Direct connection of a Saturn 2200 MS

[johnelling]

Folks, I want to use a Saturn 2200 off of an old 3800/2200 GCMS to directly sample a gas reaction chamber. Does anyone know what pressure the Saturn is trying to achieve in the MS inlet? I want the gas reaction chamber to have a pressure above the inlet pressure, of course, so that the sample draws into the MS.

Does anyone know if there is a reason the 2200 inlet requires a pressure above 1 atmosphere

If this will work in theory, my next problem will be getting the Star software to work with only a MS connected.

Thanks for any advice!

[benhutcherson]

It's been a while since I had an operational Varian trap(been threatening to get my Saturn 2000 going again for a while), but the transfer line(where the column attaches) is going to be roughly equal to the operating pressure in the trap. IIRC, that's on the order of 10^-4-10^-5 torr. A trap the size of the Saturn doesn't need as hard of a vacuum as a quad, but in practice it usually gets there. The vacuum system is designed to hold this pressure at a flow rate of ~1mL/min helium.

Operating without a GC itself isn't really a problem-I had one hooked up for a while without realizing that I needed to run a separate ethernet cable to the attached 3900(stupid on my part). It never WORKED properly(more on that in a second) but I could operate all the functions of Saturn-such as turning filaments on/off, checking vac pressures, and running autotunes or air/water checks from MS Workstation. Even though I don't have a Saturn running now, I have another Varian GC-MS, and I have full access to all the MS functions(which are a lot more complex/involved than a Saturn in this particular instrument) even with the 3800 powered off.

One important point, though-the Varian Saturns-and I think all traps in general-MUST have a small amount of helium present as a "buffer gas" to work properly. Only helium works-not hydrogen, nitrogen, argon, or some other inert gas. On a GC-MS, this is readily supplied by the normal column flow. I have a Finnigan LCQ trap(LC-MS) that plumbs helium in at a low flow rate for that very reason. The last time I had my Saturn going, I drove myself crazy for a few weeks with it not giving any useable signal because I didn't have the GC connected and consequently no He flow.

To your specific situation-you MIGHT achieve it with a length of .25mm deactivated capillary tubing attached to a gas sampling valve. You'll need a pressure regulator to supply a small flow of helium when not actively sampling. This set-up I'm describing is starting to sound a lot like a GC...and if you can find one Varian did support these sort of attachments on the GC inlets pretty nicely.

[johnelling]

Thanks Ben. I'm trying to get the system communication with the data station going now. I have to try to build a Star workstation by loading the software onto an old XP computer with a 10BaseT network card and a GPIB card. I have the GPIB card hooked up to the Saturn 2200 MS now and the Star software does not see it. I don't have a connection to the 3800 - still waiting on an old network card from ebay. You may be indicating that the GC has to be installed first so I'll try that before I panic.

Do you know if I'm going to be able to get the National Instruments GPIB card to work. I'm somewhat troubled that the NI software does not see anything connected even though the MS is connected and powered up. But maybe Varian didd something special with their GPIB protocol.

Thanks for any more advice!

John

[benhutcherson]

Mine runs off a National Instruments GPIB card-in fact trying to re-purpose its computer was my first introduction to the fact that HP instruments will only work on an HP/Agilent card . I unfortunately don't know exactly what model card it runs on. I do know that it's on a Dell Optiplex GX620, which is a Pentium 4 that I see used a lot to run GPIB/HPIB instruments. I have another of these earmarked for Chemstation G1701BA, and the person selling me the software specifically said that only an Agilent 82350B card(not the HP 82350A) will work correctly with that particular computer. There COULD be a similar situation with the NI cards, but I honestly don't know.

I do know that MS Workstation is perfectly content to talk to/control only the mass spec. In fact, on a prior attempt to get mine working, I had overlooked even connecting the GC via ethernet, and was able to still operate the MS just fine. Of course I was never able to make sense of anything going on-partially because(much to my ignorance at the time) I didn't realize how important having helium in the trap was. None the less, I was able to control all the MS functions including executing a pump down, setting temperatures, reading pressures, turning the filament on and off, and seeing the pitiful attempts at collecting a mass spectrum.

Fortunately too, mine is paired with a 3900 GC, which I like somewhat better than the 3800 even though it's a bare bones GC without a front panel. It also connects via 10baseT, not 10base2.

I'll mention too that since your post, I have had mine powered on again(it's not currently as the only place I could put it, I was overloading the circuit) and managed to get something resembling a decent tune.

With all of that said, you mentioned "Star Workstation." I've never used Star Workstation, but was always under the impression that it was meant only for controlling GCs(and maybe LCs also) with "simple" detectors like TCDs or FIDs. I was under the impression that all the Varian MSs(including the Saturns, the 1200 and 300 series, and the 500 series LC-Ion Traps) needed MS Workstation. Both my Saturn 2100D and my 300-TQ are on MS Workstation 6.9.3.

[johnelling]

Ben, I've loaded the MS workstation software 6.30 that is supposed to be controlling the GC/MS. The web of connections is a little strange. I've got the 10BaseT coax cable from the computer to the 3800 and a GPIB cable going from the computer to the Saturn 2200. Then there is a DB15 connection between the GC and the MS from the J6 connector on the GC that might be necessary according to some internet info.

So far the workstation software sees the 3800. I went through the System Control / Automation startup and the network connections to get to this point. Unfortunately I can't connect the MS. It is supposed to be a module that is added to the System Control / Automation screen but the option is not present in the software (I'm looking for address 40 in the Setup Ethernet Ports dialog box). I have no idea how to get this module connected so that the 2200 MS shows up in the System Control.

The manual is frustrating. There is a lot of info on how to connect via ethernet to the 3800 but nothing about setting up the connection to the MS! I've love a suggestion as I'm at a dead end.

[benhutcherson]

Not sure if this is any help or not, but this is the card in mine.

Unfortunately, I can't help on initial set-up because, as far as I know, mine was set up by a Varian engineer(we bought it new).

Something that I know from HP/Agilent stuff, though-each device on a GPIB chain has to have a unique ID from 0-30. I've never dug into the interface card on a Saturn to see if it can be changed, but at least on HP stuff there's set of switches that let you set the number. You then have to tell the software what the address is in order for it to find the device. There's a "normal" ID for certain device types but this can be changed. One might want to change it if one is running multiple devices from the same computer.

Also, I'll mention that you're correct that the 3800 set-up is relatively straight forward with a 10base2 ethernet card. Mine came with a hub with both RJ45 and coax connectors, but I never could get that to work. I put the 10base2 card in(I had a couple handy of a very specific 3Com card from a failed attempt to get a Finnigan LC-MS working), set a static IP address on the GC, and the next thing I knew I had the GC control window up on the software. BTW, your set-up is not all-together different from my 3800-300, with HPIB subbed for the DB-9 interface card the 300 uses. The cable connecting the GC to the MS is the sync/start cable to tell the MS that a run has started. It gets even more fun when you add the pigtail connector for the autosampler!