Problem with Saturn 2000 GC-MS, possible leak

[Potter]

I recently replaced the turbopump on my Saturn 2000 GC-MS after it had been sitting unused for around 6 months. I immediately noticed that there was a problem because my system would fail the auto-air/water check yet my spectrum showed no air or water!

After speaking with Varian tech support I found out that my system could not see ions below 40 m/z despite numerous efforts by me and Varian. I ended up cleaning the ion trap as a last resort this past Friday and letting my system bakeout over the weekend.

This monday I found that my system could now properly see ions across the entire size range (10-650). But now the detector is seeing maxed out intensity across the entire range!

I was under the assumption that if my turbopump was properly reaching 100% speed then I did not have a leak, but Varian support told me that I could still have a leak. They also seem convinced that an air leak is causing the maxed intensities on my spectrum. I have spent two days now checking every o-ring and seal that I can find but I cannot locate the leak.

I read about a popular leak-detecting method involving spraying compressed air around the possible leak locations but this does not work as my intensities are all maxed out so I cannot see the peak for the air propellant.

Any suggestions for possible leak locations or other solutions would be greatly appreciated.

Thank you

edit: I do not think the problem/leak could be in my GC because before cleaning the ion trap the system was seeing all ions fine (above 40 m/z) and I did not make any alteration to the GC while cleaning the ion trap.

[Peter Apps]

The first step is to eliminate the GC as a possible source of the problem by capping the transfer line connection to the column.

If you see maximum intensity on all ions it sounds more like a detector or electronics problem than a leak. What do you see if you decrease the detector voltage ?

Peter

[Potter]

I capped the column yesterday afternoon and still saw a maxed-out signal so I'm pretty sure the problem is in the MS. Currently my EM voltage is 2100 and I have tried lowering it to 1900 with no difference. I wasn't sure if there was a lower limit so I didn't try to go any lower than 1900.

[Peter Apps]

As far as I know there is no lower limit on an electron multiplier (except that at a certain point it stops multiplying). If there is a recommended range for your instrument it will be in the manual, or the software help, or values below the minimum will not be accepted when you try them. Try setting it at the minimum that you can get the software to accept.

This is all generic, we need an expert of Saturns to chip in.

Peter

[labstephan]

Did you check the RF adjustment and Cal gas adjustment?
I use my Saturn not on a regular basis and it is almost not evacuated. Everytime I want to use it I have to check and re-adjust this two settings.
Sometimes there a realy big differences between the old and new settings. And nobody is playing with the settings in the meantime!

After this a multiplier gain adjustent may be helpful.

Hope this helps

Best regards
Stephan

[Ultimate Science 101]

Have you ran a diagnostics check in MS Workstation? If so, does everything pass? If not, please specify what error messages are generated.

When you run the daily checks method, what is the ionization time in segment 2? Close to 25,000? Also check the ion count, hopefully it's less than 7,000. When you turn off the filament and RF but leave the multiplier on (in segment 2), what does the 100% value read? Ideally, it should be 1 or 2, but definitely less than 10.

I would check the RF adjustment under your manual control tab. Make sure it is within specification and adjust the screw inside the 2000MS door if an adjustment needs to be made. As another member said, also check your cal gas pressure (also in the manual control tab). Repeat a full FC43/detector tune after this and try an analysis to see how things look.

If all this fails, there is a reset button the SapWave board that you can try pushing.

At this point, if the system still is not back to normal, I'm wondering if the problem will turn out to be detector related. It may need replacing at this point of troubleshooting.

Please let us know if we can be of assistance. Good luck!

[Potter]

- I have determined that the range for my EM is 1000-3000 volts.

-My RF adjustment looked good yesterday and I couldnt calibrate calgas because I could not see it within the maxed-out signal.

-My diagnostic test failed once yesterday and reported that my EM was set for a different voltage than the setting I entered. The diagnostic test has passed all other times.

-Segment #2 in DailyChecks.mth is set for ionization time: 25,000 and my max intensity was 4095 counts.

I'm not sure if any of this information is valid anymore because my problem has changed. I cleaned my ion trap yesterday according to the manual (sonicate in methanol). I also sprayed the trap with some DustOff once it was installed. Upon lowering my EM voltage to the minimum (1000V) I no longer had a maxed out signal. The intensities for all masses are still quite high, but not maxed-out. I left my lab for the day and set the instrument to bakeout overnight. Upon returning this morning the situation is still the same, the signal is almost maxed-out across the size range. The peaks all have a very blocky shape.

Thank you all for your help, I know it is difficult to offer advice when my problem keeps changing.

[labstephan]

Hi Potter,

just for understanding. Do we talk about chromatographic peaks or mass peaks?

Did you check the RF adjustment again after the bakeout time? Also I would suggest you to close the Cal gas valve completely an look if the mass peaks will decrease. If they get smaller perhaps your cal gas valve was opened to much.
I have to readjust my cal gas adjustments every time I reactivate my saturn system. The equilibration time is also "very" long. It is best to wait for 1-2 minutes after each time you make a turn on the valve.
I also have in mind that the cal gas valve may be a little faulty (I think my maybe is). Purging the cal gas line before adjustment may be also very helpful.

Has your cal cas valve been changed in the past? Maybe somebody has changed it against a CI valve wich is operated counter clockwise (or vice versa the cal gas valve).

Best regards

Stephan

[Potter]

I have only been viewing mass peaks.

I have checked my RF adjustment and it looks good. I am going to retry my calgas adjustment because I didnt know that it would take so long after each turn of the valve to register a change in flow. I did isolate a leak today but I'm not sure of the identity of the part. Directly above my calgas vial is a solenoid with a yellow cap and a hose that goes to the plastic elbow on the turbopump. I made sure the hose was tight and I tightened the bolt that the hose fits over, but I didn't want to mess with the solenoid itself. I just checked and the leak is still there. Does anybody know what this part is and what purpose it serves?

Thank you

[Potter]

Hello everyone,

I have been able to fix a leak. Notice that I said A leak and not THE leak, haha. There was a leaky o-ring in the solenoid above the calibration gas vial. I used compressed air to test the new o-ring and now the leak is gone at that spot.

Unfortunately my system is still seeing very high water and moderate air. I have previously isolated the problem to the mass-spec but now I am stuck and cannot find any other leaks. Does anybody know of any uncommon locations for small leaks in the instrument?

Thank you

[labstephan]

Hi Potter,
did you check your Transferline? There is also a O-ring which may need replacement.

I had leaks in my system at the lower plate of the cal gas block. Perhaps you have to retighten the plate. Also have a look at the O-rings of the CI gas restrictor loop.
In my case these seals were missing. I bought the system used without any information about it. Stupid but very cost effective. In noticed that the turbo speed went down when I tried to use CI. This behavior was also present with the CI valve closed, so I removed the lower plate and saw that no seal (O-rings) are in place. The whole thing coct me approx. 4 weeks (I only did it at weekends) to find and fix.

Another idea is to keep the GC Injector (did you leak check the injector) in split mode. Agilent 5890 and 6890 models are known to "create" a little air leak in splitless mode.

Hope this helps.

regards

Stephan