Changing oil on a turbomolecular pump

[Don Baker]

The maintenance manual for my 5970 MS says that I should change the oil in the turbo pump every six months.

But to do so requires unmarrying the GC, removing the cabling and turning the unit on its back to gain access to the pump. This just doesn't seem right. Maybe I'm misunderstanding the instructions, but I just can't see a working lab doing that sort of a procedure every six months. It seems that you would be risking contamination, breakage of the GC column, possible electrical problems, possible damage to the ion source, possible vacuum leaks, and who knows what else.

How often do those of you with a 5970 actually change the turbo pump oil?

What am I missing here?

Don Baker

[Peter Apps]

Hi Don

There is no oil in a turbo pump Or at least there certianly should not be any !), the oil that you change is in the roughing pump, aka the fore pump, which is the ordinary vacuum pump that pulls enough vacuum for the turbo-pump (or diffusion pump) that is built into the instrument to start working. If the oil in the fore pump needs changing it will be dark or cloudy - somewhere on the pump is a little window that you can check through.

Peter

[Ron]

Some of the early turbo pumps in the 5970 needed periodic lubrication, or at least that was the recommendation. My feeling was that in most cases removing the plugs to expose the oiling points introduced more dust and contamination that would shorten the life of the pump to offset the gain from topping off the oil supply. Basically a wick was being moistened with oil, it was very difficult for the pump to actually loose all oil.

I am aware of a pump being lubed according to the scheduled times that failed prematurely and spectacularly.

[Don Baker]

Peter - They specifically called out the turbo pump. I had never heard of such a thing either. My understanding of the best way to deal with turbos is to stay away from them.

Ron - Yes, they show photos of the access to the pump and the wick.

The one thing that makes me think that I might want to dig into the unit and at least check the oil is that the machine has been sitting around for so long - certainly more than five years and possibly even ten years.

OTOH, I have already spun up the turbo twice for vacuum tests and it seemed to do OK. The first time there was a somewhat disconcerting noise as the turbo was spinning up. If you've ever heard a bearing turning in a housing, it was that sort of noise (but I was specifically listening for that sort of sound). The sound disappeared after a few seconds and when I fired the pump up the next time there was no anomalous sound. I was able to get as low as 3.5 x 10^-6 torr, though eventually the pressure stabilized near 7 x 10^-6.

Loss of the turbo pump is one of my primary concerns. If I have to replace that, I may as well go out and buy a refurbished MS.

[Ron]

The bearings were probably dry when you cranked up the turbo pump, and as the pump got up to speed and temperature the lubricant redistributed itself and the bearings are now lubed. I don't know if I would open it up and try to lube at this point, since the noise went away there is probably enough oil to keep the bearings lubed for a while. The bad news, there was probably some damage done to the bearings when the pump was making noise as it spun up.

[tima]

I have actually done this procedure on an HP 5970 GC/MS many years ago but what I can remember is there was a kit. The kit included some new "o" rings and a new oil wick and you swapped out from the bottom of the turbo pump. It was very easy to do. Maybe you can get this kit from SIS or some other second party. I have never had a turbo pump fail but I have heard when they go they can go with lots of damage. Most likely if the pump survived the initial spin up it'll be OK.

Timothy

[Peter Apps]

Hi Don

I learn new stuff every day ! - and it sounds as if the procedure is more akin to regreasing the wheel bearings rather than changing the oil ! If you can synchronize the maintenance with say a source clean - which requires venint and opening the vacuum anyway it might not be too much of an interruption, but I would be inclined to follow Ron's advice and see how it goes.

Peter

[chemstation]

FYI,
site showing the repair of a Turbo pump

http://www.rtftechnologies.org/general/ ... 0turbo.htm

[Don Baker]

Wow! Somebody has a lot more courage than I.
But, of course, if the pump is shot, you can't hurt it much more.

Very enlightening! Thanks

[Bigbear]

Don't oil it! We had an early 70 and the Agilent rep said it caused more problems than it fixed. When the turbo goes the ones available now don't requuire oiling.
The only turbo that I have had die was the oiled type on our old 70. It was replaced with a Varian pump which ran for 10+ years.
We now have 4 5973's and have had no turbo related problems. One of these systems was doing CI with ammonia!