I am using a capillary column dimensions 30m x 0.53mm x 1um. My column head pressure is set to 18 psi. My P&t has its own pressure control and its set to 6.5 psi. I have a column flow of 7 ml/min. My carrrier gas/AUX gas flow is 29.1 ml/min (that's with the column flow subtracted) My air is set to 260 ml/min and my Hydrogen is set to 38 ml/min. Those are fairly close to where they were when I was hired but I am completely open to changing them to have a more optimal run. I just have no idea what the flows should be for the best results on my instrument.
I believe both the P&T is supplying Carrier gas as well the GC has a line going to it from the tank. They have (I believe) independent controllers. The GC has a Column Head Pressure Controller and the P&T has a Trap Pressure Controller.
I have run a similar setup as you describe. What you have is the older non purged packed/direct inject inlet in a 5890. There should be only one knob on the front of the GC with a pressure gauge which reads head pressure. The carrier line is probably spliced on the left side of the GC so that it goes from the flow controller in the GC to the back of the Tekmar 3000 to a port called Carrier Gas In. It then goes through the oven and returns to the injection port by the heated transfer line which is spliced into the other half of the line that would normally go from the GC Flow controller to the GC Inlet. This is how we had our old P&T GC/MS setup years ago with the 105mx0.53 Rtx 502.2 column then into jet separator before going into the MS. I also used the same setup on an FID minus the Jet Separator for doing GRO analysis.
The pressure gauge on the P&T that is labeled Trap Back Pressure adjust the backpressure on the outlet of the trap during purging to give sharper peaks by compressing the adsorbtion bands of the analytes in the trap. This setting has nothing to do with the carrier flow at all, only the purge flow of the sample path.
If it is the older non-purged inlet you will have more problems with septum bleed. If you are not using the septa for injections you may want to us a teflon faced septa and see if that helps. You could also be having problems from the column placed too high up into the injection port. I had an analyst do this once and he was getting really bad chromatograms with humps and split peaks, I checked the inlet and he had pushed the column nearly up to the septa, it should only be installed a few millimeters above the ferrule, at the bottom of the inlet. We also have run the setup bypassing the injection port completely just using a zero dead volume connector to join the transfer line to the column. Flow control remains the same you just lose the ability to do a manual injection for trouble shooting purposes. You can actually remove the injection port completely while keeping the flow controller in place. With the heater and thermocouple disconnected the GC will simply think the port is not there and not give any errors.
As for the drifting baseline, another problem could be the multiport valve in the P&T oven. It can become contaminated or it can leak causing similar problems. It is a Valco valve with a vespel rotor. The vespel section can adsorb contaminates and cause problems or it can become scratch and leak giving erratic flows. You can remove the vespel rotor by unscrewing the knurled fitting on top of the valve(not the hex screw in the center, that adjusts the tension on the valve) then stepping the instrument to rotate the rotor which will knock it loose then lift it out with forceps or a strong magnet. You can inspect the rotor, it will have small notches that connect adjoining ports together. Clean the notches with a swab soaked in Methanol, or sonicate in methanol if you think it is contaminated. The valve on the 16 position is similar but is a total pain in the you know what to get realigned, but if you ever take it apart I do have a trick to make realignment easy
There also used to be a setup sold that would allow you to connect a syringe to the top of the trap connector, and backflush the whole flow path with methanol or water back to the sparger. I used to do this while it was hot to steam clean the system when a sample foamed.
Also make sure your column is not installed to high in the detector as it can cause problems with baseline too. A deformed jet tip can also cause strange problems with signals, have you tried another jet yet?
