Would anyone know of a way to determine what exactly the paricles are that we are seeing in an air compressor oil sample? The attached oil analysis report is from a Gardner Denver rotary screw compressor. The wear metal and silicon levels are low, but our small particle counts are getting consistently higher, even with frequent filter changes. Any thoughts?

0700AC10,_09-26-06,_MDK.pdf (39 Kb, 28 downloads)

Mark:

Are you looking for an exact particle analyser? An analytical ferrography test would indicate if it was iron metallic particles. A photograph of the particles are great from an AF test as well. An AF test usually costs between $100-$125 each, so a few tests per year is feasible.

Depending on the lab you use, try to get them to run a more in-depth test on the particles.

In your site, try a patch test with a magnifying eye piece. A drop test on a business card helps, too.

Are you having any reliability issues with the screw compressor? Does it "sound" strange or different than it used to? Is there a higher usage of oil? MTBF lower? MTBR lower? Is it continuously cycled on and off? Has the operation changed on the compressor?

This information may not give you what the particles are, but might lead you in the right direction. Your evidence will either eliminate or confirm your suspicions. Get all the evidence you can.

We just lost an identical compressor this month due to bearing issues. We had been seeing bearing faults in the vibration spectrum for the past few months, so we knew it was on it's way out. Now, this compressor just got called out yesterday for possible bearing fault frequencies. Another interesting point is that we have had 3 different types of oil in the past 3 years go through these compressors, due to switching to a new supplier, and products being discontinued.

DING DING DING!!!

You just answered your own question. Different oils, different suppliers, different additive packages, etc. Oil has additives that depending on their use can really build up inside equipment, especially when you change the oil.

Was the oil drained, flushed and filled when you converted? That would reduce the effect of settling out of the additive packages just topping up with new oil. Also, additives by their inherent nature may just build up on components.

Are you seeing the bearings built up with any kind of scale (white or black), or a buildup of sludge on the bottom of the reservoir? These are good indications of additive package incompatibilities.

The other question is viscosity. Have the oils you have used maintained the same viscosity? If it did change, it may be another good indication...

Two comments.

1 - Particle counts seem to have a way of being erratic when there is no apparent explanation. We have seen it ourselves but I have no idea why.
2 - I did hear that at the recent PDM conference in Chatanooga, a paper was presented identifying there can be a LOT of variability in the way that particle count samples is analysed, even within the same lab.