have you checked your flow in both neddles?
The flush needle should have about 100 to 120mL. If flow is too high (say 250ml) this will cause an overpressure in the vial that has to be compensated in the first minutes of the vial analysis. This is usually done already during the transfer time of the sample. From my experience an high overpressure is not what you want but should not cause the symptoms you described.
The analysis needle should have a flow of about 0.5 mL measured at the vent of the valco. Did you adjust this after you changed the nafion tubes, needles etc.? This may change due to different lengths of capillaries and slight variations in the diameter. If the helium flow through the measurement needle is too high this will dilute your sample in the vial fast and will consequently decrease your signal faster. From your description of the problem I suspect that the fast drop of the signal should be related to the helium flow. This should a cause a continuos drop over the 8 or 10 sample peaks.
If you have peaks missing or a sudden drop of one peak this may be cause by water drops under the septa that block the needle. However, this usually happens during the first peaks of the analysis.
You may also check your standard system parameter as system and signal stability and peak shape just to make sure that the system itself works fine and the problem is related to the Gasbench and not the MS.
Am 22.03.2013 um 19:43 schrieb Dana Biasatti:
> Hi all,
> I am running water samples on a gas bench and Delta V plus. On the last couple of runs, there has been a 4V drop in the signal from the 1st peak to the last (10th) rather than the normal 1 to 2V difference between the 1st and last peaks. This is causing bad precision. I baked out the gc column in the gas bench for several days, I ran about 50 standard on-offs yesterday, I've changed the sampling needle, and it's only been a couple of months since I replaced both water traps. The source heater, analyzer heater, and heating block are turned on but do not help either, so I'm thinking that water in the system is not the problem. Does anyone have any suggestions?
> Thank you very much.
> Dana Biasatti
> Assistant Research Scientist
> Chesapeake Biological Laboratory
> University of Maryland Center for Environmental Science
> 1 Williams Street, PO Box 38
> Solomons, MD 20688
> Telephone +1 410 326 7452
Dr. Robert van Geldern
GeoZentrum Nordbayern / Applied Geosciences
University of Erlangen-Nuremberg
91054 Erlangen, Germany
[log in to unmask]
room: O 2.112 (office)