Reduced sensitivity with FID on 5890

[osp001]

My solvent peak is vastly reduced, and my analyte peak (when I can find it) is reduced probably 3-4 orders of magnitude. Can't figure out why. Running waxes on an RTX-5 capillary column, hydrogen carrier gas. Column has been kept below maximum operating temp of 350C. Went from "intermittent" to "signal always reduced".

Things I've tried:

New septum. New liner. New ferrule on FID side of column (column nut was loose- can't figure out why, so I slapped on a new ferrule, no dice). FID is clean. Double-checked column "tail" length.

Tried different solvents- analyte in chloroform vs. analyte in methanol. Solvent peaks markedly reduced in both.

Attenuator set to 0.

Increased inlet pressure- no joy.

Changed out water trap on gas supply.

Tried different needles- no change.

Haven't changed the split ratio.

For a while, I was getting a normal-ish response maybe every 6th or 7th injection. Now I've gotten nothing but markedly smaller peaks for 3-4 days.

Could this be a problem with the jet? I've not torn down the FID that far before.

[Himar]

I would try a manual injection and monitor the response.

Try changing to splitless or increase injection volume just to see if the detector responds as it should.

[dblux_]

...
Tried different needles- no change.
...

What if the fault is in the barrel or plunger ? Change the whole syringe.

But first of all check tightness of all joints from the column outlet to the detector. Check if possibly present capillary adaptor isn't loose.

[Peter Apps]

A leak at the inlet end is a likely cause. Check with a leak seeker.

If you are using an autosampler check that it is actually pulling the plunger back - a mechanical connection working loose would also fit the symptoms.

Check the actual flow out of the split vent and the septum purge are what they should be - use an independent flow meter, not the onboard pneumatics of the GC.

There might be something loose in the FID, check that the jet is screwed down firmly (but DO NOT overtighten it) and than the electical contacts are good.

What instrument are you using ?

Peter

[dblux_]

...What instrument are you using ?

Peter

I hope it's still 5890

[Consumer Products Guy]

Reduced solvent peak indicates to me injection/syringe issue or a leak.

If just sample was being reduced I would've suggested some wax was kicking out of solution so less being injected, but the solvent peak thing leads me away from that.

[osp001]

Sorry to be so glib- instead of saying "needle," I should be saying syringe + needle. I get the same results even when I use a different syringe + needle. I am doing all injections manually.

I swapped out the inlet connection- trimmed the end, checked the old end under the 'scope, doesn't look like it was plugged by anything. Threw on a new ferrule, hooked it back up- still no joy.

I tried doubling up the injection, got a solvent peak that was substantially larger than usual (~6 volts instead of ~5 volts), and today I'm not even getting an analyte peak.

Tomorrow when everything's cold, I'll go over the injector port itself, as well as meter the purge and check the jet. This one is turning out to be a bit of a stumper for me.

[osp001]

It was the jet!

I tore down the detector, yanked the jet, threw it under the dissecting scope- couldn't see any gas path, so I ultrasonicated it and gave it another look. Still no joy. Not having any brass wire, I pulled some clean 30 ga copper magnet wire, unplugged the nozzle and gave it another ultrasonic bath in IPA.

Dried it out, re-installed it (finger tight), reassembled the detector. After cooking everything off, I ran a sample and- much joy.

Thanks, y'all!

[Peter Apps]

Result !

Thanks for the feedback.

Peter