Nothing but noise!

[ali4683]

Hi all,

We recently had a power shutdown in the building which houses our GC-MS (Agilent 7890-5975). As I was mid-vent preparing to shut the machine down, the power unexpectedly went off/on twice- a complete nightmare.
I shut the whole machine down as soon as I could and turned it on again last week giving it time for any air in the system to resolve. Yesterday I attempted to run a few samples (for fatty acid analysis), but got some very strange results.
Basically the samples and controls looked like noise, identical to my blank hexane runs, no other peaks present, which is very unusual. No method parameters have been changed, samples are identical to those run before the power cut so nothing else has changed. The reason for the power cut was to alter the electrical supply in our building in some way, don't ask me how, but could this result in a noisy detector and be the cause of my problems??
Any ideas welcome!

Alison

[krickos]

Hi

Generally speaking I have experianced many unplanned power cuts over the years but none resulting in a total havoc of the detector.

You need to isolate the error a bit. Did do any air leak testing and/or tuning before starting analysis would indicate if MSD is functional or not.

[ali4683]

Yes the tune looks good and there is no air leak- there was a bit of air in the few days after pumping down the machine as expected, but everything has settled now. Thanks for your reply

[Don_Hilton]

If the instrument tunes - the mass spec is working and should gibe you some idea of the noise level in teh mass spec. The fact that you saw peaks from your hexane blank again indicates the mass spec is working.

Unless you see noise indicated in the tune reports, I would look toward a chromatographic problem. Start at the gas bottle and make sure everthing is correct. (gas pressure present, septum changed, leak checks around inlet made, etc.) The try a standard mix at high conentration to see if you can find the peak.

I've known of a case were there was a lot of work put into trouble shooting an instrument and tweaking everything - only to find out the wrong bottle had been used to make a dilution for the check standard...

Check from one end of the column flow to the other.

If there is additional information, the area to search could be made more specific.

Is it safe to assume the inlet was cool when the power failure occured?

[ali4683]

Hi,

Thanks for your help. Yes I have realised that since my tunes are indicating no noise that it is likely to be a chromatographic problem rather than with the detector. I will need to do some detective work.
To answer your question- no, unfortunately the power shutdown was unexpected therefore the inlets and oven were not cooled at the time. Do you know what kind of problems this could cause? I am not totally sure what kind of damage to look for.

Thanks,
Alison

[AICMM]

ali4683,

I would start by manually injecting a uL of air with a separate syringe and watching my scan profile for the air peak. If you don't see the air then you have either an inlet or column problem. If you see the air peak, I would then look at the autosampler as the source. This would be the first divide and conquer I would do.

Best regards,

AICMM

[Don_Hilton]

The question about hot inlets was because of potential damage to the inlet end of the column.

First check to be sure you can get an air peak through the column as mentioned above. This will show that you have working inlet and the flow path from inlet to detector is working.

The nest question is the condition of inlet and column. The question about inlet temperature was becase some columns to not do well if they are hot and exposed to air - depends on the stationary phase.

If I have a question about the condition of the inlet, I change liner and septum. (And if it analysis that is particularly sensitive to the condition of the seal plate, that too.)

Next step is to run a mixture where you can find peaks. You may need to try somethign less sensitive to active sites that fatty acids - perhaps some paraffins.

If you can find peaks from a standard - look at the shape. If the peaks tail badly, there may be activity from damaged stationary phase. Trimming the front end of the column may take care of the problem.

[Don_Hilton]

And, by the way, if you are not sure about what you are doing on diagnosing a column - ask before trimming the column. It is always easier to cut a column shorter than to make it longer again...