GC/TCD to FID

[wrigcman]

I need to know how to adjust the sensitivity of my TCD. I've troubleshot all I know how to, to no avail. I have a 6890N GC and installed a TCD that vents to an FID, using a HP-PLOT Q of 30mx320x25um for the analysis of fixed gas and water samples and I can not get the TCD to show any significant peaks and just injected a methane sample to test FID and everything looks good on that detector. I tried just injecting air to test the TCD an nothing. Any suggestions? I used Agilent's on-line TCD troubleshooting tips and everything seems to check out, but no sensitivity. The filament was tested and is within range, both cold and hot. Maybe I just need optium parameters for this application, but I can't figure them out. Help?

[AICMM]

wrigcman,

First thing I would try is just hooking the column up to the TCD and disconnecting the FID. Or, perhaps a splitter (i.e. parallel) at the back of the column rather than serial.

What is the current signal on the TCD and what does it do if you change the make-up or reference gas flows? What are the flows through the TCD right now? TCDs are concentration dependent so that will have a big effect on sensitivity.

Best regards.

[wrigcman]

I'll try the suggestions... The current signal is suppose to be <30 I'm told and depending on the flow configurations it goes all over from 6300 to 35000. The only way to get <30 is if I run a undesirable column flow of around 0.5 mL/min. Still with no peaks from an air injection. I'm currently trying a 3.0 mL/min, 30 mL/min Reference flow and a makeup of 6. Signal around 6500. I'll test out the TCD w/o the FID connected and see what happens...

[wrigcman]

Ok, I performed your suggestions of disconnecting the FID and test TCD by itself. Same results, nothing as far as peaks. What is the correct parameters for my setup? Flows, temp program, etc... ? I'm using an HP-PLOT Q 30mx.32mmx20um and am intested in seeing O2/N2(Air) and Water in my injections. Thanks for any help

[AICMM]

wrigc,

You need to run a balance of flows on a TCD. That means that you should run at least 30 mL/min from the column plus make-up if you have 30 from reference. Set your column flow where you need it to be and then add make-up to get the flow up to near 30 mL/min. I would then adjust either the make-up or the reference at that point to get a reasonable zero value.

What column flow are you trying to run at right now?

Best regards.

[wrigcman]

I would like to run column flow anywhere from 1.0-2.0 mL/min with the particular column I'm using (HP-PLOTQ, 30mx.32mmx20um).

[wrigcman]

Ok, I tried your latest suggestions. The only problem is that I can only go as high as 12 mL/min on the makeup flow and still can not achieve the <30 value and still do not have peak response or sensitivity on the TCD.

[AICMM]

wrigcman,

sorry about the delayed reply, out of town for a couple of days.

You should be able to get more make-up flow than that and it is critical for balancing the TCD. Either of two things going on. One, the supply pressure is not high enough to get the flow you need. Two, the make-up controller is limited or clogged. The next thing I would do if it were my instrument is by-pass the make-up controller with a needle valve and supply the TCD with more make-up gas, till you get balanced flows.

Three by the ways. One, the TCD I work with most often is running at something like 80-90, not 30 so anything below about 200 will probably work. Two, did your signal drop as you increased the make-up? Three, turn the TCD filaments off when you measure flows because you could hurt the filaments by turning flows on and off.

Best regards.