FRONT INLET PRESSURE PROBLEM

[che313]

In Agilent 6890 an HP-5ms 30m, 0.25mm, 25um was installed. The flow was kept constant at 1mL/min and the pressure was not monitored. Some times an inlet pressure shutdown during running was faced without tracking at which temperature that was happen, an agilent rep. advice to change the septa whenever face this situation.

Due to stick ghost peak the coulmn is changed to RTX-1 30m, 0.32mm, 0.1um. Then chemstation run it refuses the flow of 1mL/min and the minimum accepted flow was 1.5mL/min.

As scheduled, service task was done by Agilent rep. where the inlet port was cleaned. He notice this problem and try to identify the reason but he doesn't sucsses!! We are running know under 1.2mL/min and the pressure from 1.4psi at 80C increses to 9.6 psi at 300C.

My questions are:
1- What is the ideal inlet pressure at 80C knowing that the cylinder outlet pressure is kept at 5 bar.
2. What is the source of decraesing the inlet prssure from its ideal to just 1.4psi??

[larkl]

1 psi sounds too low a pressure for the EPC to handle. What's the total flow (are you running split)? The EPC can't control well at very low flows.

[che313]

we are working on splitless but would like to work on split.
total flow of what?

[Peter Apps]

With a 0.32 mm i.d. column you need a flow of about 2 ml/min to provide an optimum linear gas velocity in the column. Also, if you set the flow at 1 ml/min the vacuum in the MS sucks faster than the EPC supplies gas, and so you get very low pressure or even a slight vacuum in the inlet, which is highly undesirable for all sorts of reasons. Set the linear flow to 35 cm/s and let the software sort out the pressure. The cylinder outlet pressure is irrelevant as long as it is about half a bar above the inlet pressure.

Changing the septum when you get a pressure shutdown is good advice, even better is to change the septa regularly enough to avoid the problem.

The total flow required for a split injection is the volume flow into the column multipled by the split ratio, plus the septum purge flow. You can set the split ratio on the software and let the electronics do it all.

Do you really call service reps in to clean your inlets ?

Peter

[che313]

thanks peter for your contripution

restek say the typical flow for .32 column is 1mL/min
look here http://www.restek.com/guide_gccolsel_sect3.asp

and regarding the service reps not only for inlet, mainly for ms source and this is done becuase the instrument is under warntey

[sassman]

The optimum flow rate in mL/min depends on temperature (and column diameter as you have discovered). Like Peter said, you should set the linear velocity instead (35 cm/sec) and let the instrument calculate the other parameters.

[gpronger]

What I suspect is happening is that the tank pressure is not sufficient to deliver the pressure needed at the end of the run. Depending upon the distance of the tank to the instrument and the ID of the tubing used their is always some level of pressure drop.

With older (manual pressure control) you had constant pressure, so when you set up the instrument (probably at the lower temperature) it just kept the pressure the same (actual column flow was decreasing, which may be why historically there was issues with later peaks being broader).

Go to EPC, and the system is trying to keep the flow constant. The narrower the column the more pressure is needed. Go to a wider column and you'll need more overall flow (ml/min) to maintain any particular linear velocity. With the Agilent system, when it can't achieve the needed pressure the GC starts making annoying noises and then shuts down.

My recommendation is increase the tank pressure significantly from where it is set and try another run with your original GC (including inlet) parameters.

[Don_Hilton]

I've run into this one before, as I often try to run closer to 23 cm/sec. And with a 30 m x 0.32 column running into a mass spectrometer, you have to have the inlet pressure at less than (or really close to) one atmosphere absolute pressure to do that. (And inlets don't work that way.)

Note that you changed from a 0.25 ID column to a 0.32 ID column. When you do this you must increase the flow rate (mL/min) just to maintain linear velocity. And because the pressure drop across a 0.32 ID column is less than a 0.25 column, you have to watch the inlet pressure. The reason you saw a drop in column head pressure was the increase in column diameter.

