Page 1 of 1

Varian 3400cx baseline issue

Posted: Thu Jul 20, 2006 2:34 pm
by erazumich
Hello.

I've come to see if anyone could give some advice on my problem.

I have a two Varian 3400cx GCs that are used for PCB analysis.

They have Ni 63 ECDs and J&W DB-1 columns.

We have a recurring problem with high baselines.

Optimal conditions: we have a millivolt reading around 2-3mV

We can operate, with care, on a 4-5 mV baseline.

But recently, we've seen readings climb as high as 20-25mV.

By running a few 'blank' samples, I can bring the baseline down slightly, but not quite to where they can be used effectively.

We think it may be stem from two things.

The baselines will rise, and have always risen slightly, after periods of inactivity, if for example, we do not do an overnight run, and even more so over a weekend. But it has become more extreme of late.

Another factor may be that we have had a recurring issue with brief power outages. Blips.

These may be incidental factors.

Has anyone else had issues with these instruments like this? Or have any advice?

Thank you.

ER

Posted: Thu Jul 20, 2006 3:03 pm
by gcguy
I have seen a similar effect on the Agilent ECD. Best way to sort it is to bake the detector. Consult your handbook to check for the best conditions for doing this. Basically the hotter the better but be aware of the column, you may need to remove it and blank off the detector.

gcguy

Posted: Fri Jul 21, 2006 2:05 pm
by erazumich
Thank you.

We are attempting a bake out today.

ER

Posted: Mon Jul 24, 2006 6:45 pm
by Rick
Hi!
First, disconnect the column to detector and cap the detector inlet.

Start your oven temperature program and check your baseline.
If you have your problem again, the problem is probably gases, contamination or leak

If you have not your problem, the problem is your column, injector contamination, gas or trap contamination.


If you have a second detector ECD on the same GC,you can connect your column on this second ECD..if your problem is already present, injector, gas or column is your problem