fully loaded and hot gas bypass is off).
  • Evaporator air in (normally 75F to 80F).
  • Evaporator air out (normally 55F exiting the machine).
  • Evaporator temperature difference (18F to 22F is normal).
  • Evaporator approach (evaporator leaving air temperature minus the saturated suction temperature). This is 20F to 23F on older systems, 10F to 15F on newer systems.
  • Condenser air in. Normally this .s the design outdoor ambient temperature. On certain installations, however, a lack of a natural breeze, a dark roof and a sunny day can push the inlet temperature higher.
  • Condenser air out.
  • Condenser temperature difference (10F to 15F on newer systems, 20F to 25F on older systems.)
  • Condenser approach (saturated discharge temperature minus condenser leaving air temperature). Normal is 10F to 15F .
  • Volts per phase (leg to leg) where ±5% of nominal rated voltage is normal, ±10% is the maximum.
  • Voltage imbalance between phases (2% maximum).
  • Amps per phase.
  • Amperage imbalance between phases (up to 10% is normal).

Begin System Checks

Once you’re satisfied the compressor is operating safely, begin the system checks. Until all checks are complete, don’t operate the system unsupervised. System checks include the following:

  • Check any controls internal to the unit (live test). Operate each control and safety to ensure that they operate as expected and are set to operate at the proper pressures or temperatures.
  • Check all timers to ensure that the compressor is not short-cycling.
  • Check all controls external to the unit such as third-party control systems. If a third party-control system is installed, obtain a sequence of operation from the system provider written in plain English so you can understand how the system is expected to operate and so you can pass judgment on their method of control. Remember, you own the compressor warranty.
  • Check all interlocks between internal controls and external or third party controls. It’s not uncommon to find a condensing unit off on time delay and the liquid line solenoid energized. If no interlocks exist, give the customer a price to install them.


Detailed service reports are to be prepared daily. Documentation that must be prepared at the end of the job to formally close it, includes:

  • A complete description of the work performed.
  • An explanation of the cause of failure. If no cause is found, it must be clearly stated.
  • A description of any mandatory work required to make corrections so the warranty can remain in force. This should include any work required of the third party controls contractor that must be tested and witnessed by you (the technician). Without successful completion of this work, you cannot accept warranty responsibility.
  • A description of any optional or recommended work.
  • The last complete set of operating readings taken.
  • A qualified representative of the owner must sign all documentation. A qualified representative is a person of authority who can understand the financial repercussions to his or her company if the specified repairs are not made expeditiously.

When you, the technician, close out the job, you’re putting your seal of approval on it. Your reputation as well as the reputation of your company is at stake.

Harry Hartigan is the president of Mechanical Service Corp., a 27-year old commercial/industrial contracting firm based in Whippany, NJ. The company specializes in process cooling, facility automation, and HVAC service. This article is based an a technical bulletin Hartigan put together for his service technicians. He can be reached via e-mail at [email protected].