ill functioning MSD...high EMvolts, leak and flow error

[JFo]

Hey guys,

Our GC-MS is playing up at hte moment. we have a GC 6890 connected to a 5973 MSD. I starting to notice someof my lower sensitivity compounds had dissappeared in my runs and even at 2 x concentration as previously being able to be detected I could not detect these specific compounds.

We then realised the EMvolts on the machine were increasing. Having been previously at ~1800-1900 EMVolts, they went up to 2200 volts. We cleaned the source which didn't change much.

Over the next 2 months the EMvolts went up to 2400 volts and over the course of a week went up to 2600 volts.

We have noticed a leak which isn't consistent with any specific method parameters i.e. flow rate, split or splitless. Generally the leak is at ~7% nitrogen, which can go up to 40% using certain parameters, but is not consisten with these parameters.

We also recently noticed in the auto-tune that the flow rate is not the same in the tune report as the flow-rate that we set the method or machine to.

Does anyone know a reason for any of these problems? and how to fix them? At one stage I thought it may be the EMhorn dying, but I'm not sure whether this is correct now that we have discovered all these other problems.

Any help would be greatly appreciated.

Thanks,
Jade

[tlahren]

At 2600 V your EM is dead. I would get a new one and then see if you have any other troubles. Make sure you have NO leaks though. I usually order a new one when my auto-tune hits around 2000 V. I replace it at about 2200 V. After about 2000 V the noise get's exponentially worse. It's no longer a log-linear relation ship between gain and EM Volts. There is a way to print/view the gain calibration through Chemstation but I forget exactly how to do it. I think there is a file under /MSDChem/diag/gain.xls that will plot the curve. I don't remember how to get the values into the file though. I think you may have to run the gain.mac macro. Anyway, you should see that it gets less linear around 2000 V. At this point all that increases as you increase Voltage is the noise.

The tune file should have nothing to do with your He flow rate. Unless you are running CI then it will control the reagent gas (Methane) flow rate. Split rate etc. should have no effect on tuning as all that makes it to the MSD is the flow through the column (i.e., whatever you have set up for a flow rate in your analysis method).