Liquid CO2 cooling of injector (1077) not working?

[north90ty]

We have a Varian 3800gc with a liquid CO2 cooled injector.
The injector heats up to 260C and has to cool down to 60C for the next cycle.
The injector is equipped with CO2 cooling, but this doesn't seem to work:
- It takes between 30-45 minutes for the injector to cool down.
- Also it seems that the cooling takes longer after each injection.
- Shutting off the liquid CO2 doesn't really slow down the cooling.

My questions:
- can anybody give a reference of how quick their injector cools down on liquid co2?
- how should the liquid co2 be hooked up to the gc? (diameter lines/regulator pressure/etc)
- did anyone experiece similar trouble and found the cause?

Thanks,
Frank

[Ultimate Science 101]

As a rough estimate, the 1079/1077 should cool (with liquid CO2) around 2-3 degrees Celsius per second. This is at least the case when the coolant turns on and the injector is quite hot. As it gets down to below 100C, you will probably notice it cools slower. Maybe 1-2C/sec, just as a rough estimate. It should be able to reach around -65C with liquid CO2 cooling.

Did you hookup the coolant kit yourself, or was it already factory assembled on the 3800 GC when you received it? Guessing if it was done at the factory, they used the recommended inner diameters and lengths of PEEK tubing. We can check early next week and let you know the inner diameter of PEEK we've used in the past for installing cooling capability on a 1079 injector.

One issue we have noticed is with the cooling valve. Sometimes it "sticks" for lacks of a better word, not turning off when it should. This can damage the valve if left "clicking" for too long. In any event, if this is part of your problem, power off the 3800 and disconnect the valve from the GC electronics. Carefully take the valve apart and clean it thoroughly. There is a filter inside that may be clogged. Clean appropriately, reassemble, and then see where the system stands.

Regarding the liquid CO2 tank, have you tried multiple tanks and still experience the problem you're having? What type of regulator are you using? The ones we've worked with in the past don't have pressure gauges. Try changing your tank if you've only used 1 so far.

Good luck!

[Peter Apps]

Hi Frank

The simplest explanation is that the CO2 tank is empty - you need to check that before you go any further.

I have the later version of your hardware (a 1079 inlet on a 450GC) and I have simply connected the CO2 tank to the back of the GC using 1/8 inch copper pipe. You must use a CO2 tank with a dip tube that takes liquid from the bottom of the tank, not gas from the top - the gas industry calls this an eductor tube I think - the tanks I use have a yellow stripe down the side to show that they have the dip tube. For effective cooling the CO2 needs to reach the inlet as a liquid - so you should not use any regulators.

Peter

[north90ty]

The GC in question has been equipped with CO2 cooling for years and is probably installed by the manufacturer. It did work well in the past (according to collegues). But the cooling feature hasn't been used for a while, the gc has moved around labs etc.

We have had multiple tanks connected so i don't think it is (just) the tank. We do use tank designed for liquid co2 delivery.
We did have a regulator between the tank and the line, which we now removed.
The liquid CO2 flows from the tank through 1/8" copper to the GC.
There is no filter inline betweren the tank and the valve on the gc. We ordered a filter.
The valve on the GC seems to be working, we tested the valve independently (but with no CO2 pressure on it).

I am starting to think that the ?expansion nozzle? in the injector might be plugged with dirt?
Can we take the injector apart and check for this? Any other ideas?

Thanks, Frank!

[north90ty]

SOLVED:
some line in the injector was plugged. We forced compressed are backwards through the liquid CO2 cooling system to blow out the particles that possibly plugged the lines. After reconnecting the CO2 the injector cools down from 300 to 60 degrees in about 6 minutes.
Thanks for all the advise!

Frank

[Peter Apps]

Hi Frank

Thanks for the feedback - the backflush is a good idea.

Peter