UPLC Retention time / Pressure Fluctuations.. help

[mnrevoy]

Hello all,

We are having significant issues with our Acquity H-class UPLC.. We recently noticed that the retention times of all of our analytes were shifting longer and longer, forcing us to continually alter the processing method. The pressure is also all over the place. There doesn't seem to be a trend one way or another. In order to try to figure out what was going on, we started to systematically check all of the system parts. When we ran the static leak test, it failed for the primary. After a plethora of further testing, we replaced all of the primary seals, the primary pump head, the connection between the primary and the accumulator, and the needle seal and sensor . The leak test finally passed when we excluded the needle and needle seal, i.e. passed from the QSM straight to the column, but failed when included. The needle seal readiness test also keeps failing. We are at a complete loss as to how to proceed... The UPLC have been out of commission for quite a while now..

Mobile phase: 75:25 ACN:0.35% formic acid using a BEH C18 column

Does anyone know where the issue might be? Any and all insight would be much appreciated.

I will also note that we performed a PM this past November, and are quite worried as to why the system seems to be failing already...

Thank you!

[tom jupille]

Did you do the PM yourself, or was it done by Waters or a 3rd party? And did the system pass PQ afterward?

[Gerhard Kratz]

Did you check for leaks??

[mnrevoy]

Hello,

Tom, we did do the PM ourselves. Our lab manager has been trained by Waters to perform the necessary maintenance, and yes the system did pass afterwards.

Gerhard, as I mentioned, the leak test we performed failed, indicating that there must be a leak.. somewhere. The leak however is probably so minute that it would be impossible to see with the naked eye, i.e. ~1500 nL/min of MeOH, making it very difficult to figure out where in the system it is leaking..

[Gerhard Kratz]

Take off your column from the system, place white paper tissue to the critical parts like pump heads back side, injection system and than flush your system with maximum flow rate.

[Hollow]

Use a strip of thermoprinter-paper for searching micro leaks. organic solvents (even vapours) will turn it black too

I'm not familiar with the H-class especially with injector.
are there any seals or tubing connections failing?

Beside failing some 'tricky' pressure test, what about the good old volumetric-flask flow test, and gradient step tests?

[adler_LAB]

check with a volumetric flask if the pump is able to have the flow you need. when the retention time gets worse it could be the piston seals when there is no visible leak. cause it can't do enough pressure with the needle, but without its able

Also check your data. does the software store the pumppresure in every run ? how did it change ? can you see a difference with A and B if you run gradient.

Sorry i didn't have time to read your whole posting

[mnrevoy]

Thank you all for your advice.
We had a Waters technician come in and fix our problems.

[kjones0000]

What was the required repair, out of curiosity?

[mnrevoy]

The Waters technician performed a whole new PM on the SM-FTN. He replaced the injection valve, seals, needle, etc.