Vacuum Pump Oil for rotary vane pumps.

[bunnspecial]

I had someone stop me at work today and ask if I though vacuum pump oil would make a good gun oil. My immediate response was no, as I think it's too viscous and also stinks to high heavens(I suggested some alternatives I've tried, both gun specific and general purpose light oils).

It got me wondering, though, just what exactly is in it. Does anyone know this?

Vacuum pump oil-at least in a chemistry lab-gets abused pretty badly. Just mechanically speaking, the vanes undergo pretty high sheering forces. My rule of thumb is that a good rotary vane pump should be able to pull 10^-2 torr(although I've seen some do better) which necessitates the oil having a fairly low vapor pressure. The oil has to deal with all the [censored] that gets pumped into it-namely solvents-and many don't get drained until they get noisy or start having other problems. I've had some badly neglected pumps where I drained a few hundred milliliters of sludge out the bottom. In most of those, I refill, leave the pump pulling against a vacuum for a day and then drain again-I can usually get a lot more sludge.

The smell of "fresh" oil makes me think it likely is something similar to automotive gear oils-as mentioned, it's quite viscous and also has something of a sulfur smell. Does anyone have any specs on what a typical rotary vane vacuum pump oil might look like?

 

[Sunnyinhollister]

Most rotary vane pumps will use something like this:

http://inlandvacuum.com/products/mechanical-pump-fluids/inland-19-vacuum-pump-oil/

A good two stage pump will go into the low milli-torr range.

Pumps that will see hydride or oxygen service will use something like this:

http://www.pchemlabs.com/product.asp?pid=4929

 

[KingCake]

It's just mineral oil. Nothing magical.

 

[bunnspecial]

Thanks.

BTW, I've used many gallons of Inland 19.

Quote:
A good two stage pump will go into the low milli-torr range.


10^-2 torr is 10 millitorr. I've seen freshly rebuilt pumps get close to 10^-3(1 millitorr) but it's been my experience that 10^-2 is more realistic over the course of a rebuild. There's a reason why we call these "rough pumps" in high vac applications-in something like a mass spectrometer they're virtually always backing either an oil diffusion pump or a turbomolecular pump. Depending on the chamber specs and how good the high vac pump is, they will often get you to anywhere from 10^-6 to 10^-10 torr. Most of our quadrupoles runs at 10^-6, while the time of flight runs at 10^-8.

 

[kschachn]

Originally Posted By: bunnspecial
I had someone stop me at work today and ask if I though vacuum pump oil would make a good gun oil. My immediate response was no, as I think it's too viscous and also stinks to high heavens (I suggested some alternatives I've tried, both gun specific and general purpose light oils).


What's the brand of that oil? The DuoSeal oil I have for my little Welch pump doesn't have much of a smell at all.

 

[bunnspecial]

Originally Posted By: kschachn
Originally Posted By: bunnspecial
I had someone stop me at work today and ask if I though vacuum pump oil would make a good gun oil. My immediate response was no, as I think it's too viscous and also stinks to high heavens (I suggested some alternatives I've tried, both gun specific and general purpose light oils).


What's the brand of that oil? The DuoSeal oil I have for my little Welch pump doesn't have much of a smell at all.


I'd have to check. For a while I was working out of an old stash of Welch branded pump oil, and I never recall it being particularly potent smelling either. It was of course designed for the old low-speed belt driven Welch pumps, and I'm not even sure if those are made anymore(although we probably have 50 or more still in service).

I had both an Alcatel engineer and someone at a local independent scientific instrument company tell me that the particular oil is no longer appropriate and the local company recommended either Inland 19 or an Edwards product depending on the specific pump. Of course, Alcatel told me that their oil was the only appropriate one. The Edwards oil is the stinky one, and it's the only one I keep around since the local company claims it won't hurt anything in an older pump.

 

[tom slick]

It might be good for keeping your gun lubed in Space.

 

[Sunnyinhollister]

Originally Posted By: bunnspecial
Thanks.

BTW, I've used many gallons of Inland 19.

Quote:
A good two stage pump will go into the low milli-torr range.


10^-2 torr is 10 millitorr. I've seen freshly rebuilt pumps get close to 10^-3(1 millitorr) but it's been my experience that 10^-2 is more realistic over the course of a rebuild. There's a reason why we call these "rough pumps" in high vac applications-in something like a mass spectrometer they're virtually always backing either an oil diffusion pump or a turbomolecular pump. Depending on the chamber specs and how good the high vac pump is, they will often get you to anywhere from 10^-6 to 10^-10 torr. Most of our quadrupoles runs at 10^-6, while the time of flight runs at 10^-8.


