UHV Turbo pump failure in Thermo Exactive

[hfredricks]

Hi everyone,

I have an Exactive Plus (Thermo orbitrap) and the ultra high vacuum turbo pump just failed. I've never had a turbo fail in my 13 years (I know, I'm lucky!) and this one is less than 2 years old. After talking with some friends & colleagues it seems that the Exactive UHV turbos fail a lot, from the 3 Exactives that I know of, 8 turbos have failed in a space of a couple of years. The turbo is a Leybold / Oerlikon TW 290/20/20.

I wondered if anyone else has experienced such a high rate of turbo failure?

Helen

[Or Not]

Hi Helen,

We had one fail last year, after having our Q-Exactive for about 6 months. Someone told me they had a bad lot of them, but don't know for sure if that's true.

[hfredricks]

It sure does seem like it!

with your turbo that makes 10 turbos that have failed in 6 instruments within 2 years of purchase - mostly within one year. unbelievable. We did not figure on a yearly replacement of a turbopump.

Helen




[quote="Or Not"]
Hi Helen,

We had one fail last year, after having our Q-Exactive for about 6 months. Someone told me they had a bad lot of them, but don't know for sure if that's true.[/quote]

[zacmcd2509]

Ours is about to fail too. Just over 18 months old. Thermo needs to explain what is going on here. How can these pumps start failing from 6 months!!

[dr_bahram1977]

If your pump fails too spn, and you've run it normally, it's manufacturing problem for sure.
I've Pefiffer pumps with over 50,000 hours of work time and still no problem. I've seen same about thermo's GCMS instruments too. write them and ask for replacement pump from other makes.

[James_Ball]

Is it just the bearings that are failing like normally happens in turbo pumps, or something else?

We have a Leybold on our ABI3200 that has been running pretty much continuously since 2006, so hopefully Leybold just got a bad batch of parts at some time. May not be Thermo's fault if it was a problem with some sub component along the way.

[adler_LAB]

Are you sure that the pump is the problem. ?

Is the drive frequency ok ? Is the current power usage more "mA" than normally ?
Is the room temerature not to high so the turbo controler doesn't shut off ?


When you extract the pump, does it still run, can you manually move it ? cause thats the main thing that changes the price of the repairing.

we refurbish them usually. if you just send them to the factory and they change the bearings its about 4000 Euro here. If you want a pump in exchange its about 7000 Euro. This price changes a lot from pump to pump and what your actual damage is.

There are no used pumps to sell, cause they manly repair them in exchange. if you do have luck its probably to much money to repair it.

There is no need to exchange the pump every year, but if your device has lots of leaks and if you have a high flow with your gc, the pump live can variate a lot. we use hydrogen as carrier gas, that affects the live of the pumps a lot cause it runs with full speed 24/7

So if you really want to complain about the system make sure you have all the readbacks and the details of your method with flows. Also make a leakcheck to figure if your device doesnt cause the problem itself.


the warranty for a refurbished on is 1 year

[hfredricks]

Hi Adler_LAB
Speaking for myself and most of the stories I've heard from close colleagues, yes, the pump is the problem because it often fails catastrophically. When you're pulling parts of the rotor out of your mass spec its not difficult to see where the problem is.
I have a bunch of new & old mass specs here and never had to deal with a turbo problem, it seems to be specifically this model of pump, the Leybold TW 290-20-20 UHV.

I got a refurbished pump from mascom, they will send out a refurb. pump as soon as they receive yours for fixing, so its a fairly quick turnaround, that was the only company I could find. 3100/4250 euro depending on bearings or rotor repair.

http://www.mascom-bremen.de/html/en/vac ... 20290.html


Talking to Oerlikon/Leybold I got an offer to buy a new pump for ~$9,000 - that's with a 'generous discount' (Thermo asks ~$6,500!!) and they gave me some advice to spin up my turbo every week or so if it sitting not being used... ?? not at all useful when you never turn them off. They said the lube on the bearings separates if it is not spun up regularly, well guess what, the lube fails even if it is spinning, to the point of spectacular failure... metal shavings or torn up rotor everywhere. ugh.

[zacmcd2509]

Just heard about another catastrophic turbo pump failure in a Q Exactive...And now apparently even the replacement pumps are still a problem.

Thermo/Leybold do the right thing and recall the problem batch of pumps! Is avoiding the problem really worth the breakdown down in trust?

If your pump is making a high pitched whine get hold of your Thermo engineer asap, don't wait.

[amitkumar]

We are facing the same problem with the UHV Turbo pump. We got it replaced two times within six months and again very recently it failed for the third time. Although the pump is covered under warranty, we are wasting the instrument running time. We hope, the manufacturer comes up with the improvised version which will last long.

[zacmcd2509]

@amitkhumar
What happened when it failed? Did the pump seize up?

We have also replaced the turbo for a second time as we were getting unexplained loss of vacuum. Not sure if pump was actual problem, but interested to know what is experienced when turbo fails or misbehaves

[amitkumar]

@zacmcd2509,
UHV TMP speed went to zero when the it failed. When restarted, UHV TMP could not clock the speed and forepump was running OK.

And unfortunately, again the TMP failed today.

[oursquiparle]

Hi everyone,

We got our TW290/20/20 replaced and I thought I'd share the knowledge with you.

It turns out it is a known problem.
Main cause is baking out the qex too long.

When baked out, orbi heats up, turbo heats up by conduction. When it's for too long, the turbo grease breaks down, the helixes start to wobble more and more, leading to contact and catastrophic failure.

In some cases, they would replace the turbo and the staff would go for a 40 hours bake, hoping to get the best vacuum, which would kill the turbo right away. The issue only stopped when they did bake-outs less than 10 hours.

In the new qex systems they have upgraded the cooling fan and software for the UHV turbo.

But if you don't have the new system, the main point is: never ever bake-out more than 10 hours. Just bake 8 hours and it should give you a decent lifetime.
If you want the best vacuum possible with a long bake-out, you'll have to think of a clever way to cool down the turbo during the bake-out.

I confirm that getting a new TW290/20/20 with controller is around 7k-ish.
I tried to just get a pump from Leybold/Oerlikon dealers and they'd ask at least 10k for the pump only so you might as well go with TF. And they are pretty quick and competent.


Hope it helps.

Cheers