MSD 5975C leak

[sausuma]

We are using Agilent MSD 5975C with GC7890A. The system is not giving proper air and water check. (N2 % > 15). We have done various trouble shooting checks.
1. Plug MS at the interface : Air water check passed (No leak in MS side)
2. Connect the transfer line column to the MS but plug the other end going into the splitter (Dean's switch) : Air water check passed (No leak in transfer line and interface nut)
3. Connect transfer line to the Dean's switch but plug all other outlets of the switch (viz GC column and FID). : Air water check failed. We start seeing N2 peak > 15-20 %
(and it varies with different runs) This made us conclude that there is a leak somewhere in Dean switch. But we could not identify the exact location by using Dust removal spray. So I have 3 options :

1. Go for on site agilent maintenance. (Very expensive and will take time)
2. Buy a new dean's switch. (Expensive and risky too as we are not 100% sure that it will fix the problem)
3. Continue with the small leak (around 15%) (Cheap but I dont know how will it affect our result and other parts of instrument)

Please, need advise.

[chemstation]

Hi Sausuma,
Please provide more information, you mention N2%, but not O2%?, as you may have a leak in
your Helium supply before the OxyTrap.

Have you plugged off the Dean Switch Inlet port, as you may have a leak in your Injection port, eg blown septum. Or configured the Switch to direct the flow away from the column connected to the MS.

Have you connected the column directly to the injector (with a large Split ratio), to also check the helium quality?
As you may have a bad Helium connection going into your Aux EPC module.

Have you turned the AUX EPC pressure off, and only pressurize the Deans Switch via the
Injection port gas Flow (you can check by seeing the pressure reading rising even though the EPC is turned off) and see if N2% changes.


Kind regards
Alex

[Peter Apps]

You need to check for leaks at all the connections with a leak seeker - if you do not have one then buying one will certainly be cheaper than a new Deans switch. You could try puffing dichloromethane vapour around the connections and looking for it on the MS, but it is difficult to localise the leak on a small device like the Deans switch.

Is the Deans switch at the MS end of the column or at the inlet end ?

What is on the other side of the switch - the side that does not go to the MS ? You might have back-diffusion inside the switch if the flows or pressures are not correct.

Peter

[sausuma]

Dean switch is after the GC column. It directs column flow either to FID or to MS. I plugged ports coming from GC column and also the port going to FID. So even if there is a problem of He contamination or leak in GC inlet septa etc., it should not affect MS.
I also tried turning off the auxillary EPC but no change in fact the N2 peak increased further.
N2 to O2 ratio in water air check is about 10-15:1

[Peter Apps]

Comments in blue

Dean switch is after the GC column. It directs column flow either to FID or to MS. I plugged ports coming from GC column and also the port going to FID. So even if there is a problem of He contamination the Deans switch is fed with helium via the auxillary EPC, so plugging the column port will not have made any difference to the helium reaching the MS or leak in GC inlet septa etc. the test has eliminated that, it should not affect MS.
I also tried turning off the auxillary EPC but no change in fact the N2 peak increased further unless you did this with the FID port plugged this is just the MS sucking air down the transfer line from the FID. You are lucky that you did not burn out the filament. If this was done with the FID port plugged then it is very strong evidence that you have a leak in the Deans switch.
N2 to O2 ratio in water air check is about 10-15:1

Try puffing dichloromethane vapour around the switch, and get yourself a leak seeker.

Peter

[chemstation]

The N2:O2 ratio seems atypical, as the ratio is usually 4:1, I don't know if you have tried this,
have you connected the GC column directly into the MS, bypassing the Deans switch ?, as need
to check that your main Helium supply is good, (before it splits off to the Aux EPC at least.)

Swapped the FID and MS outlet positions, as a fracture on one side may not be detected
if the MS is on the other side. (presuming it is leaking at the plate in the oven)

Have you checked the % when the oven is at 50'C and when it is 280'C, you may need to
calculate the change in flow using the Flow Calculator programme, to normalise the result.

As I diagnosed an instance where a fracture in a CI weldment into the transfer line would leak
at 280'C, but not at room temperature from a new installation, which had confused the
Agilent Techs, as they come , clean things, test the system and it would pass, by when
we went to use it after they had gone, there would be a leak.

Alex