Drift in the area during analysis

[demper]

During calibration (mix of gases) in the GC we have a drift in the area (upwards or downwards) for 6 injections, RSD is over 1%. We have to reanalyze the calibration to get stable values. What’s the problem?

The analysis takes a lot of time and I don’t want to increase the injections to get stable values.

[Johnny Rod]

You'll need to give us rather more info than that I'm afraid

[Don_Hilton]

What instrumentation are you using? Detetor type might play a role. And, describe the inlet sampling system including the way that you ensure the pressure of the gas is constant from sample to sample.

[demper]

We are using Agilent 6890 GC. Molsieve and Porapak in series to a TCD detector.
We have a flowmeter to controll a stable sample flow into the GC.
We have a mixture of CH4, CO, O2 in N2.
We run QA samples each week to verify that instrument is ok until next QA sample. These QA samples have not been stable lately and have to run more than one calibration a week.
The instrument is in a room inside the production local and the room has same atmospheric pressure as in the production local.

Im not so good in Chromatograpy and need help to know whats happening?

[chromatographer1]

1% RSD values are pretty good. 5% not so good.

You may have a leaking sample valve.

If your sieve column is getting contaminated an extended time at 200C might be a good idea.

You may also not be getting enough sample flushing the sample loop before it is injected.

Your valve timing may be slightly off. You may be losing a part of a peak which can affect your values.

Good luck,

Rod