Chromatography Forum

We are experiencing issues with some of our GC-MS methods becoming corrupted and no longer useable. This has happened to us recently, twice, on two different systems. On one system, our 6890 GC with 5975 MSD, the run method error message was displayed after changing out an ALS. The second time, two methods were loaded into a sequence list. The first method ran fine, and the second method accidentally had the wrong column length entered into the method and the sequence list halted, generating a RUN METHOD ERROR. Both of the methods loaded in the sequence list were subsequently unusable after this run method error. It appears that if you make a mistake on a sequence list, the method becomes corrupted. Has anyone else experienced this problem before? Methods becoming corrupted for no good reason, simply because you made an error on your sequence list? A similar problem was noticed recently on our 6890 GC - 5973 MSD system when a sequence glitch corrupted several of our methods. Another weird observation made was that 3 saved methods were corrupted, would give a run method error when doing a single injection or a sample list. 1 method worked just fine that I tried, but after I edited this working method to change the oven ramping condition, the run method error came back. At one point I thought that maybe a Windows update had messed up ChemStation since both of our GC-MS systems experienced this run method errors pretty close together in time (in the past 2 weeks). Somehow, everything seems to be working fine again, not exactly sure how the problems got fixed but essentially we had to recreate our original methods. If anyone can shed some thoughts on this, I'd appreciate it as we don't want this to happen again. I'm not exactly sure of the version number at the moment since I am at home, I can post it tomorrow, but it is the Enhanced Chemstation, a version that was released in 2006 or later.

We have also heard different stories regarding the computers being used to run an Agilent GC-MS. Can the computer be connected to our internal company network? Does this cause instrument problems? We have the computers removed from the domain and there are no firewalls turned on. Is this the way it needs to be configured to run in a stable fashion over time?

Thanks again.

You can connect your 5973/5975MSD computers to an in house network by installing another NIC card but getting it to work correctly is a bit of a pain. Basically each time you restart your computer you may need to do some fiddling (toggling the in house network and MSD network connections on and off) to make sure the computer correctly assigns the Network and Chemstation correctly.

For example what we do:

Restart the computer, turn off the network connection to the GCMSD, use the in house network to open up a printer tab, close the in house network, reopen the GCMSD network connection and start chemstation, then reopen the in house network. If we don't do this sometimes chemstation assumes the MSD is already in use and refuses to load.


Also if you do add them to the network you'll need to make sure you keep the .net framework running at the current version, upgrading to 3.0 caused severe issues for us.

Please DO NOT use two network cards, it is not supported or recommended by Agilent. Use a switch or router.

There are many ways in which your methods can get corrupted. Copying a method from a different instrument, incorrect column configuration, things like gas saver or auto prep run, and then just random little mistakes that wind up getting propagated.

I would recommend going from the default method (default.m or def_gc.m) and then recreate all your methods, once you have them how you like them make a backup copy on a CD or somewhere else on the drive.

Wow, there are so many possible reasons for your problem including a corrupt operating system and/or ChemStation software (you did not say which version?). Obviously, your software has been corrupted at some point (again, a million ways this could happen). The methods and sequences that are corrupted are not going to cure themselves w/o fixing the main source of corruption.

Your statement:"It appears that if you make a mistake on a sequence list, the method becomes corrupted. Has anyone else experienced this problem before? Methods becoming corrupted for no good reason, simply because you made an error on your sequence list?"

Your observation is just that, an observation. If you make a mistake in a ChemStation Sequence, it should not corrupt your methods unless ChemStation too is corrupt. Backup any important methods onto a separate drive/disk too for security. If your ChemStation software is corrupted and an error occurs it can indeed corrupt everything it "touches". Without detailed information about your issues the quickest fix is to clean up your computer right down to the operating system (as in wipe it clean and install it fresh) and then install ChemStation onto your computer. If you transfer any corrupt methods or sequences to the new system, then they will of course still be corupt so you may have to re-create them from scratch. This may be the quickest way to get the system back to running 100% once again. *You could also try and remove just the ChemStation software alone, reboot the system (cold), wipe the ChemStation from the Windows Registry and then re-install it fresh w/o refreshing the O/S too. At this time it does appear you have at least corrupted the ChemStation core files or kernal so this might fix it.

As for what may have caused the problem to begin with ? Again, too many possibilities, but the most common ones we see are when people use their instrument computer to surf the Internet or load unsupported or corrupt files on the same. The computer is always the weak link. As a system gets used it will eventually fragment itself to the point where errors begin to add up and a noticable failure will result. *Remember to defrag the hard drive every week; keep the system clean and do not use it for other "non" ChemStation tasks. Having two separate NIC's (if they are set-up correctly) is fine and can reduce network traffic on the acquisition side of things (yes, so can a properly setup network Switch or router too), but I suspect that has nothing to do with your current problems.

You're right that it probably has nothing to do with his method problems but really, do not install two network cards. Windows XP plug and play can randomly adjust network priorities and cause communication problems. Its true that if the MS network is set ahead of the other network card in priority that it'll probably be OK but there are other ways it can mess up and just using a router is much safer.

Don't put it on a domain either, and use static IPs. Don't include the chemstation directory in virus scans. Don't turn automatic updates (do them yourself when you aren't acquiring data).

I had an error like this with Agilent chemstation. My sequences would randomly stop.

I'll tell you what Agilent told me.

Defrag your hard drives and create new methods from scratch.

It solved my issues.