5973 High N2/H20

[benhutcherson]

I've posted a decent bit here about an ongoing 5973 project, which is now doing pretty much full time service running EPA 522(1,4-Dioxane in water).

For all intents and purposes it's running great and consistently passes BFB tune with rarely needing more intervention than just running "BFB Target Tune." The only issue I sometimes have is BFB target getting 176 to 104% or so(relative to 174), and even if initial tune passes it will go there after ~48 hours, but that's yet to be a problem that's unsolvable. I'll mention too that I'm currently running helium. Column is a CP Select 624 .250mmx30m, and I'm running at 1.2mL/min. Hi-vac is reliably around 1.3x10^-5 torr(gauge) at analytical flow rates.

For as long as I've had this instrument in service, even when all my other numbers looked good, N2/H20 has always been high. Right now it's running ~200%. I've done everything I can think to rule out a leak at the MS itself including a fresh TINY amount of Alpezion-L on the door and vent valve O-rings(and I've also replaced the vent valve O-ring). I've used dust-off to check every potential leak source at the MS, and I can't seem to find it. BTW, I should mention that this problem has carried across a change from a diff pump to a turbo pump(and accompanying vac manifold change).

I'm chasing the last little bit of sensitivity, and am afraid that my high N2 levels are holding me back.

I'm paying a bit of attention to the inlet now, and am wondering if I should start chasing issues there. I did try today snugging up the column nut. I was watching absolute counts of N2 on manual tune/scan and did see it drop after doing that, but tried several other things including playing with the tightness of the septum nut(new septum, never pierced for this) and didn't see a ton of difference. The gold seal, liner, and liner O-ring are all relatively new.

One thing I've never really paid attention to is the split line, which I know is sort of a consumable. I know Agilent sells these pre-cut, but is there any reason why I can't make my own? I have plenty of pre-washed 1/8" copper tubing on-hand, and it looks like nothing but a cut piece of tubing with Swagelok fittings on the end.

Aside from this, is there anywhere else I should be looking for a potential leak?

[Consumer Products Guy]

One thing I've never really paid attention to is the split line, which I know is sort of a consumable. I know Agilent sells these pre-cut, but is there any reason why I can't make my own? I have plenty of pre-washed 1/8" copper tubing on-hand, and it looks like nothing but a cut piece of tubing with Swagelok fittings on the end.

I replaced a somewhat clogged split line on a 5890 GC-FID unit with 1/8 copper tubing, and had no issues.

[Consumer Products Guy]

Did you try a new helium cylinder?

[benhutcherson]

Did you try a new helium cylinder?

No, I will need one before too long but, no joke, this one has been here 15 years. Buying anything here is complicated, and unfortunately stockpiling(like I did at a previous job-I was on the phone ordering if I didn't have 2 in my reserve stash, especially since the way we had them plumbed a careless TA could kill one overnight...).

I can say that when I was running this on hydrogen, I had the same issue, and it was consistent across the couple of H2 cylinders I went through in a few years.

[Consumer Products Guy]

Understand. I'd plug the transfer line end in the oven compartment and check on the N2 and water, to try to isolate if such are entering from the MSD itself or from the GC part.

[benhutcherson]

You know, I'm not sure if I paid attention the last time I had a blank ferrule on it. It was while I was doing some troubleshooting/diagnostics and the only tunes I have from then were when I temporarily had a GPIB eMod in the instrument and was running it on G1701BA, which IIRC doesn't do air/water as part of autotune.

I need to do inlet maintenance this week. so I'll toss the blank ferrule in(I keep the nut/blank inside the oven door for a reason) and see what happens.

I guess as an intermediate, though, without actually venting, I could plug the inlet end of the column with a septum while I'm working on the inlet. I'd normally do that anyway if I were going to have the inlet open for an extended time without venting the MS. I know that doesn't rule out a leak at the column nut on the transfer line, but it should isolate the inlet.

[Consumer Products Guy]

I guess as an intermediate, though, without actually venting, I could plug the inlet end of the column with a septum while I'm working on the inlet. I'd normally do that anyway if I were going to have the inlet open for an extended time without venting the MS. I know that doesn't rule out a leak at the column nut on the transfer line, but it should isolate the inlet.

Yes, that could help to isolate the inlet as the cause.

[benhutcherson]

I wanted to follow up a bit on this.

I spent some time breaking the inlet down as much as I could think to do. I pulled the S/SL block out and swabbed the whole inside down with DCM, something I found suggested in the service manual. I polished the base, where the gold seal would sit, with some very fine mesh Micromesh(the same that I've used to clean the source). The nut that retains the gold seal got a good cleaning-it might need a replacement as I was not totally happy with it, but I got it as good as could reasonably be.

I broke the septum weldment down as much as I possibly could. I honestly think this may be an issue that I need to address eventually, but not a chance of the money now or any time soon(unless I can do some horse trading and get a serviceable one off a parts system). In a past life, this system ran a P&T, and the line was spliced very close to the weldment itself. The union I have in it makes routing the line very, very difficult and it's hard to do it in such a way that you're not putting upward and uneven pressure on it. This did not have an ALS when I set it up(well, I guess the a lot of P&T systems probably don't) and I'd notched the original manual injector cover to handle the nut. I've not modifed the ALS mount for this, and had an "a ha" moment when I realized that the lines really are supposed to come from the opposite side as compared to the manual injector(left vs. right). For now, I just bent the lines such that they come down 90º, which seems to work well, although the stress on them can't be good either and of course I'd be up a creek if I broke one.

I understand that SIS is good at jobs like rebuilding these weldments with new tubing, and that might be a reasonable alternative to $600+ for a new one(or hoping I could salvage one off another 6890 that wouldn't have its own issues) although can't afford the down time right now to do it.

In any case, I also pulled the C-clip and took the weldment apart as much as I could for cleaning. That let me pick out some septum crud remnants that had been there a while-some of which I knew about, some of which I didn't. I sonicated all the parts in DCM, then hit the lines with it and blew them dry.

I actually ended up not replacing the split vent line. I had everything out and ready to do it, but couldn't really justify a reason to do so-no solvent I ran through it came out the other end as anything other than clean solvent, so I just hooked it up to the compressed air cylinder and let it backflush a bit and reinstall. I went ahead and cut one so I cna change it if I do find that i need to address that.

Long and short of all of this is that later that day, my N2/H20 was down to ~60%, which is 1/4 of what it was before I did all of this. We've been closed for spring break this week and I haven't been in since doing this, but I gave the inlet nut an additional snug before leaving for the day(I figured that would have given the ferrule a chance to shrink-it moved a few degrees) and let it be. Normally my resting flow is .5mL/min through the column, but manually set the pressure to 10psi(which worked out to .9mL/min with the oven at 31º) to avoid having the pressure too low in it. I did cut back septum purge a bit(to 1.5mL/min), as I'm below 1K PSI on helium and ordering can be...well an adventure here and certainly not a fast process.

So hopefully I can report more tomorrow on performance.

[LALman]

I chased a similar leak for some time. I bought the RESTEK splitvent block to see if I could isolate leaks to either the EPC or the split vent. I bought a new split vent filter with new seals. In the end it turned out to be my Tekmar P&T unit. So, try this. Blow some argon at your purge and trap switch valve and check for the Argon peak increasing. Or you could switch to Argon as "purge gas" for a short time (I ran with argon purge for years) Those valves have super tight tolerances but it could be sucking some nitrogen especially if you are using nitrogen as your purge gas. I grabbed a valve out of another P&T unit and swapped it in and problem solved. Got that valve resurfaced for a very reasonable price.