HELP Vacuum Problems

[sarettars]

Dear all,

I'm submitting you my problem, hoping that someone could help me. In my laboratory we have an Agilent 6890 GC system coupled with a 5973N MS detector. The problem is that the air & water check gives us higher values (expecially of nitrogen, > 20%), which change also during the day, thus leading to irreproducible analysis.
In detail, once we turn on the instrument, doing the pump down procedure, the N2 values goes down but, once reached a minimum, it strarts to grow again. At first we thought that the problem was the rotary pump, so we change it. But the issue still happens. We also see that the pressure showed by the gauge controller is from 8*10-6 to 2*10-5. We have also performed every kind of "cleaning" procedures (changed the filaments, cleaned the ion source, clean the inlet and injection parts, clean every kind of part inside the oven ). Now we have noticed that the final part of the transferline (GC/MS interface) is partially curved. The technician of Agilent said that that situation is not normal but that it could not effects the performances of MS detector.

Any of you have already had a similar issue? Do you have any suggestion?
Is it possible to personally replace the transferline?

Thankyou in advance. We are becoming mad!!!

Sara

[tkubowicz]

Hello

I'd recommend:

-cap MSD transfer line with blank nut and try to run autotune (this will cut GC from diagnostic). If tune is OK it means you have problem with GC.

-check carrier gas trap (perhaps is saturated and needs to be changed)

Regards

Tomasz Kubowicz

[Peter Apps]

Hi Sara, welcome to the forum.

I agree with the previous post - isolate the MS by disconnecting the column and capping the connection using a no-hole ferrule if you have one, or a ferrule with a piece of wire in place of the column.

Remember the troubleshooting rules; break the problem down into smaller pieces, change only thing at a time, and do the cheap easy things before you do the expensive difficult ones.

Peter

[sarettars]

Thank you! Now we will try to cap MSD and we hope it will be successful!

Sara

Hello

I'd recommend:

-cap MSD transfer line with blank nut and try to run autotune (this will cut GC from diagnostic). If tune is OK it means you have problem with GC.

-check carrier gas trap (perhaps is saturated and needs to be changed)

Regards

Tomasz Kubowicz

[sarettars]

So.....
Moving from your suggestions, we capped the transfer line with the no-hole nut and we performed an autotune and air & water check. We observed that the signal of H2O and CO2 were good (but changes of about 10% while time was passing) and the signals of N2 and O2 were still high and in the ratio 4:1.
Therefore we took an Argon tank and we tested the m/z 40 signal at first on the nut, but we didn't find any leak.
Then, probably just for a stroke of luck I confess , we flushed the Ar on the external door of the quadrupole and we saw a great increase of m/z 40 signal. Now we will try to check the internal seal.
Do you have any other suggestion?

Many, many thanks for your help!

Sara

[Rndirk]

That's good news, you have found your leak!

You might have to replace the big O-ring (check if it's visually damaged) on the inside of the analyzer chamber. It's a pain to replace, have fun

Obviously don't use grease or something to seal that chamber...

[Bigbear]

I beg to differ, apezieon (sp) grease can be used on that O ring. If used use a very very small amount, then use a kim wipe to remove as much as you can.

Works for me.

Something I always do when buttoning up the MS is to wet a kim wipe with methanol and wipe the O ring while it's in place.

[a.pennetta]

If the O-ring is not damaged, you can try to remove it and to pass it through your fingers. Also do not turn the screw, but push the quadrupole cover while you turn on the mass. I noticed that when the screw is tightened the quadrupole cover does not adhere perfectly.

[James_Ball]

I beg to differ, apezieon (sp) grease can be used on that O ring. If used use a very very small amount, then use a kim wipe to remove as much as you can.

Works for me.

Something I always do when buttoning up the MS is to wet a kemwipe with methanol and wipe the O ring while it's in place.

I always use Apezieon L grease on the o-ring. Put a little on your fingers and pull the o-ring through until it is lightly coated, then pull it through a kemwipe several times to clean it off. Also if you do not remove the o-ring, at least wipe it with a kemwipe or other lint free wipe, even a small piece of lint on the o-ring will give you a leak. You can also blow it off with some Dustoff or argon to remove lint and dust.

As another said, don't use the screws to hold the door in place, they cause more problems than they solve. Only use them to seal it up when shipping or moving it. And never tighten up the one on the rear, because some day you will try to open the door and forget about it and end up bending the door and it will never seal again.

[sarettars]

Thank you everybody!
We carefully looked at the big O-ring in the MSD door and it looks like it is not damaged. After, we run our finger on the O-ring surface and we let the system equilibrate for two days (but we held the door using the screws). On Monday the autotune and air and water check were almost OK (N2= 10%), today we observed again irreproducibility in the chromatograms and, performing these tests again, the N2 was 15% and the ab. ratio between 69, 219, 502 was 100:30:1. Therefore we supposed there was still a leak in the same place, confirmed by the argon test.
We called the assistence service and for replacing the O-ring (88 €) they asked us about 1400 €. For this reason, we want to replace it by ourselves. Is there anyone who has a manual for replacing it?
In this way we can also test the solution previously proposed by all of you (grease or not grease).

Thank you again for everything! =)

Sara

[Bigbear]

I don't think there is a manual for changing the O ring. It is not difficult, but it can be trying as you will be attempting to put a round o ring in a sort of rectangular opening.
Take your time and it can be done.
As an earlier poster said, when pumping down push the door closed by applying pressure on the aluminum box with the 4 rubber feet. If you feel the need to use the front screw, only screw it in so it's a mm or so away from it's seat, and ONLY do this after the pump down is complete.

[James_Ball]

To replace the o-ring you need to remove the side plate(door) and set it aside, on the rubber feet.

Open up a paper clip and if you look in one of the corners there is a notch in the outside of the groove that the o-ring is in, use the paperclip to gently lift the o-ring up enough so you can begin to pull it out using your fingers. Be careful not to damage the o-ring with the paper clip.

When installing the new one, start it in one corner and work around from both directions at the same time. Gently push it into the groove with your fingers working around the groove slowly. You will probably end up with a excess of o-ring where both hands meet, but you just keep working until that falls down into the groove, it will compress back along the length of the o-ring.

If you try to work around the groove in a single direction, you will always have the excess popping out ahead of you as you go around and will just keep chasing it around and around.

Once the o-ring is in, slide your finger around it to make sure it is completely seated in the groove and that there is no dust on it. Reinstall the side plate, and pump down.