MSD Chemstation wont read 7673 autosampler UPDATE:SOLVED!

[benhutcherson]

Email me at hutchersongcservice@icloud.com and I should be able to send images of BA. I'm also happy to send the 5970/71 patch for it, which I realize you don't need but is handy to have.

For some reason, I can't locate the compatibility matrix(I need to just save it because I never seem to find it when I need it) but for some reason I thought CA was the last version that officially supported HPIB. I know DA can be modified to run a 5890/5971 and there's a company offering that service(and tie it to a HASP to do it).

Just out of curiosity, and to eliminate other issues, have you removed the ALS completely(or at least disconnected the G1512A) and attempted to run the 6890/5973 alone? Is it otherwise operational?

Having never used DA, I don't know exactly when things changed, but BA only supports 9 character file names. I know that for me, that can be SUPER limiting for setting up a sequence. I've gotten creative with naming, of course judicious use of folders, and sometimes just having to stick to vial numbers and cross referencing my lab notebook(which I generally will need to do for quantitation anyway). With that said, when, for example, I'm doing a calibration curve it's super easy to be able to just dump concentration info right into the file name for when I'm working it all up.

Also, in the grand scheme of things, it's probably not terrible to look at doing a LAN upgrade on the system if you plan to continue using it. JetDirect cards are easy enough to install in 6890s since AFAIK all have the slot there for them. You can salvage them from old printers, although you need a BootP server to get those to work and Geoff has told me that it's iffy when running EA under Windows 10(the JetDirect card you get from Agilent now doesn't need one). You need a SmartCard III, which you can get from Agilent still, on the 5973. Geoff does a really need job of installing them-I hadn't really looked at the install before a few months ago when my mainboard went out(I'd basically left it alone since he did mine in May 2021) but IIRC the remote port lines up well vs. the SCII++ and he uses the standoffs there to hold the SCIII in place. I'm working a lot on memory, but I think he basically used a sheet metal nipper to open up the original slot for the HPIB port large enough for the LAN cable to plug in. When you look at the cost of the "good" JetDirect card, the cost of a new SCIII, and a G1701EA license, the $10K or so that Geoff charged us a few years ago really was pretty reasonable for his time coming and setting it up.

[boulderco]

Thanks for the detailed reply, email sent. I've been running the 6890 with a single FID for a couple years, and about a year ago bought the MSD for it and got it hooked up no problem. The system has been running completely smooth with chemstation and manual injection, and only since getting the ALS setup has it encountered the problems. When chemstation isnt running, the ALS/MSD/GC setup works great which leads me to believe it is the software. Id love to upgrade to LAN or a whole newsetup altogether but am on a tight budget at the moment sadly. I appreciate all of your help tremendously!

[boulderco]

Still no luck!!!! Thank you to the above members who sent the ChemstationsI have tried both the DA and BA chemstations to no avail. When I have the controller turned off, chemstation loads the GC-MS just fine in both versions. I can even turn on the controller while chemstation is running and the "ALS ON/OFF detected, restart chemstation to connect to sampler" message pops up so I know its being recognized....

But when I try to run chemstation with the controller on, the software says it cannot bring up GC and will remain offline, but it appears it is seeing the ALS. Very strange.

Any and all ideas are welcome at this point, no matter how far fetched I'll give it a shot!

[benhutcherson]

Sorry that I never found my images to send you...

Do you have another G1512A you could try swapping around on this system and see if it works? Do you have a different set of cables? I'm not sure if any DB-9 cable with the correct gender on each end will work, or if there's a specific way the ports are connected. That COULD make a difference.

I can try to check mine(probably won't be this week) and report what the connections are. That would actually be a handy thing to have since generic DB-9 connectors are not that expensive and genuine Agilent cables are $$$.

