Baseline jumps FID

[scottythree]

I am just starting to troubleshoot this for an analyst. They sent me a pic of what they are seeing but have not sent me parameters yet. I'm hoping the pic will trigger a response as to where the problem may be. Thanks up front.

[Peter Apps]

If you are doing troubleshooting you need to train the analysts from the beginning that you are not gong to waste your time by even looking at a chromatogram that does not have full operational details to go with it.

Peter

[Johnny Rod]

Peter's right. Someone isn't killing the pressure in your nitrogen (FID make up gas) line?

[scottythree]

Got some info:
Agilent DB-Wax 30-m x 0.32-mm, 0.5-um df
Oven 90C for 7 min, 10C/min to 220C, hold 10 min
Nitrogen, flow rate 0.9 mL/min, constant flow (5.4 psi, 19 cm/ sec with Oven at 90C)
Inlet temp = 250c
Split ratio = 10
FID at 250C
H2 =40
Air =400
Constant make up N2 at 20.0 mL/min
Range =0
Diluent is 20 mL of GAA in 1L of water (internal std of IPA too)
Tower injection, 3uL

Retention of peaks are OK. Peaks are not shown in the chromatogram though. The problem is with the random vertical jumps in the baseline.

[dblux_]

Peter's right. Someone isn't killing the pressure in your nitrogen (FID make up gas) line?

Very likely. Is the GC in question the only instrument connected to this source of nitrogen ?

Or (if GC electronics is technically obsolete) voltage fluctuation may cause sudden baseline shifts.

Leaks at FID end of column connected with temperature changes of the oven ?

[scottythree]

The GC is a 6890N built in 2004. It is the only instrument connected to the gas tanks. I'll check the detector end for leaks.

[Peter Apps]

It is also very necessary for the analysts to include the time scale on the chromatograms so that you know whether you are looking only at the couple of minutes before the solvent peak elutes.

And they should not use acronyms that other people might not understand.

Peter

[CE Instruments]

Water based injection.... Check for gas flow changes and leaks. Could be electronic and your FID amp is at fault. However it might just be a dirty FID, they don't like water injections it could even be corroded. Remove the jet , clean and inspect. 3ul injection at 10:1 split, fast injection ? If so the volume is too large for the liner , could it have backed up into your pneumatics and they are at fault ?

Take a second opinion but I would be looking for a better method like headspace if possible ?

[dblux_]

The GC is a 6890N built in 2004. It is the only instrument connected to the gas tanks. I'll check the detector end for leaks.

This instrument is very resistant to voltage fluctuaction so at least one reason may be excluded.


BTW - my 6890N works properly even when voltage drops to 200 V AC (Europe - 230 V nominal).

[GasMan]

You do not mention what you are using to show the chromatogram. I presume you are using a data system of some description. You mention that you are using Range 0. This would only be used with an analog output going to an A/D converter. You need to check if this part is not giving the problem.

Is the problem repeatable? Does it always appear at the same position in the chromatogram. A simple test would be to put the 6890 into the test chromatogram output mode. This produces three peaks over a period of 1.5 minutes. The second 'peak' is 10 times smaller than the first peak and the third peak is 10 times smaller than the second peak. The cycle is repeated until you press stop run. This test will show you if you really have a FID problem or if the problem is in the data handling side of your system.

Gasman

[dblux_]

I would start with checking whether 1/4'' nut of adaptable adapter (the one hidden under the insulating cup) is tight - of course if there is adaptable adapter, not capillary optimized fitting.

[scottythree]

The graphite ferrule on the adapter to the FID has chipped off in places and the adapter is looking a bit well used. I'm changing it out so we will see!

We collect data in TotalChrom using the analog outs. The analyst tells me they have seen this problem off and on with other runs (not set up with these exact parameters). They did install a new jet for this run however so can't attribute this issue to a clogging jet. The jumps are random, sometimes before, during and after peaks.

[Johnny Rod]

I'm not clear what you've posted a screen shot of - is this a run, or just the baseline monitoring with or without one or more runs? Have you waggled the connections of the data cables? Checked the connections to the FID collector etc. in the GC?

[chromatographer1]

I think it may be a cracked (or dirty) electrical connection which occasionally disconnects, via heat expansion or vibration.

best wishes,

Rod

[scottythree]

We have not seen the jump since changing the adapter. I'll be keeping an eye on the system. The pic shows part of the baseline from one chromatogram zoomed in to show the problem.