Most of the two stage pumps we used (Leybold and Edwards) would get down to 5 millitorr or under. The last diffusion pump I saw was on a very old CHA evaporator. Most of our high vacuum pumps are turbo or cryo pumps. I also worked on SEMs at one time which used ion pumps as well as mag lev turbo pumps.

Fun stuff

 

[bunnspecial]

Originally Posted By: Sunnyinhollister
Originally Posted By: bunnspecial
Thanks.

BTW, I've used many gallons of Inland 19.

Quote:
A good two stage pump will go into the low milli-torr range.


10^-2 torr is 10 millitorr. I've seen freshly rebuilt pumps get close to 10^-3(1 millitorr) but it's been my experience that 10^-2 is more realistic over the course of a rebuild. There's a reason why we call these "rough pumps" in high vac applications-in something like a mass spectrometer they're virtually always backing either an oil diffusion pump or a turbomolecular pump. Depending on the chamber specs and how good the high vac pump is, they will often get you to anywhere from 10^-6 to 10^-10 torr. Most of our quadrupoles runs at 10^-6, while the time of flight runs at 10^-8.


Most of the two stage pumps we used (Leybold and Edwards) would get down to 5 millitorr or under. The last diffusion pump I saw was on a very old CHA evaporator. Most of our high vacuum pumps are turbo or cryo pumps. I also worked on SEMs at one time which used ion pumps as well as mag lev turbo pumps.

Fun stuff


High vac systems are always fun stuff.

I think you'll find that diffusion pumps are still really common on things like simple quadrupole or ion trap MSs. They are simple, reliable, and compact with the downside that they are slow. Our main GC-MS will come down in a couple of hours, but it really takes about two days to stabilize.

There are even a couple of mercury diffusion pumps floating around on Schlenk lines.

Our TOFs have turbopumps primarily because there's a LOT of space there to pump down and you also get a lot of air in them when you're loading the sample plate. I know that the sample chamber on our main one takes about 5 minutes to get down to 10^-8 after you load the plate.

I've also seen turbo pumps self destruct. The newer ones require you to plug an ion gauge into the controller and won't spin the pump up without a sufficient vacuum. One our our techs, though, fired up an old one without pumping it down first, and he ended up shaking pieces of the rotors out of the vacuum chamber-fortunately the body is tough enough that it's rare for something to go through.

The last cryopump I dealt with was on a Finnigan triple-quad. That probably was my favorite high vac pump I've used-it gives you the speed of a turbo pump with the overall reliability of an oil diffusion pump.

 

[kschachn]

Originally Posted By: bunnspecial
Originally Posted By: kschachn
What's the brand of that oil? The DuoSeal oil I have for my little Welch pump doesn't have much of a smell at all.

I'd have to check. For a while I was working out of an old stash of Welch branded pump oil, and I never recall it being particularly potent smelling either. It was of course designed for the old low-speed belt driven Welch pumps, and I'm not even sure if those are made anymore(although we probably have 50 or more still in service).


Ehh I just wondered. Something having a smell equates to volatiles and I'd think that's the last thing you want in a vacuum pump oil.

 

[Sunnyinhollister]

Originally Posted By: bunnspecial
Originally Posted By: Sunnyinhollister
Originally Posted By: bunnspecial
Thanks.

BTW, I've used many gallons of Inland 19.

Quote:
A good two stage pump will go into the low milli-torr range.


10^-2 torr is 10 millitorr. I've seen freshly rebuilt pumps get close to 10^-3(1 millitorr) but it's been my experience that 10^-2 is more realistic over the course of a rebuild. There's a reason why we call these "rough pumps" in high vac applications-in something like a mass spectrometer they're virtually always backing either an oil diffusion pump or a turbomolecular pump. Depending on the chamber specs and how good the high vac pump is, they will often get you to anywhere from 10^-6 to 10^-10 torr. Most of our quadrupoles runs at 10^-6, while the time of flight runs at 10^-8.


Most of the two stage pumps we used (Leybold and Edwards) would get down to 5 millitorr or under. The last diffusion pump I saw was on a very old CHA evaporator. Most of our high vacuum pumps are turbo or cryo pumps. I also worked on SEMs at one time which used ion pumps as well as mag lev turbo pumps.

