Hello,

For sometime now I have been trying to fix a problem baseline drift problem with my ECD signal.

I have a 7890 GC configured with a TCD, FID, and ECD. The ECD is in line with the back inlet and is set up with two columns and a 10 port switching valve that switches back and forth between the two. So in position A, one column is running a sample the other is back flushing and visa versa for position B. I am analyzing greenhouse gases (N2O on the ECD) and as such use packed columns (PP-Q Restek columns 3m, 2.0mm ID 1/8OD, 80/100).

My GC is just now a year old as well as my columns. When I first started running N2O samples on the machine the baseline was very nice and the switch transition between the valve position/columns was smooth. After I changed my first tank of p-10 carrier gas and after I had run maybe 150 samples, I noticed an increase in the baseline. Then I noticed a slow downward drift in the baseline after the valve switch. I have attached a screen shot http://i.imgur.com/E2l1b.png illustrating this.

On the screenshot the first peak in a air peak, then N2O, then a pressure peak indicating the valve switch. You can see the different valve positions/columns and how the one drifts downward throughout the run.

I immediately suspected a contaminated cylinder so switched it out, no change. I have checked for leaks from the cylinder to the inlet and back...no leaks. I have swapped around columns to see if it was a bad column, but no change. I had a service technician in who eventually replaced the switching valve. It seemed to temporarily help the problem, but after a few sample runs the drift came back.

Even when I do a run without making injections, I notice the baseline drift after the valve switches.

Any suggestions as to what could cause this or other things to try to and pinpoint the problem?

I would appreciate any help or suggestions. Let me know if you need any more information.

Thank you for your help.