Unstable digital-like baseline on an Agilent 6890 NPD

[marcg]

Hi folks

I'm having some trouble finding the cause of an unusual baseline on our dual NPD. After months of working fine, suddenly one signal intermittently started showing an unstable baseline over a few pA (not huge but annoying enough) where it would suddenly drop & rise through the ramp (60 ->280, 15C/min), like digital "blips" - not your usual unstable baseline due to near-expiring bead). NPD cond's: 325C, H2 3mL/min, Air 60 mL/min, N2 7mL/min - no changes here for 10 years (if it ain't broke...).

After swapping the front & back NPD electronic boards and replacing the NPD bead on the affected detector, the problem was still there, which pointed to electrometer / NPD interconnect or a mainboard electronic problem (?). However, on swapping electrometers Front<->Back, cleaning collectors & replacing ceramics, the same problem was observed now on both detectors (although my background offset is now nice & low). I doubt the gases would be an issue since nothing has changed there fore a long time.

What am I not considering here (apart from calling an Agilent engineer)? Any tips would be much appreciated.

regards
Marc G

[AICMM]

marcg,

Not much help here but one thing to consider. You did all of the swapping I would have done and the problem appears to have stayed put. So now the only things left are column to front (let's assume front is the problem detector) and detector EPC to front.

Best regards,

AICMM

[marcg]

Thanks AICMM for the suggestions.

I managed to locate the problem to a poor interconnect-to-collector connection. After cleaning the spring at the end of the interconnect & reassembling, the signal was back to normal. Nothing obvious had changed here, but I'm wondering if the tarnishing of the surfaces over time/use could affect the connection between these components.

As an aside, I noticed that the Agilent instructions for reasembling the NPD detector are difficult, ie screwing the weldment down followed by inserting the interconnect into the side hole. I find it much better to install the electrometer first, ensure that the spring makes good contact with the side of the collector and then (carefully) screw down the weldment.

thanks
Marc