HPLC is Leaking

[Jumpshooter]

I am a novice user of HPLC 1090. I have been told that leaks are common to this analytical system---where exactly are leaks most prone to occur?

[Kostas Petritis]

Valves, tubing connections and "pump heads" (i.e. pistons seals etc). In analytical HPLC you should be able to track them down pretty easily. Also the HP/Agilent pumps have several leak detectors in key places.

A quick test you can do, is to add a column in order to create some backpressure, have your pump working at the adequate flow rate (i.e. 1 mL/min if your column is 4.6 mm ID) and then measure your flow by collecting at the end of your column. If your nominal and measured flow rates are the same and you do not see any leaks after careful inspections of the places I mentioned, you should be OK. Keep also in mind not have a max pressure cut-off for your HPLC system (it will depend on the column you use, otherwise for the HPLC system 400 bar is a good max pressure for your pumps).

[Mark Tracy]

I'm sorry to hear that you were given the 1090. They were great in their day, but that was 15-20 years ago. The design is not like any other, and it has unique problems. I don't even want to remember the gruesome mistakes I made with my first 1090 back in 1986.

If you can find it, look at the operator's manual and the maintenance section. It will explain what is going on in there. Do a thorough visual inspection. I'm going to assume that you have the automatic injector; look at the needle seat, and the injection valve. Remove the solvent reservoirs, and the covering behind them. Look at the metering pistons; you may have more than one depending on the model. Check the oil in the hydraulic booster pump. There is a heat exchanger in the column compartment; the fittings may be worn out, and never seal properly; if so either replace the tubing union, or bypass it altogether. The flow cells on some of the old UV and DAD detectors were put together with epoxy glue, and that can fail.

[Jumpshooter]

I guess the collective opinion is that the LC 1090 that I'm using is not the best choice---but given our lab's budget it's the only thing we could afford. WE purchased it at an instrument auction website (can't rember the name?). At any rate, this is the card I was dealt ("on-the-river Queens") so I need to make do with this system. Furthermore, I cannot find a User's Manual anywhere!

[Mark Tracy]

I'm going to assume that you just had to take someone's word on it that the machine is in workable condition, and all sales are final. Have Agilent come out and service it; if Agilent won't do it, there are independent service outfits. If necessary beg your boss for it. Neither you nor s/he can afford the complications of learning a new technique on equipment of uncertain condition. The validity of your work is at stake. Then when the service rep shows up, watch closely and ask lots of questions, because you will be doing it the next time.

You might go back to the auction site and see if any Operator's manuals are for sale.

[Jackus]

Dear Colleague,
If you will search for manuals, there are two parts. The most important is part II (Maintenence guide, Troubleshooting guide, Touring the HP1090). I will try to scan important parts to PDF for you.
It is true that concept of 1090 system was revolutional but its sensitivity and noise is not comparable with present instruments.
Regards