Even assuming yours are correct, though, it might still be worth checking them with a VOM and making sure you don't have an internal break or lack of continuity that's causing issues. I'd worry less about the sync cable than the comm cable from the 6890 to the G1512A-IME with using a G1512A on a 5890, if the sync cable is missing I think it will make the injection but just not start the run(or will it just stall out and not make the injection? That was 5 years ago when I last tried that).

Just thinking through possibilities.

[boulderco]

No worries at all Ben, cannot thank you enough for your help thus far!
(Right now) it actually appears the software was not the issue after all. I got a copy of older MSD chemstation from KM-USA and had weird issues that I could not find another report of anywhere on the net (after much digging, trust me).

I have a strong suspicion you are spot on about the cable, the G1530-60600 cable (the fancy agilent/HP 9pin serial cable ) was just tested for continuity and gave back some weird results. It looks like the cable may have an internal short between a few pins so I am going to try to pick up another one tomorrow and give it a go. This could possibly explain why the 6890 would read it but chemstation crashes.

I purchased the controller from a friend who bought it refurbed with a warranty but upgraded to LAN before they could use it so I hope its not that, but if all else fails I will definitely purchase a second one and give it a go.

None of the manuals or docs I've read mention upping the baud rate, but the 6890 is set to 9600 (I think), and the controller defaults to 2400. Changing the controller to match the 6890's rate hasn't done anything, but figured I would mention it.

I may also try using a Y APG-remote cable instead of the two I'm using now (one for mass spec, one for ALS controller).

Once again, thank you for the help everyone. This has had me quite stumped!

[boulderco]

Update:

Got a new cable and tried it to no avail. After a closer look at the pin out diagram, I realized I was testing it wrong and all the cables have now been tested for proper continuity. Back to square one

Either has to be the GC or the ALS controller, since an ebay G1512A is a lot cheaper than a 6890, Ill probably try that route first, but any other ideas are quite welcome

[boulderco]

Cannot thank everyone enough for their help, I owe ya one big time.

The problem turned out to be the Firmware on the G1512A.....

As soon as I updated it, sure enough everything works like it should!!!!

Alas, victory!

I am having some pretty bad coring and my septa are starting to leak after only a few injections. Ive tried multiple different Agilent ALS syringes (even the tapered gauge one). Im using Supelco Thermogreen LB2, I usually get a couple hundred injections out of them when doing manual injection using my Hamilton Point Style 5 thin gauge needle. Any better septa recommendations? Is it possible the ALS is misaligned slightly? This is a whole new world for me

[Consumer Products Guy]

I am having some pretty bad coring and my septa are starting to leak after only a few injections.

We experienced same about 4 decades ago with fresh batch of septa from our usual supplier. I called them and insisted something was wrong, and we returned them and found out that there was indeed something wrong with the entire batch, so replacements were sent.

This was so long ago that I will not name the supplier, as that would have no value.

So it sure could be a bad batch of septa !!! I doubt that it is autosampler or syringe-related if you've tried different syringes or needles. And sometimes we did use syringes with side holes.

[Jake]

I am having some pretty bad coring and my septa are starting to leak after only a few injections.

We experienced same about 4 decades ago with fresh batch of septa from our usual supplier. I called them and insisted something was wrong, and we returned them and found out that there was indeed something wrong with the entire batch, so replacements were sent.

This was so long ago that I will not name the supplier, as that would have no value.

So it sure could be a bad batch of septa !!! I doubt that it is autosampler or syringe-related if you've tried different syringes or needles. And sometimes we did use syringes with side holes.

I cannot second this enough. Ever since COVID, we've had nothing but quality and supply issues from our GC/sample prep consumables vendor, where we spend tens of thousands of dollars a year. I'm not surprised at all now to hear of quality issues with consumables.

[boulderco]

I believe my coring issues are related to the ALS towers being slightly wobbly and not completely stable. Im going to 3D print some better stands for them and I think that will solve the problem. The issues are intermittent and I can see the tower move slightly when injecting. Oddly enough, I used a pack of septa from the late 70's that had been sealed well and they worked perfectly

Luckily I have had no problems with these thermogreens while manually injecting and only intermittent with the ALS. These are all I have ever used though, can anyone point me in the direction of something they prefer more?

