Troubleshooting - Your thoughts welcome!
Posted: Sun Mar 12, 2023 8:57 pm
Hi,
I am heavily reliant on a GC-MS for my research work and for that reason, I tend to try to be reasonably conservative/careful in order to minimise downtime which (as a heavy user) hits me hard. I normally develop basic methods on FID and then move to GC-MS, keeping my sample concentrations for splitless at or below 50 PPM where I can. A colleague who is supposedly 'very experienced' at GC recently decided to run a bunch of quite viscous samples through our GC-MS at 'PPM concentrations' with no solvent delay after assuring our technicians that it would be fine.
As it turned out the dilutions were way off (they turned out to be at parts per hundred level!) and this led to quite a bit of downtime due to the filaments burning out, the technicians having to spend a lot of time cleaning the source etc. and various other issues.
It's now back up and sort of running, but sensitivity is way, way down. After running an inlet leak detection/restriction test, it would appear that there is a whopping 50ml/min leak in the injector.
My interpretation of this result (which is repeatable) is that the split valve is not closing properly, possibly down to being clagged up with the viscous sample referred to earlier in this post, leading to my splitless runs not being splitless at all. The other thing that I think might support this idea, is that the liner seems remarkably easy to remove and put back in - it normally needs a bit time to work its way down and requires a little bit of force, presumably as any gas in the inlet which is being displaced would normally have to works its way around and through the new liner.
Does this interpretation sound right, or is there perhaps something else that I haven't considered?
Many thanks for any thoughts on this.
TD2
I am heavily reliant on a GC-MS for my research work and for that reason, I tend to try to be reasonably conservative/careful in order to minimise downtime which (as a heavy user) hits me hard. I normally develop basic methods on FID and then move to GC-MS, keeping my sample concentrations for splitless at or below 50 PPM where I can. A colleague who is supposedly 'very experienced' at GC recently decided to run a bunch of quite viscous samples through our GC-MS at 'PPM concentrations' with no solvent delay after assuring our technicians that it would be fine.
As it turned out the dilutions were way off (they turned out to be at parts per hundred level!) and this led to quite a bit of downtime due to the filaments burning out, the technicians having to spend a lot of time cleaning the source etc. and various other issues.
It's now back up and sort of running, but sensitivity is way, way down. After running an inlet leak detection/restriction test, it would appear that there is a whopping 50ml/min leak in the injector.
My interpretation of this result (which is repeatable) is that the split valve is not closing properly, possibly down to being clagged up with the viscous sample referred to earlier in this post, leading to my splitless runs not being splitless at all. The other thing that I think might support this idea, is that the liner seems remarkably easy to remove and put back in - it normally needs a bit time to work its way down and requires a little bit of force, presumably as any gas in the inlet which is being displaced would normally have to works its way around and through the new liner.
Does this interpretation sound right, or is there perhaps something else that I haven't considered?
Many thanks for any thoughts on this.
TD2