Tube leak testing for the PE TurboMatrix 150

[passy]

Hello all, hope I'm posting this in the right place.

Ok, a bit of a story. I've just restarted work with thermal desorption-GC-MS after a long layoff.

We have a Perkin-Elmer TurboMatrix 150 Automated Thermal Desorber going into a Varian 3800 GC-MS.

The problem (actually one of a few) we have with it is when a tube fails its leak test, the ATD will then try the next tube and if that succeeds, it will continue the thermal desorption process and inject the sample. Because the ATD and the Varian GC-MS are controlled with different software, the sequence we have with the GC-MS will be then be out of step with what is injected from the ATD. When a tube fails a leak test the ATD will then attempt up to two more subsequent tubes before shutting down with an error. The GC-MS acquisition sequence can be well out of step with what is going onto with the ATD. After each run we have to make sure to check the logfile before being sure of what is actually been injected.

Its a ridiculous state of affairs. I can't recall this being the case when I worked with the PE 650 series but may be mistaken (its been a number of years now).

Reading the TurboMatrix manual on pages 309-310 it states that the number of leak test retries can be set between zero and two (two being default, which is what we have) - I wish to change this to zero so if it fails a leak test, the process then shuts down immediately and we'll have no confusion.

My problem is *how* to change this value - the manual says nothing apart from it can be. I've looked over of the preferences/setting/method editor pages of the ATD touchscreen and no luck.

Anyone out there know?

We're getting a Markes system in the next few months but this is our go to instrument until then.

Many thanks in advance.

[iainconlin]

I think what you need to do is a small wiring fix so that, when the tube fails to run on the ATD it sends a signal to the GC to run. You will have a blank chromatogram but the sequence order is kept I believe this is irrespective of turbomatrix model number.

The section of the manual you need to look at is the Fail Output Signal (page 44 in the manual I have), you must set your cycle time to zero though.

This fix certainly works well for our labs.

Interesting that you are moving away from PE to Markes we may do the opposite.

Iain

[passy]

Many thanks for your reply.

You must have the same manual - that fix will do exactly what we want. Cheers!!

> Interesting that you are moving away from PE to Markes we may do the opposite.

We've had nothing but trouble with this PE system; I'm here at work over the Easter weekend because it is so unreliable. I came here to discover that the run I put on Thursday stopped two tubes after I left. Now as its been more than 24 hours since the standards have been run I now have to spike another set for the remaining tubes on the list. Not happy.

I'd be interested in hearing your opinions/experiences of the Markes systems.

Many thanks again for the reference.

[iainconlin]

We have two Markes Unity's with Ultra's attached. Our instruments will be an older model than the instruments you are getting so I would be slightly cautious of making a direct comparison.

I do not think the leak check on the Markes is as good as the PE system, I had some trouble a few months ago and was very dissappointed in the service response from Markes. I've had excellent service from PE engineers in three different companies I've worked for.

I also prefer being able to control the thermal desorbers directly on the instrument rather than through a PC, saves a little on lab space which is not abundant here. On the system we have every time I switch them off the firmware has to be reloaded (this is a design feature not an instrument fault).

I'm based in the UK, my colleagues in the US who have the Turbomatrix also have had good service from their local PE engineers. It is their positive experience with the Turbomatrix which is making me consider the PE system.

I also prefer how the Turbomatrix controls the column flow and splits compared to the Markes system which seems to be GC dependant.

Have you spoken to the engineers regarding the poor performance of your system?

Although not a problem with the compounds I am analysing at the moment the Peltier cooler on the Unity I have is certainly not as good as the Perkin Elmer, this is something I hope they have corrected on the new versions (to get to 0oC I have to take a cover off!).

If we do get the Turbomatrix and I have the same problems as you it will be a PE engineer losing his weekend not me!

I hope things improve with your new system. If you have any things to watch for on the Turbomatrix let me know.

Iain