Chromatography Forum

I have a Tekmar 7000/7050 headspace autosampler connected to an Agilent 6890 GC. I am encountering a problem where the GC inlet pressure suddenly drops as the Tekmar is returning the sample vial to the carousel after it has injected. The valve actuator motor in the Tekmar is very loud and it is clear that the pressure drop only occurs when the valve actuator is turning. This seems very strange to me that a motor inside the Tekmar is causing a pressure drop in the GC inlet. Could this be a power issue (insufficient power to the GC?) or a valve issue? The time the valve actuator is on is not consistent and sometimes the valve actuator is engaged for over a minute.
As soon as the sample is returned to the carousel, the pressure returns to the setpoint (setpoint 16PSI drops to ~6PSI and returns to 16PSI after a minute or so). This is hurting reproducibility as the retention time shift due to the pressure drop can be anywhere up to 0.5 minutes. I'll list my running conditions below:

• Tekmar 7000:
  Platen temp: 140 C
  Valve oven: 160 C
  Transfer line : 160 C
  Platen Equil: 0 min
  Sample Equil: 30 min
  Vial Pressure: 20 psi
  Pressurize time: 2 min
  Pressurize Equil: 0.25 min
  Loop Fill: 1 min
  Loop Equil: 0.25 min
  Inject Time: 1 min

• 6890 GC:
  Inlet Temp: 250 C
  Inlet Liner: Sky 1.0mmID straight
  Split Ratio: 10
  Linear Velocity: 33 cm/s
  Oven Ramp: 35 C, hold 1.5 min, 12.5 C/min to 300 C, hold 3 min
  Detector: FID
  Detector Temp: 320 C
  Detector Makeup: 45 ml/min Helium
  Detector hydrogen: 40 ml/min
  Detector air: 450 ml/min

Edit:Valve actuator motor, not platen motor or vial load/unload motor.

I take it you have cut the gas supply line suppling gas to the inlet. You then connected the 7000 in series? Pipe going into inlet is connected to the transfer line, and the other end going to the 7000?
If configured this way it should work. On first glance I think your injection time is too long . You only need 3-5 sample loop volumes to assure consistant results. At 33cm/sec flow rate will not take long.

I take it you have cut the gas supply line suppling gas to the inlet. You then connected the 7000 in series? Pipe going into inlet is connected to the transfer line, and the other end going to the 7000?
If configured this way it should work. On first glance I think your injection time is too long . You only need 3-5 sample loop volumes to assure consistant results. At 33cm/sec flow rate will not take long.

Yes I have the transfer line connected just as you described. Earlier, I thought it was the platen or vial load/unload motor that was causing this, but I realized the inlet pressure is dropping only when the valve actuator is engaged.

I will try a shorter injection time and see if it fixes anything.

Teledyne Tekmar tech support suggested removing the right side panel of the Tekmar and cleaning the optical sensors. I used some air duster and a cotton swab with methanol to remove the dust which had accumulated on the optical sensors. Hopefully this fixes the issue, otherwise it could be the actuator control board which is about $1000 to replace.

I will update this thread if the problem persists. Otherwise, assume cleaning the sensors fixed the issue.

-cwp

The valve might be sticking - taking longer to switch between pots and so creating an interruption in gas flow to the GC.

Peter

I was thinking of what Peter said. Maybe the valve is sticking. You can remove the rotor by cooling the valve oven and un screwing the knurled "ring" on top. If you have a small magnet tie rotor will pop right out Don't pry it out with a tool. Do not mess with anything that would change the spot of paint on the part. Be aware that there is a mark on the under side of the rotor , make sure it's pointing the same way when you install it.

I can't remember if I connected the gas supply line from my GC directly to the 6 port valve.

Update: Cleaning the optical sensors on the control board for the valve actuator solved the problem. I used a cotton swab soaked in methanol and a couple blasts of electronics dust-off.

The issue was that one of the optical sensors was dirty and was not recognizing the position of the valve, so it kept turning the valve until it reached a time-out error. I'm very glad the solution was this easy and I didn't have to spend over $1000 on a new control board.

Thanks, everyone, for your suggestions

Thanks for the feedback. Kudos to Tekmar support for suggesting it.

Peter