ELSD Signal

[wchriste]

Hello,
I have recently inherited an HPLC equipped with a Sedere ELSD. The instrument/detector was fully functional when it went down. Someone flipped the power switch and the system sat intact with ACN/H2O mobil phase for about a month. Powered back up the system seems to work properly, but I get half the signal. Standards calibrate with eachother, and check sample numbers come out green. It doesn't matter if I inject 10uL or 20uL, the signal is exactly half of what it was previously. We have burned off the detector a couple of times, then just gave up and replaced the lamp, nebulizer, and glass tube. Changing the HPLC column yields similar results. It is eerie that the signal is almost exactly half. This leads me to look for something in the settings, but detector temperature and gain are consistent with previous operating conditions.
I am new to the ELSD. Is there something I have missed?

Thanks,
Wendy

[Kostas Petritis]

Which Sedere model are you using? Have you played around with the gain control?

[wchriste]

It is a Sedex Model 75, and I have played with the gain. Increasing to 9 gives me a more intense signal, and 7 gives me less. But I have been warned not to run the method with the larger gain because the method was validated at 8.

[Kostas Petritis]

From what you are saying, you are sure that when the results with 2 fold more signal were obtained, the gain was set at 8? Can you confirm that?

Also what about the signal to noise ratio? Are there the same or different?

[wchriste]

I am 98% certain that the detector has always been run at 8 gain. (2% chance that this is some joke meant to have me chase my tail for a while. )The signal to noise ratio is good (the same if not a bit better). I recalibrated the detector yesterday with no change in intensity.

Thanks for taking an interest.

Wendy

[wchriste]

Going through the data generated with the same standard in the past year, I have noticed that the peak heights are quite variable. (1x, 2x, and 4x) Unfortunately mine are the 1x - smallest on record. Any ideas what would cause this variability while maintaining the integrity of the method? i.e. standards calibrate and checks pass

Thanks,
Wendy

[tom jupille]

The signal to noise ratio is good (the same if not a bit better)

Just to check: your peaks are half as big as they were before, but your noise level is only half (or a bit less) of what it was before, right? This is about what you would expect if the photodetector was dirty.

[Kostas Petritis]

Wendy,

Just to clarify, you get peak heights that are variable (1x, 2x, 4x) but the signal to noise ratio is always the same (i.e. let.s say 20)? Is that correct?

If this is the case, and the gain set was always the same, I do not think that the detector has nothing to do with it... A pollution or drifting of the detector would give you random values and not exact multiples...

Maybe there has something to do with the software you are using to monitor the ELSD (although I can not think of something specific)...