As pointed out in a couple of the posts above, the critical parameter is linear velocity. When you change column diameter, the mL/min must change to maintain separation efficiency. Your ideal inlet pressure is determined not by the cylinder pressure, but by what you need to maintain optimum flow (as linear velocity) and sufficient head pressure to avoid problems with the inlet.

[Consumer Products Guy]

Agilent web site has a good, and free, GC Flow Calculator utility at http://www.chem.agilent.com/CAG/servsup/usersoft/files/GCFC.htm

Also, your topic should be in the "GC" section because it's a GC issue, even though you may happen to have that GC hooked up to a MS detector.

[che313]

thanks for all of you

I set the velocity to 35cm/min and the flow becomes 0.9ml/min which can't be accepted by instrument.

So I increase the flow to 2ml/min and the pressure now is 6 psi. I run a sample and I'll see the result.

[Peter Apps]

Calculate the volume flow rate through a 0.32 mm tube at a velocity of 35 cm/s (NB not 35 cm/min) and I predict that you will get 1.7 ml/min (0.16 squared x pi x 350x60)/1000, so your GC giving you 0.9 ml/min suggests a problem with the EPC, or with the software, or that it is calculating on the basis of their being a vacuum at the end of the column, which is fine but you then need a higher linear velocity, say 60-70 cm/s, which will bring the volume flow into the head of the column back up to about 1.6-1.8 ml/min.

To make sure that you do not have a hardware or software problem you need to check actual flow rates, using a bubble flow meter on the split and septum purge outlets, and by injecting something unretained (any permanent gas) and measuring how long the peak takes to come out. It's good that the machine is still under warranty.

Both "modern" and "old fashioned" carrier gas flow control hardware control flow through the column by controlling the pressure in the inlet, although the way that they control the pressure is slightly different. In the (good) old days the pressure regulator was upstream of the inlet, and there was a needle valve downstream that controlled the split flow. In modern units the pressure regulator is downstream of the inlet (and controls pressure upstream of itself) and the flow controller is upstream of the inlet, and you set total flow which is column flow + split + septum purge. As long as the inlet pressure stays the same, changing the total flow does not change the column flow or the septum flow but it does change the split flow (and therefore the split ratio). All of this can be achieved with mechanical gas controls, but (nearly ?) all GCs now have electronic gas controls, which allow flows and/or pressures to be programmed during a run, which can be helpful under some circumstances.

Peter

[JI2002]

First of all, you need to make sure the column dimension was entered correctly because 25 um film thickness for a 0.25 mm column and 0.1 um film thickness for 0.32 mm column don't sound right.

Second, if you calculate the inlet pressure by using Agilent GC flow calculator pointed out by CPG, you will find the column head pressure will be negative 1.5 psig for a 30m X 320 um column at 1 ml/min. Basically EPC doesn't have the ability to control the pressure below o psig, That's why GC keeps shut down on you. You have to run at a higher flow rate for the column.

[gcguy]

It really sounds like the column dimensions have not been entered correctly into Chemstation. I would have no hesitation in running a column with the dimesions stated at a 0.9 ml/min flowrate if required.

GCguy

[Don_Hilton]

I set the velocity to 35cm/min and the flow becomes 0.9ml/min which can't be accepted by instrument.

So I increase the flow to 2ml/min and the pressure now is 6 psi. I run a sample and I'll see the result.

Check the inlet pressure for the flow of 0.9 mL/min. I suspect that you are either below atmospheric pressure in the inlet or so close that the instrument can not control the pressure. (Remember that the GC controls pressure and calculates flow based on the column geometry that you enter into the column configuration.) Also check the configuration for the other column. I've seen a configuration set for the second column in a GC and specified as being installed as in the same inlet as the column actually present. Check it on the GC front pannel. (I ran into this with a third party data system.)

[thohry]

With the column geometry (30x0.32), the flowrate of 1.0ml/min is normally accepted, except for the case of MSD at the outlet.
You should check the column geometry, the column outlet (it should be Ambient), gas type ...