Pump cal failed

[lcrandall]

Hello,
I just did a calibration on my Agilent G1311A pump
h2o vs h2o 0.1% acetone
step gradient detected at 240nm
10%
50%
90%
100%
comparing all values to the 100% level, with +-1% spec.
My 50%value for lines A:B is 52.5% and fails.
Where do I go from here? Everything else looks fine. What type of maintenance repair should I investigate?
Thanks!

[ARBE98]

Have you tried repeating the calibration with the other lines (i.e. use A and C or A and D)? This could help you narrow down whether you had a problem with your multi-channel gradient valve or if it's one of the other valves in the pump. If you're not in a highly regulated environment, you might just be able to use the other lines that do work.

We've had trouble with the lines that pump ACN; these seem to go out more often and we end up switching the MCGV and it fixes the problem. Agilent offers a refurbished MCGV that's about half the cost of a new one - you just have to send in your old one.

Also, just to check - have you repeated the calibration a second time to verify that you really are having trouble at this particular step? Sometimes we get in a hurry and don't purge our lines (and especially the degasser) long enough and we get funny results on the first run.

[lcrandall]

Thanks for the reply,
I haven't tried to narrow it down to A or B. I will do that next.
We aren't regulated, so I can definitely put the failing lines out of commission. I would prefer that over an $$ fix!

[Bryan Evans]

If my memory is correct - our Agilent service engineer
used to use 0.5% acetone (he said 0.1% can sometimes fail the test).

[lcrandall]

Howdy,
I tried 0.5% acetone and still failed. Then, I narrowed it down to line A. Since we aren't in a regulated environment, I can just say "don't use A for gradient elutions".
But, I'm a bit obsessive and don't want to let this go just yet
I found an Agilent note for "EMPV Cleaning". (Electromagnetic proportioning valve, right?) Do you think that might help? I don't want to waste too much more time on this.
Thanks!

[mlsabe]

Have you try the pressure test and check if it passes?

[ARBE98]

I've never tried the EMPV cleaning. Let me know if it works!!

[detox]

lcrandall

Are you running the Agilent gradient tests from the Chemstation software in the automated test features? The tests are within the verification tab.

Agilent does use 0.5% acetone for the gradient test.

If you are using the chemstation software to run the automated gradient tests, then when the test is started the channel with the acetone tracer is purged for 9 minutes prior to the actual gradient test starting. If you are not using chemstation to run the gradient test then make sure you purge the channel with the acetone very well before the test. The G1322 degasser has about 20mL of volume in each channel. The G1379 degassers have 1mL of volume in each channel. When the acetone tracer is "trapped" within the degasser it is at a slightly different concentration then what is in the bottle (because of the acetone volitility and the degasser pulling vacuum on the degasser tubing). This difference in concentration can cause the test to fail.

Here is a simple test to determine if your MCGV is bad. Make sure all four channels are primed to the valve (use a priming syringe, don't use the syringe on the G1379 degassers it damages the membrane, ok to use on the G1322 degassers). Now disconnect the line from the MCGV to the active inlet valve (AIV), hopefully your mobile phase bottles are higher than the pump. If so you should see siphoning from the teflon line that goes to the AIV. Turn the pump off, this should close all four of the channels in the MCGV, if the siphoning continues with the pump powered off then one of the channels is stuck open. This will cause your gradient test to fail. To find the bad channel try to lower each bottle one at a time below the level of the pump, when the siphoning stops thats the channel that is bad. To try to unstick the channel flush that channel with about 4L of LC grade water that is heated to 60C, that will hopefully clear the channel. Repeat the siphoning test. If the siphoning still continues then most likely that channel is done and you will have to avoid using it. This quick test is not 100% but it will find if one of the channels is stuck open.

Hope that helps.

[lcrandall]

Hello,
Thanks for the tips. We ended up having to run an isocratic method on that instrument, so I haven't followed up with further troubleshooting.
Lisa