Use exposure time for normalization

[YooSunYoung]

For now, white beam, histogram mode detector data doesn't come with shutter-speed(exposure time).

Per measurement (per file) it doesn't change, but it can vary in different files.

The current normalization method doesn't take it into account.

Update From ECDC
ECDC will add this field as a NXlog and it'll be somewhere under the detector group such as /orca/acquire_time

[YooSunYoung]

@MridulS This is related to the normalization.
I think the exposure time should be exposed as a parameter (from the perspective of sciline) so that user can set it.
It won't be easy to automate finding this parameter in a non-ess-nexus file.

But this issue is about nexus file, so we should be able to find this exposure time in the ess-nexus file.

[nvaytet]

If the detector is the same as the CMOS detector on the TBL, then I expect the data to be recorded as a 2d histogram, with dimensions frame_time (=event_time_offset) and time (equivalent to event_time_zero).
So I guess the exposure time is trivially computable from the length of the time dimension?

[YooSunYoung]

So I guess the exposure time is trivially computable from the length of the time dimension?

Yes, but only if the exposure time per frame is the same.

It'll be trivial to implement this but just need to have this in place....

[nvaytet]

We also talked about using monitor counts for the normalization instead of counting time.
This would be more accurate if the source is not perfectly stable?

[YooSunYoung]

We also talked about using monitor counts for the normalization instead of counting time.
This would be more accurate if the source is not perfectly stable?

Yes. But we also want this option.
And I don't think it's one or the other. They can be used at the same time.

[nvaytet]

Yes. But we also want this option.

Would it be like a choice of normalization method like for the powder reduction when you create the workflow?

And I don't think it's one or the other. They can be used at the same time.

Not sure I follow: could you explain how it works to normalize by monitor counts and exposure time at the same time?

[YooSunYoung]

Not sure I follow: could you explain how it works to normalize by monitor counts and exposure time at the same time?

Exposure time is more about the shutter speed so it can be an extra parameter for normalization or you can simply apply it in parallel to the monitor-normalization.
I don't have specific workflow in my mind but it's just about utilizing the values we have for different situations.
Like... monitor not available for some reason...

[YooSunYoung]

And I think it's more about the optical imaging ...? so no monitor in that case... 🤔