Fun stuff


High vac systems are always fun stuff.

I think you'll find that diffusion pumps are still really common on things like simple quadrupole or ion trap MSs. They are simple, reliable, and compact with the downside that they are slow. Our main GC-MS will come down in a couple of hours, but it really takes about two days to stabilize.

There are even a couple of mercury diffusion pumps floating around on Schlenk lines.

Our TOFs have turbopumps primarily because there's a LOT of space there to pump down and you also get a lot of air in them when you're loading the sample plate. I know that the sample chamber on our main one takes about 5 minutes to get down to 10^-8 after you load the plate.

I've also seen turbo pumps self destruct. The newer ones require you to plug an ion gauge into the controller and won't spin the pump up without a sufficient vacuum. One our our techs, though, fired up an old one without pumping it down first, and he ended up shaking pieces of the rotors out of the vacuum chamber-fortunately the body is tough enough that it's rare for something to go through.

The last cryopump I dealt with was on a Finnigan triple-quad. That probably was my favorite high vac pump I've used-it gives you the speed of a turbo pump with the overall reliability of an oil diffusion pump.


Except for the SEM, all the vacuum systems I've dealt with were for manufacturing. One of the reasons that the diffusion pumps were phased out was due to contamination from the pump and dealing with the contaminated oil. They were primarily used on ion implanters which pumped all sorts of nasties like PH3, AsH3, B2H6, ect. Implanters now a days use turbos in the source and beam line areas. For this reason most of the wet pumps are now gone. A lot of our techs have never even seen a rotary vane pump.

Sounds like the chambers I dealt with were a tad larger. It would take a lot longer to pump down one of our implanter wheel chamber or sputtering systems to
Our turbos were mostly 1000 torr liters per second variety and when they failed they could be quite spectacular. I remember looking into the inlet of one pump after it failed and it looked like the end of a wood screw.

Some of the failures I've seen on these large systems interesting and comical. I had one idiot tech dump the vacuum system on an older implanter so bad the oil in the diffusion pump coated the whole machine. Then there was the time an operator was manually trying to tune the beam on a high current implanter and burned a whole through the stainless wall of the chamber. Of course that was right before a holiday weekend and I had to spend the whole weekend there to fix it. I remember when one of our facilities lady was walking past one of our machines and accidentally hit a breaker which powered off the mag lev turbo pump. The sound it made when the rotor came down on the emergency bearings can best be described as someone violating a moose.

Now most of the systems I work on only get down to 100 mbar.

Boring....

 

[bunnspecial]

By the way, as far as stinky oils go, one of the worst ones I ever did was an old Welch pump that had been running(on and off) daily for 2 years on a Rotovap pulling off Carbon Disulfide. When I checked the vacuum of it before the change, it was struggling to pull about 5 torr, and made a horrible racket. It had been used with a dry ice trap, but many of the folks running it were lazy. They'd load it with dry ice in the morning(not dry ice and methanol or acetone) and then never bother to top it off until it was nearly empty. The trap almost never got drained, despite the fact that it had a stopcock on the bottom that made it stupidly easy to do.

I spent three days changing the oil in it. When I did the first one, the oil actually came out in three layers with the top one floating out like water(given how bad the plumbing on the line was, it may well have been) followed by not very viscous pump oil and then sludge that dripped out of the drain cock for a couple of hours. I refilled it and left it running overnight(stopper on the inlet). Fortunately, it was much quieter, but was still only pulling about .1 torr. I drained another 300mL or so of sludge out of it(along with the rest of the oil), left it running that day, then drained and refilled again before I left for the day. On the third day, the oil sample I took out actually smelled like the stuff I put in and the pump was nearly dead silent. By that point, it was in the 8-10 millitorr range, which I figured was the best I could ask for of a pump with unknown history.

I was working a summer internship at a biofuel lab at the time, and EVERYONE thanked me for how much better the pump was running. Unfortunately, my supervisor wouldn't let me replumb it-which needed to be done(if nothing else to get rid of the dry rotted natural rubber vacuum hoses). Given that we were primarily pumping off CS2, it should have been used with a liquid nitrogen trap. With that said, I didn't advocate for that as I didn't feel like the plumbing on it was tight enough to use liquid nitrogen safely. For reference, I could get 8 millitorr right off the pump, but only about 100 millitorr at the line going into the rotovap.