[benhutcherson]

Glad it's working!

I would have never thought to update the G1512A firmware. Just out of curiosity, how is it done? Do you connect it to a computer by the RS232 port and do it that way, or by the HPIB port? Or is it done through the 6890?

I figure it's good to know.

As to your coring issue and mentioning movement in the tower:

I've worked with probably dozen G1513 towers, both 7673 marked and "6890 Autosampler" marked(same thing, just the latter from when HP was branding everything 6890...) and all were rock-solid whether mounted on a 5890 or 6890.

I'd double check first of all that your stand-offs are screwed all the way down tight. IIRC when Geoff installed mine on the 6890, he used a screwdriver for the last fraction of a turn. I've installed a fair few on 5890s and I've done the same. I've not tried to remove my 6890 ones, but I'm pretty sure I'd need one to get them loose.

I've only taken the ALS mount off my 6890 once, but I seem to remember there are maybe 6 or 8 screws holding it down? Be sure they're all there. Also, just if I may be permitted to post a reminder that you probably already know, but the screws are Torx head, I think T15? but don't hold me to that. In any case, be sure you have them down snug with a properly fitting torx driver.

If your tower is really flexing enough to cause problem, something is going on. Across the 7673 and 7693(I've never used a 7683) I've found HP/Agilent seems to fundamentally have a rock solid design for mounting and alignment just really doesn't seem to be an issue unless the tower is so badly worn that it's probably in need of a full rebuild. I'll also compare that to the Varian ALS I fought with for a while(8600? Don't hold me to that) on a 3800 where I felt like alignment was a daily procedure, and there was still a good chance I'd end up with a bent syringe...when I bought that system(second hand) I was given the option of that or an older PAL sampler, and I really regretted not getting the PAL.

[benhutcherson]

BTW, as far as septa go-

It seems that septa color go in and out of fashion. I've been a long time Thermogreen user, but all the stock I have is pre-COVID.

At a previous job, I had a connection at Chromatography Research Supplies in Louisville, KY. He was one of our former PhD students who landed there after graduation, and would come often to use our FTIR and other instruments they didn't have.

In any case, their "standard" septum they offered was red BTO rubber. I have a few jars of these around with their branding, but it's also the same red septum that Agilent sells in a blister pack-and yes I know this because I had more than a few blister packs around he'd given me that QC pulls.

The times I've used these septa, I've been super happy with them, whether out of an Agilent blister(I don't have any of these now) or from a CRS jar.

[boulderco]

Thanks for the advice! If youre ever in Denver, come swing by the lab and I'll buy you a beer! (We are next door to a brewery, just ran some of their beer through the GC-MS and let me tell ya, there is a lot of compounds in there)

The ALS controller hooked directly up to the computer with a serial cable and I had to download agilents firmware tool to update it.

Luckily I think I've solved the coring issue, the towers I have on my Frankenstein system dont quite fit the mounting tray so they were a bit wobbly upon injection. I drilled a few holes in the towers and the mounting tray and secured them with brackets and it seems to be working great now. Although out of nowhere the MS wont tune. It looks fine at first and then gets extremely noisy and the abundance goes down. Theres plenty of PFTBA in the vial and ive tried to reseat it multiple times and purge it to no avail. Any ideas? Perhaps the solenoid went bad and its a coincidence?

[70 eV]

Why is your tower moving? Stop drilling random holes. I'd say buy a merlin if you were worried about coring but you'd probably just Z the needle at this point.

[boulderco]

The tower was moving because the ALS mounting bracket wasnt configured correctly for it so the towers would shake ever so slightly on injection. Drilling the holes through the mounting bracket and securing the towers better fixed the problem completely. I was looking at the merlin microseals but I already have a large surplus of septa and theyve been working fine for years, do you think it would be worth upgrading?