Leak problem with 6890

[emsman]

Good Morning,

Can anybody help me with troubleshooting on a 6890 GC? I did a maintainance last week and don't get it running again since then. I changed liner, o-ring, gold seal, column ferrule and split vent line. Problem was then, that I didn't reach a pressure higher then 90 kPa. I checked all off the changed parts again. It got better so far, that I reach a higher pressure, but only as long as gas saver mode is on. I can start the sequence and instrument gets the sample, but just a moment before injection the pressure breaks down. This because the gas saver mode is turned off for injection. If I make a splitless method, then there is no problem. Can there be a problem with the EPC-module? If there is a leak, splitless mode should also fail, doesn't it?
Please help me! Thank You!

Best regards
Florian

[Peter Apps]

Hi Florian

Please tell us the settings for flow and pressure etc that are entered on the GC, also the dimensions of the column.

You also need to leak check (with a leak seeker) all the connections that you "maintained".

Peter

[emsman]

Hello Peter,

Column: DB-17MS, 30m x 250um x 0.25um
Pressure: 109 KPa
Split: 20:1
Flow: 0.9 ml/Min
Total Flow: 23ml/Min
Split Flow 18.7ml/Min
Oven starting at 150°C

In the beginning there was a mistake in the method (Split 5:1 and lower velocity resulting in a total flow of only 8 ml/Min) and I thought that this might be the problem, because with higher total flow the instrument could establish the pressure. However, with the method above there was a pressure shutdown after the first run was completed.

So the problem is still existing, don't know if there are one or more reasons.....
I will also check the connections with a leak seeker in a minute.

Thanks for Your help
Florian

[gcguy]

Have you done the proper Agilent leak test?

Put a blanking nut in place of the column. Set the injector to splitless. Blank off the septum purge. Set a pressure of 50psi. Turn off the gas and monitor the pressure drop. This will tell you if you have a plumbing leak.

GCguy

[emsman]

Hello,

The Leak Test according Agilents User Manual was successful, i.e. there was no pressure drop.
However, the problem is still remaining. Sometimes one or two injections are completed, but then there is a pressure shutdown again. With constant pressure mode it is a bit more stable than with constant flow.

Is it possible that the split vent line is plugged and therefore it works randomly? That's the next step I try, change it.
How can I check if the EPC module is OK?

Thanks in advance for further help
Florian

[AICMM]

Emsman,

The reason it does not fail with splitless but does with split is because with splitless on the 6890 the split vent valve is off completely (unlike the 5890 where flow is simply diverted.) Therefore, there is usually ample primary pressure to drive only the carrier flow.

If I understand you correctly, if you make a split injection, it doesn't work, if you make a splitless injection, it will inject but stop at the end of the run with a pressure error? Sounds to me like a primary pressure problem, that is, not enough total gas supply to the injector to provide sufficient back pressure when the split is on. Considering the maintenance you did, I would strongly suspect the culprit to be the split vent line.

Best regards.

[Peter Apps]

Hi Florian

Almost certainly there is a leak somewhere that is bleeding off about 20 ml/min of carrier gas. If the pressure drop test passes with the septum purge and the column connection plugged the leak might be through the septum purge controller, of at the column connection. What was the outcome of your leak seeking ?

What carrier gas are you using ? - a pressure of 109 KPa is on the high side.

You also need to check actual flows out of the split and septum purge vents using a bubble flow meter, not the readout on the GC or the software.

Peter

[emsman]

Hello Peter,

Carrier gas is helium. The pressure of 109 kPa is at 150°C, it's getting higher during the run, as I use constant flow mode. I haven't found a leak until now.
What seems strange to me, is that I get the leak a while after starting the run as the temperature increases. Can there be a leak which gets more at higher temperature? Should be the column connections in the oven, but that is all checked.
The other thing is: when i use a higher split ratio the pressure keeps longer stable. I.e. with 5:1 the breakdown is at a certain pressure (and temperature) and with 20:1 the breakdown is at a higher pressure or there is even no breakdown.

Stupid instrument! Luckily I have some more of them, which are running OK!

Florian

[AICMM]

Emsman,

What is your cylinder pressure at and how many GC's are you trying to feed from that cylinder?

Sounds to me like you are starving the inlet for total flow in, especially since it passed the leak test.

Best regards.

[emsman]

There are two GCs running on a cylinder with 5.5 bar. Shouldn't this be enough?

Emsman

[Peter Apps]

Hi Florian

All the symptoms with different split ratios and temepratures point to a leak, unless there is a problem with an EFC. There is a mass flow controller upstream of the inlet that lets through the calculated gas flow to supply column, septum purge and split. If there is a leak as well this flow is too little to supply all the gas, and the pressure in the inlet drops.

Have you checked the flows out of the split and septum purge with a bubble flow meter ?

What kind of leak seeker are you using ?

Peter

[emsman]

Hi all,

Thanks for Your help.
The instrument is working again. There still was a leak at the inlet, which leaked only above certain pressure and flow. Somehow it was gone after I completely rechecked all parts once again.

Greetinks
Emsman