Servicing a turbopump - DIY

[olof.gross]

Hello!
I just inherited an old (12yrs i think) MS with a worn-out turbo. MS is a Perkin-Elmer Clarus 500, turbo is Edwards EXT255H. Changing the turbo is a bit expensive on a student budget (especially considering i'm definitively not sure i'll ever get it to work - reason they gave it to me was that there's been some problems with the electronics, so they're buying a new one), but i'm the kind of guy who's not afraid to try things that you're not supposed to do yourself just to prove 'em wrong.

Have anyone here ever taken apart a turbo? D'you know exactly why the manufacturers "recommend" returning the pump for an exchange, does changing the bearing require specialized equipment to balance it, or something like that? The vanes spin freely when i move the pump so i guess the bearings are just beginning to wear out, can that result in a loss of pump efficiency so the MS won't work? Or do they break down completely before loss of efficiency occurs? Any idea where i could find a suitable replacement bearing?

[Alp]

I wish you luck.
If the problem is purely electronic, and you have easy access to the circuit components and know what you are doing then you might get away with it.

However, Turbo-molecular pumps are a serious bit of hardware and have to be respected. I would guess that the circuitry driving them has to be perfectly tuned, in order to control the RPM and current loads when spinning up and if you do not know what you are doing you just might make things a lot worse.
On second reading of your email I get the impression that you suspect the bearings might be worn out. I think an experience service technician would be the best person to ask in this case.

Myself, I'd look for a replacement turbo with a warranty but then i am not on a student's budget. Or maybe buy a dead instrument with a similar and working turbo pump.

Alp

[gapeev]

Is it feasible to rebuilt a turbo?

If it has not failed (i.e. still pumps as designed, may have some abnormal noise), then maybe, albeit unlikely. You will need a rebuilt kit (~$2K) and may need to balance the rotor. The latter is not DIY as you will need special equipment worth >$50K

If it has developed some mechanical problems (e.g., broken blades, frozen rotor, loud rattling noise) rendering it nonfunctional, forget about it.

If the problem is with a controller or power supply, you may be able to have it repaired or replaced.

For replacement components, Ebay is an excellent place to look, for example item no. 130943896748

If the issue is not about the turbo, well... every mass spec I dealt with had some sort of self check to facilitate troubleshooting. I'll have to stop here as not to take bread from MS service people.

Good luck!

[leadazide]

Hi

Being a service engineer on mass specs my self I would warn against trying to do any maintenace work on Turbo pumps your self.. I don't touch the things! I replace them! I have seen and heard cases where turbos have catastrophically failed and I have the utmost respect for them! Those suckers are packing a lot of power!

I think the rebuilding on pumps need to be done in an alignment tool to ensure balance and stability..

Bets of luck in getting the old system up and running.. Like already mentioned your cheapest bet would to find a similar dead instrument with a working turbo..

Good luck!

[paul_b]

I agree with the above, do not consider dismantling a turbo without the alignment tools, they run at collosal speeds and the tollerances are extremely tight! Some more modern models have magnetic bearings and as such are 'maintenance free', older ones had oiled wicks and it is worth checking these - I managed to get another two years out of one which had been removed previously by re-soaking the oil pad.

We do have a great photo where a turbo failed and the blades went through the protective mesh - it made full volume motorbike impressions at the same time .

[olof.gross]

Thank you very much for your answers, all of you.

At the bottom of the pump, there's a cover secured with 8 screws. I removed it, and underneath was some rings made of some felt-like material, drenched in oil - this must be the wick that keeps the bearing lubricated. I didn't have time to dig any further but i'll return to the problem later today hopefully. If the bearing can be extracted from this side, there won't be any need to mess with the rotor, and hopefully it won't need realigning.

paul_b: you say you didn't even change the bearing, just added some lubricant to the oil pad? That's very interesting indeed, exactly what i would need so to say! What kind of oil did you use? I guess the vapour pressure must be minimal...

If i succeed in this, i promise to post pictures for your enjoyment!

[MSCHemist]

There are companies out there that service turbo pumps. I believe their prices are reasonable ~$1k. I'd rather give it to them.

[James_Ball]

When I first started working here in the lab 22 years ago, the manager had rebuilt the Balzers turbo in our 5970 by himself. Every time you would pump down or vent it would sound like a large truck hitting the engine brake going down a steep hill, and sounded like a jet plane sitting on the runway idling all while it was running. It was actually louder than the roughing pump, but the thing lasted 5 years!! I came in one morning and the lab was totally silent, that is when I knew it was dead. Replace it and thought the replacement was broken. Pumped it down twice before I realized a turbo pump was not supposed to be so loud.

So, it can be done, but I definitely would not recommend it. I also saw the one at another lab here in town that had a catastrophic failure. It was in another 5970. The blades literally turned to dust and the whole analyzer and quads were covered with aluminum powder. The whole instrument was toast after that.

Best test to see if your bearings are shot, hold it level in your hands or on the counter and rotate it quickly, if the blades remain stationary while it is turning and only turn a little at the end of your movement, then they are good, if they turn with the body of the pump, the bearings are bad. Good bearings are really that good. You can also blow gently on the blades and they should spin freely and keep spinning for several seconds if the bearings are good.

[paul_b]

It was genuine Pfeifer tubo-pump oil - I actually binned my last litre bottle in June as all of mine now have magnetic bearings.