Dear Suzanne,

I doubt that it is the Valco. But of course it can be an issue. However, since you replaced it this seems not the reason. MY first suspects would be

- contaminated Helium (means your Cylinder is bad/contaminated or something between your cylinder and your Gasbench).
- check Helium flow of both flush measurement needle (and at the outlet of the valco). If your flush flow is too low you will have traces of air in your Exetainer that might look like that (set flush flow to 120 ml/min).
- the “carry over peak” might indicate that you flow out of the measurement needle is too low. That means that you still have the previous sample or air in the transfer line. You can either try to increase the system He flow a bit (however, this also lowers your signal in the ion source as more helium enters the source). Or you change the transfer time (allow 20 sec more).
- check positions of the capillaries in the sample open split. Maybe it sits too high in the split? It must go all the way down to the MS transfer capillary in closed position.
- Check the regulator and the helium supply line of the open split. Blow some argon gently(!) above the open split. The helium must prevent that the argon enters the open split. If you see a clear argon response the helium supply capillary might be blocked or leaky at the regulators ferrules.

regards,
Robert


> Am 04.07.2018 um 11:28 schrieb ankerssm <[log in to unmask]>:
> 
> Dear colleagues,
> 
> I am chasing a leak in the Gas Bench and need some advice. The problem began with elevated backgrounds. We thought there was a leak in the Valco so I replaced it with a refurbished valco. After making sure all the port connections were tight, my backgrounds were good again. However, when I ran helium blanks, I got very small and regular peaks in the mass trace and larger peaks in the ratio trace (see attached image).
> 
> I inserted the sample needle in a helium flushed vial to test the valco. When I scan mass 40 and switch between load and inject, I get a large peak (~8,000 mV). If I leave the GasBench in inject mode, the peak quickly falls off to zero. I should not be seeing any peaks when I run a blank or switch between load/inject, right? I was certain that the problem was at the valco so I have replaced it (3 times!) without any improvement. While chasing this leak, I have almost rebuilt the Gas Bench.
> 
> Here are all the other things I have tried so far:
> 
> ·        replaced the sample needle (twice)
> ·        replaced the 1st water trap
> ·        replaced the GC column
> ·        swapped and cleaned the sample loop
> ·        switched helium tanks
> ·        leak checked all points upstream and downstream from the valco
> 
> I am stumped. When the device is in load mode, everything is tight. When I go to inject mode, I get a large pulse of atmosphere/background. Also, there is a carryover peak that shows up on some samples during the CO2 reference peaks at the beginning of each sample. At first, I thought this was due to opening and troubleshooting the device between runs. But I see it in different samples throughout a sequence so maybe it is part of the overall problem. Has anyone had a similar experience? Any advice for where I should look next?
> 
> Thanks,
> 
> Suzanne
> 
> 
> 
> -- 
> Suzanne Ankerstjerne
> Instructor/SIPERG Lab Manager
> Science Hall Building Supervisor
> Iowa State University
> 2237 Osborn Drive
> Geological Science
> 
> Office: Rm 55, Science Hall
> Email: [log in to unmask]
> Phone: (515) 294-6514
> https://siperg.las.iastate.edu/stable-isotope-lab-sil/
> 
> 
> 
> <image.png>

-- 
PD Dr. Robert van Geldern
Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) | GeoZentrum Nordbayern
Schlossgarten 5 | 91054 Erlangen | Germany 
[log in to unmask] | www.gzn.uni-erlangen.de