Users with only a few vacuum pumps, or those not in constant use, don’t think about pump oil very much. On our small deposition systems, I’ve heard from customers about decades of running without an oil change. However, oil gets dirty, changes chemically, and often just leaves the pump one way or the other. We can only assume that the hydrocarbon versions at least, are going to get expensive along with everything else oily.
Let’s keep these costs down!
If you’re pumping something particularly reactive, like pure oxygen, effluent from a plasma reactor, CVD tool, etc, you already know how fast oil can get degraded. The mechanisms are chemical change, usually in the direction of increasing viscosity, corrosive additions, accumulation of abrasive particles, and just plain dirt. A rotary vane pump running hydrocarbon oil may fail in just a few hours, and a Fomblin or other inert oil pump may fail surprisingly soon as well, especially because the impurities will not mix with the oil, form a supernatant film, and rot through the pump. A bathtub ring of the nasty kind!
Some medium and larger pumps sport automotive-type cartridge filters , and these can help , especially with particulate effluent, or, if they’re designed for it, some chemical reduction. Then too, there are external oil circulating filters available that are even better, having generally more capacity. The pump manufacturers can help pick the proper filtration media.
Small systems, running benign process, though, have two special problems. One is oil being sucked back in the direction of the high vacuum pump or chamber by misoperation of valves, a power failure, or anything that slows or stops the mechanical pump. Modern pumps have ‘anti-suckback’ valves as a rule, but it’s not good to rely on these. The second problem is pump oil leaving by the exhaust port- everything else is going in that direction, after all. Operation at too high a load, in startup, or against a leak, will cause the pump to eject its oil. It really gets interesting when it starts to pool in the building’s exhaust lines!
The one problem can be alleviated both with education and a coalescing filter between the mechanical and high vacuum pump. The other problem has both of these cures, plus attention to how the exhaust is routed.
Just reading the pump oil level is no indicator that everything’s ok- the pump could be exchanging its oil with that of say, a diffusion pump, or you could be reading something like water, coming in through the system.
The best thing to do is, first, get a tag and tie it to the pump. Record oil check dates. Second, Figure out how to drain the pump oil, and remove some periodically for inspection. Typically, you can’t tell much by looking at the sight glass. The pump oil should be clear to amber, or about the color of booze. Any darker, and there’s reason to change it, or diagnose a cause.
In some impoverished climes (are we talking about university labs here?) people have been known to drain oil and run it through a coffee filter before replacing it. I doubt that any pump maker would recommend this, but I would, if that’s all you can afford. While you’re at it, check for unusual smells, unmixed liquids, metal shards, etc. Be especially suspicious if you have a diffusion pumped system, as oil swapping between the pumps is not good.
If you have a large facility, and a great deal of oil, you can look into re-refining services. Just be aware that when you do this, you’re effectively allowing different vacuum systems to swap contaminants. For some purposes, this is also not good. Along those lines, many semiconductor fabs will throw out a very expensive pump rather than have it rebuilt, and risk cross-contamination.
