I have Waters' Millennium32 ver 3.2 with 4 HPLC systems attached (all Waters equipment: (1) seldom used older system connected with SIM box (2)2690 with PDA and 474 fluorescence detector (3)2690 with refractometer and 486 UV (4)717 autosampler with two 510 pumps, 486 UV and 474 fluorescence detectors. Last June, the 2690 with PDA and 474 were installed. I upgraded Win95 to Win98 and upgraded Millenium. My memory is a little fuzzy about which came first the upgrades, the new equipment or the problem because the problem is intermittent.
The problem: I put a sample set on to run overnight (any system) and part way through, it quits and gives the message 'Instrument Failure', but I have no way of knowing what instrument. Sometimes the pump(s) stop and sometimes they continue pumping (wasting my mobile phase). When I return the next AM, I can click on the green button and the set will continue - sometimes. Maybe several samples will run, maybe only one, maybe I get the 'instrument failed' message again and have to exit everything and do a cold boot on the computer, at which time the set will run.
Remedies tried to date: new buslac/e board, shortened cables, uninstall-reinstall Millennium, and new ver 4.2 chips in the 486's.
I've been running Waters' equipment for years and cut my teeth on Baseline and Maxima. This is the first time I've had a real problem. Is it hardware, firmware, software?
I'm open for suggestions or ideas.

maybe the millennium-forum on yahoogroups.com is a better place for this question.

One question: Is your computer on a corporate LAN? I experienced the same problem with our network. Our IT dept. backs up the network every night at midnight, so if you are still logged in, you get logged off. This caused Millennium to quit running in the middle of a sample set. Also, I would recommend that you don't run Millennium 32 on a Win 95/98 environment. Windows NT 4.0 SP 6 is a much more robust OS. I'm not sure if Millennium 32 has been validated on Win 2000, but that might be an option as well.

Good Luck,

Chris

Because I've heard horror stories of HPLC systems that are on networks, mine is a stand-alone system. I've considered electrical interruptions, but no other equipment in the lab indicates this as a problem. Also, the problem has occurred during the day. The times of the problem are inconsistent. I don't think that Waters has Millennium to go with Win 2000. I usually upgrade the software as it becomes available.
I realize that most of the problems listed at this site are chemistry related, but I thought possibly someone has seen this and conquered it.

We have Millenium 32 on a client server, and I have seen similar problems (though usually our run quits in the middle of the night, much to my chagrin). After talking with Waters, we have concluded that the problem is that the instrument somehow loses communication with the computer. Since the software can no longer find the instrument, it gives an instrument failure, and the pump just keeps on pumping valuable solvent because it has not received instructions not to do so. When this happens we have to reboot the computer. We are unsure as to what causes the instrument to lose communication with the server.
Sorry this is not much help, but it may be a reason.

We had similar problems early on with Millenium and Waters equipment and the cause seemed to be isolated to a printer sharing software program called "ParaShare". Once we got rid of it, the problem dissapeared.

Anonymous,
How did you isolate the problem? The problem is so inconsistent, it is difficult to pin down the cause. I have very few programs on this computer, mostly MS Office. I am also having a problem with the Oracle program (which is part of Millennium32) not closing properly but don't know if the 2 problems are related. I didn't have the problem until I had the last 2690/PDA/474 installed. At the same time, I upgraded software. Which, if any, is responsible?
Chris, did your Waters service rep not have any solutions? I know that there is a lack of communication between the computer and the injector at that point, but why? When I leave 120 samples on to run overnight, it's a little disheartening to come in and find them unanalyzed the next AM and mobile phase going into the waste container.

This may be completely worthless but we had similar problems with Shimadzu systems. In our case, there was a delay line in a .INI file that (if I am remembering correctly) controls how long the computer waits for system readiness. By extending the delay time, we were able to fix the problem. It seems the newer computers are too fast and needed to be slowed down for the system. I don't know if there is anything comperable for the Waters system.

Sue,

No, I didn't even mention it to our Waters service Rep. Our problem was remedied by telling our IT dept.not to automatically back up or log off that particular computer with the Millennium software. Since, you are not networked, it might be something quirky with your Bus/Lace board, or as mentioned above-your computer. Did your computer come from Waters along with your system? If so, that may be your saving grace. If you purchased your own computer, it will be harder getting support with Waters regarding this issue!

Chris