HPAC Magazine

In Defence Of Compressors

Steps to getting at the root cause of compressor failures.

December 1, 2014   By DAVE DEMMA

I am a great believer in HVAC/R technicians thinking of their craft as being similar to that of being a doctor. As I have said before on these pages, the course of study to become a doctor is fairly rigorous. If one makes it through that lengthy apprenticeship, they have earned the right to practise medicine. Why does it take so long? Simply put, there is so much at stake.

While there is still much we have to learn about the human body, some basic parallels exist between it and a mechanical refrigeration system. The heart is nothing more than a pump, while the veins are “pipes” supplying blood to the intended locations. The nervous system is a complex electrical circuit, with the brain as the central processor, which controls everything.

If you can agree with the comparison, would it then be such a stretch to think of the refrigeration technician as a refrigeration doctor? Perhaps the technician can benefit from emulating the doctor’s methods in analyzing technical problems? For example, would a doctor suggest a heart transplant because he is “pretty sure” that the patient’s heart is not operating properly? Of course not. Any hint of mystery surrounding the operation and diagnosis of a heart would have disappeared long before having to make the crucial decision to recommend the transplant.

Now, let’s look through the eyes of a doctor and see if it might be of benefit in helping to analyze a few system failures.

First, a few facts from compressor manufacturers:

• Approximately 30 per cent of all compressors pulled from the field, returned to manufacturers and torn down for failure analyses do not have observable defects. (Copeland – Emerson Electric)

• 60 to 70 per cent of returned failed compressors are the result of system/service related issues or the result of misdiagnosis. (Carrier Corporation)

• 80 per cent of compressors returned for electrical motor failure were system caused mechanical failures that progressed into an electrical failure. (Carrier Corporation)

• 95 per cent of alleged warranty failures turned out to be caused by external influences from the refrigeration system itself – the compressor was not at fault. (Bitzer Corporation)

• .25 per cent (¼ of one per cent, or one out of every 400) of total UK compressor sales resulted in actual valid warranty claims. (Bitzer Corporation)

I could cite more facts, but you get the general idea: many compressors are misdiagnosed as failed when in fact there is nothing wrong with them; for many compressors that have suffered a failure in the field, the cause of failure was misdiagnosed; and the rate of compressors that fail during the warranty period due to some kind of manufacturing defect is quite small.

Let me illustrate an example of the latter: years ago I was dispatched to a supermarket with the complaint of “store too hot.” I found the 20 HP hermetic compressor on the condensing unit scrambled. A replacement was not readily available, so I picked up a 20 HP semi-hermetic as a replacement. Another technician and I went back the following day to complete the repairs.

I re-piped the system to accommodate the new compressor footprint, and my co-worker worked on the controls and electrical portion. The original compressor was set up as a part winding start, two contactors, each wired to one half of the motor. I reminded my co-worker that the two halves of the motor would need to be phased correctly so they would rotate in the same direction. He acknowledged that he understood and would wire the compressor accordingly.

Well, he might have understood but he did not wire the compressor correctly. The moment the breaker was switched, and power was applied to the compressor, it destroyed itself. The action of half the motor rotating clockwise and the other half the motor rotating counter-clockwise resulted in an immediate motor burn. While this was clearly “the result of system/service related issues,” and not the result of a defect in workmanship/materials, the compressor was returned to the supplier as a warranty failure.

Now, let’s look at two specific compressor failures to see what can be learned from approaching the repair as a doctor, or better yet, a medical examiner performing an autopsy.

Service Call #1

Upon arrival to the job the compressor circuit breaker was found tripped. After verifying that the compressor crankcase was not full of liquid refrigerant (due to refrigerant migration during the off time), the circuit breaker was reset, followed by an attempt to restart the compressor. The circuit breaker immediately tripped again.

The three-phase power supply to the compressor motor terminals was removed and the compressor windings were checked with an ohm meter to see if they were shorted to ground. Not surprisingly, the compressor motor was shorted to ground…a compressor motor burn.

A new compressor was obtained from a supplier and installed. The filter-drier was changed, a suction filter-drier was added, the contactor was replaced (per the compressor manufacturer’s requirement) and the new compressor was started. Oil pressure, voltage supply and amperage were checked and seen to be within the manufacturer’s recommendations.

So, the easy part was completed – changing the compressor. The difficult part was to ascertain why the compressor failed. Realizing that most compressor motor failures were the result of a mechanical failure, the compressor was partially disassembled to see what clues were available to assist in this diagnosis.

Figure 1 shows the cause of the motor failure: a suction reed that had become imbedded in the motor windings. Obviously this was not the result of a motor failure, insufficient or unbalanced voltage, defective contactor, or any other electrical related issue. This is a mechanical problem that escalated into a failed motor.

