Abnormally high first value(s) when ROM scanning on older SX100

[Karsten Goemann]

All,

When ROM scanning (peaking, integrated intensities...) on our SX100 (#846, installed October 2003) we regularly get intensities which are anomalously high at the start of the scan. Most of the time this happens on our spectrometer 3, sometimes on 4, but almost never on the others. Most of the time only the first value is high, every now and then the second or third are also affected.

With John's help we seem to have fixed this now by increasing the SX100ScanSpectroMilliSecDelayBefore value in probewin.ini to 2000ms. Originally this value was set to 200ms. At 1000ms the high values still occur.

We don't see these spikes in PeakSight so Cameca seem to have some way of dealing with it as well.

Cheers,
Karsten

[Probeman]

We don't see these spikes in PeakSight so Cameca seem to have some way of dealing with it as well.

Yes.  I assume Cameca is using a similar delay.

Interestingly, Gareth and I with newer Sx100 instruments do not see these anomalous intensities in the first point or so of scans, even with only a 200 msec delay, so they must have fixed some noise problems apparently caused by the spectrometer moving to the start scan position (I added the delay Karsten mentions above, just after moving to the spectrometer scan start position).
john

Edit by John: Karsten finds that adding a 2 second delay after moving to the start scan position, prevents the anomalous noise he was seeing in the first one or two data points.

There must be a firmware difference between the old and new SX100s, as Gareth Seward reports that the newer Sx100s do not seem to require this short delay before a spectrometer scan starts.