Varian Galaxie/ProStar HPLC Injection Problem

[DSV]

Hello,
I'm attempting to run a semi-prep affinity separation to extract an enzyme from blood plasma using a self packed low/medium pressure glass column. We recently bought a prep LC kit to allow us to do this work on an existing HPLC system from Varian, running Star, to save on bench space. This had been going OK until recently. The problem is that the kit I'm using doesn't want to inject the entire contents of a sequence list. Typically I'll get all but the last injection through, but now it's started to give up after the first 2 or 3. Also the system can take several attempts at starting a sequence, although at other times it starts first time

The stand alone (i.e. not networked to an intranet etc...) system I'm using consists of a Model 410 autosampler with prep option installed (Varian badged, but made by Spark Holland), ProStar 240 Ternary LC pump (Varian), 325 UV detector (Varian) and a Model 701 fraction collector. The CDS is Galaxie v. 1.9.301.220, also from Varian.

I've read a comprehensive debate about the merits (or not) of Galaxie on this forum but no one mentioned a problem such as this. Things were going quite well with Star, but that wouldn't control the fraction collector so Galaxie was installed to allow that functionality.

Does anyone have any thoughts or suggestions? The engineer has been called but won't be available until mid-november. Through talking to Varian they seem to have an issue with the PC used, which formerly ran the system when it was running Star. However, the spec is fine according to their own (Varian's) literature.

Thanks,
Dan.

[Chris0000]

Hi

there are several possibillities for this behavior. Strangely, sometimes it seems to work, and sometimes not, is that right?

This can be caused by a blocked arm of the fraction collector. At start, the 701 goes to the zero position (backwards). If the arm cant reach this, run will not start or sequenze will be abortet. also the Tubing to the arm can cause this, if it´s too short.

Another option is to check all cable connections, if theres a weak contact;
power save settings of the computer are a good reason for aborted sequenzes, disabling them will help and you can check you hard disk, if theres enough space left.

And last not least, maybe Varian service might help you



I hope I did help you
Chris

[DSV]

Hi Chris0000,
thanks for your thoughts. Yes, as you say, sometimes it goes and other times it doesn't. The original input from Varian was to change the comm cable, which is a ribbon type cable that daisy-chain's several of the modules together. Could be the source of the loose contact you suggested. It has just taken 3 sequence restarts to inject 8 samples. It went as a 2, another 2, and then it did the last 4 injections without any intrvention from me. Luckily me enzyme is very hardy so it hadn't denatured and it could still interact with the medium. I'll post the outcome once Varian have been to try and sort the issue out.
Dan.

[sdegrace]

Regrettably, we have much experience with things that worked without a hitch with Star that suddenly become chronically problem-prone with Galaxie. Things like sequences not wanting to start, or freezing for no apparently reason part way through have been common. 9 times out of 10 the problem is with the software, not the hardware. We use Galaxie to run Varian LCs.

The only thing I can say is that eventually Galaxie became much more reliable, to the point where it does not actively cause me physical pain to work with it and I can feel reasonably confident that if I walk away from the sequence it will finish. The solution was to have the Varian people come in and apply several rounds of software patches over a period of more than a year. There were also some network incompatibilities that made our installations less stable. There are some things that are very likable about Galaxie as compared to Star (although it still has quite a ways to go to be as good as Empower, IMO) and now that the system is more stable these elements are more obvious. There are a lot of genuine improvements in Galaxie. Exhibit A is checking your sys suit! Also I love the way they expose an API to let you define custom variables - the concept at least is cleaner than the equivalent feature in e.g. Empower.

My advice: If you haven't done it already, get a Varian rep to come out and apply every known software patch. You definitely don't want to do it yourself if you can avoid it... I've watched them work, and they have to twiddle a lot of settings deep in the bowels of Windows. That will probably help, and if you do also have hardware issues it will make them a lot clearer.

Stephen

[Chris0000]

Hi

If I remember right, the 230/240 Pumps have a GPIB connection.
I think there had be some issues between GPIB and Galaxie and a patch or new driver has been released...maybe its worth to ask the Varian rep for that...and the other updates

best regards
Chris

[DSV]

Hi,
the problem has been solved following a visit from a Varian engineer. I should have mentioned this before, but didn't, but the LC has a fluorescence detector in addition to a UV detector. I don't use the fluorescence detector for this work so it's switched off but was not removed from the configuration. In the method the 363 fluorescence detector was described by me as being "offline or not in use", which I thought would be enough for Galaxie to know to ignore anything to do with this 363 module. Unfortunately not. The detector was taken out of the config and the hard disk defraged and so far after 3 rather long sequences all is good.
Dan.