A thorough monitoring of system operation was required, and resulted in finding a TEV was set such that the superheat was nonexistent during periods of low load conditions. The resulting refrigerant flooding was the culprit in the mechanical damage to the compressor, which then resulted in the damage to the compressor motor.

 “Instead of paying for a service call to address the decomposed oil issue, the customer will now pay to address the decomposed oil issue plus buy a new compressor.”

Service Call #2

What a surprise, yet another call where the compressor had tripped the circuit breaker. Again, after ensuring that the crankcase had no liquid refrigerant in it, the circuit breaker was reset. As was the case in service call one, the breaker immediately tripped. Once again, the three phase power supply to the compressor motor terminals was disconnected and the motor windings checked to see if they were shorted to ground. Once again, it was verified that they were.

So, another compressor motor failure, which resulted in another compressor replacement (and contactor replacement). Once again, the amperage, supply voltage and oil pressure was verified to be within the manufacturer’s recommendation.

This was a particularly bad motor burn. The oil was filthy dark brown. At the very least, a return was required
the next day with an acid test kit to verify whether the system was still acidic or not. This had nothing to do with ascertaining the cause of failure, but rather ensuring that the new compressor is going to live a long life. If there is still acid present in the system after the replacement compressor has been installed, then the new motor windings are at risk of degradation, resulting in a potential repeat motor failure.

As with the previous service call, once the system was operational it was time to turn our attention to what caused this motor to fail, which again required a partial compressor teardown (the autopsy).

As mentioned above, the oil in the crankcase was a dark brown, with a fair amount of sludge in the crankcase. In fact, there was so much sludge in the crankcase that the oil inlet screen in the crankcase was nearly completely plugged. This made it impossible for the oil pump to supply oil to the compressor bearings.

Why didn’t the oil failure switch prevent the compressor from operating? The L and M terminals in the oil failure switch had been jumped, eliminating the control’s ability to shut off the compressor during periods of low oil pressure. This was not a smart move by whoever did it. Instead of paying for a service call to address the decomposed oil issue, the customer will now pay to address the decomposed oil issue plus buy a new compressor.

Further teardown revealed that the compressor is main bearings failed. Of course, with the oil failure switch jumped out, the compressor continued to run until the failed bearings caused the motor’s rotor to drop down and make contact with the stator. After what was likely a brief amount of time, the contact between the rotor and the stator resulted in enough stator damage to cause the motor short.

To properly address this issue the technician needs to discover why the crankcase was full of decomposed oil and sludge. This was a low temperature R22 refrigeration application. The thermodynamic properties of R22 are such that it will operate with elevated discharge temperatures during low temperature applications. A quick inspection of the condenser revealed that it had not been cleaned in years, adding to the already high discharge temperature.

Once mineral oil exceeds a temperature of approximately 350F it will start to decompose. The process is called polymerization. In laymen’s terms this means the oil molecules combine to form larger molecules (thick oil), even larger molecules (sludge) and finally very large molecules (solid material). In the process, the oil loses its ability to lubricate and the addition of sludge in the crankcase can easily plug up things such as oil pump inlet screens.

So, we have done the autopsy and found that the root cause was a dirty condenser (lack of maintenance). This resulted in elevated discharge temperatures which led to oil decomposition, and ultimately, a lack of lubrication.

This normally would have resulted in an oil failure control trip, but someone jumped out the control, likely due to repeated oil failure trips that for some reason were not able to be diagnosed.

The jumped out control allowed the compressor to continue to operate without sufficient oil pressure to properly lubricate the bearings, resulting in bearing failure. As the compressor continued to operate with failed bearings, the stator dropped down, making contact with the rotor and eventually destroyed the motor, with the final failure manifesting itself as a compressor motor failure, yet again. We see a mechanical failure leading to a motor failure yet again.

The three-step fix for this system is:

1. Clean the condenser.

2. Continue to monitor the acid level in the system until it is within a safe level. Once the acid level is at a safe level, replace the liquid line filter-drier, remove the suction line filter-drier, and replace it with a suction line filter.

3. Enlighten the customer to the benefits of a preventive maintenance program.

Now imagine if the technician who responded to either of these service calls was simply a parts changer. Yes, he might have done a perfect job replacing the failed compressor and might have even gone the extra mile in properly cleaning up the system after the motor failure.

But doctors treat symptoms and seek root causes of sickness. They do not guess. If the technician was not willing to perform a proper autopsy and ascertain the root cause of the failure and eliminate the conditions that allowed that root cause to progress into a failure, then the customer would be faced with replacing another compressor in a relatively short amount of time.

You can take this to the bank: if a compressor has failed and the cause of failure has not been determined and eliminated, the replacement compressor will fail too.  <>

Dave Demma holds a degree in refrigeration engineering. He worked as a journeyman refrigeration technician before moving into the manufacturing sector where he regularly trains contractor and engineering groups. Demma can be reached at ddemma@uri.com.

Advertisement