Chromatography Forum

I have a phenomenon that I'm sure has an easy and quick explanation, I'm just to the point where I'm overthinking it.

We have seven Perkin Elmer Series 200 HPLCs. Each component is the same model number for each instrument. They all have identical sized loops and syringes. We recently performed a generic PQ for the systems. We used the same solutions on all systems with the same injection sequence. Six of the systems provide approximately equal responses, however one system was consistently ~50% of the response given for each injection. This was baffling. We theorized that someone might have put in a different flow cell (we had bought all the systems used), so we switched the flow cell out with one of other instruments. Again we received the reduced response. The autosampler appears to be performing well. The system passes injection linearity, injection precision, as well as a 10 standard linearity run. It is working just fine. We are just very confused.

Like I said, there is probably a very simple explanation. If anyone has any ideas I'd love to hear them.

Is there an attenuation or range setting on the detector? You might get some insight by cranking up the sensitivity and looking at the noise level. It that's also 50% of the other systems, that would point toward an electronic problem. If the noise level is consistent with the other systems, then the next suspect is the autosampler (just because it's precise and linear doesn't guarantee that it's accurate!).

That said, since calibration has to be done on the instrument you're using, I wouldn't be too concerned about it unless you're doing trace analysis.

The baseline does not appear as "silent" as it does on the other systems. We did attribute this to the reduced response. Typically it isn't an issue, except on methods where we don't inject a high amount of analyte.

I've made 10 100uL injections of water and have the expected 1 gram difference in weight.

I will look into trying to find an attenuation or range setting.