By Hazem on Friday, June 11, 2004 - 12:14 pm:

Hi every one
My alliance autosampler can not inject accurate injections also I have changed everything inside and even technician of agency could not fix it nny help please

-------------------------------------------------------------------------------------------------------
By A.Mouse on Friday, June 11, 2004 - 02:43 pm:

I don't have this system, but my first question is, if the inaccuracy depends on the injection volume or not.

-------------------------------------------------------------------------------------------------------
By R C on Friday, June 11, 2004 - 08:53 pm:

O.K. So the instrument tech has replaced components and it still fails. Let's see what we can figure out ...

First, is it working properly. There's an injector calibration protocol given to you when the instrument get preventive maintenance. Basically, you put 1000 uL of water in a vial, mass it, inject 100 uL 5 times, and remass. You do this with no column, water as eluent, 1 mL/min flow rate, short run time to minimize evaporation from the vial. Use the density of water at ambient to see if the injector is pulling the correct volume from the vial.

Once you're sure the instrument works according to vendor specifications, you can repeat the procedure with your volumes. What volume do you start with in the vial, and how much do you inject? Try those volumes with the mass test to see if the instrument can work with the volumes you need. Example, many instruments tend to lose precision if you try to inject 100 uL from 105 uL total volume. Sample viscosity, bent injection needles, misaligned vials, overfilled vials creating a vacuum during injection, are all sorts of problems that seem to come and go.

'Course, we haven't even touched analytical problems -- integration failures, sample degradation during run, incorrect wavelength, and so on. But massing the vial before and after will let you know if the injector is functioning.

-------------------------------------------------------------------------------------------------------
By henrik on Wednesday, June 16, 2004 - 12:27 pm:

Normaly a change of the rotorseal, and wash with 5-10% HNO3 would do the job. Maybe you sample is not stabil. also check the syrring

-------------------------------------------------------------------------------------------------------
By A.Nonymous on Wednesday, June 23, 2004 - 02:26 am:

Try some injections of coffeine (on a short RP18 column with methanol water 40/60 to keep the retention time low). You can also add some Uracil to see if the void time variates (to see there aren't any leaks that cause your problem).
Instead of using coffeine, you can also use a paraben like methyl paraben, ethyl paraben, ... but be sure that the peaks aren't in the void volume.

Make 6 injections and watch the %RSD. If you post your handles and values, maybe we can help you.


Regards