Experimental Physics and
Industrial Control System

Mark Rivers via Tech-talk <[email protected]> · Fri, 8 Nov 2019 20:03:01 +0000

Hi Matt,

In ADAndor R2-5 we fixed 2 problems:

- The AndorPreAmpGain record, which contained the enum choices for pre-amp gain, previously actually did nothing.  There was no code in the driver to handle that.

- The ADGain parameter, which is a float64 ao record, was controlling the pre-amp gain via this SDK call:  SetPreAmpGain((int)value).  That is really not very good, we were using a double to select an enum.  There is no guarantee that different cameras use the same enum values, or even that they will be the same from one firmware release to the next for the same camera.  So ADGain is no longer used to control the pre-amp gain.  Rather the AndorPreAmpGain  mbbo record does this.

The available enum choices for the AndorPreAmpGain are determined at run-time using the asynEnum interface, so the choices presented to the user are guaranteed to be valid and those that apply to the selected camera and firmware.

I think this fix justifies breaking backwards compatibility, since ADGain was really not the right PV to control an enum feature.

Mark

-----Original Message-----
From: Tech-talk <[email protected]> On Behalf Of Pearson, Matthew R. via Tech-talk
Sent: Friday, November 8, 2019 9:44 AM
To: [email protected] Talk <[email protected]>
Subject: ADAndor iKon CCD low saturation point

Hi,

We have an Andor iKon CCD that we recently updated to RHEL7 from RHEL6. This involved updating the Andor SDK2 from version 2.98 to 2.102, and the ADAndor module from 2.2 to 2.8. 

We’re seeing that the CCD saturates at a lower point compared to before. Before, we could get >50,000 pixel counts and not see any saturation effects, but now we only get about 35,000 counts and we start to see saturation. That’s with the PreAmpGain set to x1. If I increase the PreAmpGain then I can get reasonable results and use the whole 16-bit ADC range before we see the CCD saturate. 

I’m wondering if anyone else has seen this issue, or can point in the right direction? 

The driver in ADAndor has some extra support, related to setting the frame transfer mode, vertical shift period, EM gain, read out mode. Not sure if they affect the CCD saturation point. I tried varying vertical shift period but it didn’t affect things much. And the iKon doesn’t have an EMCCD so i don’t think EM gain applies.

There was a change related to setting the pre amp gain on Jan 3, 2017, which seems to change which PV is used to set the pre amp gain (ADGain to AndorPreAmpGain)… and, as I’m writing this, I realize this might be the issue:

My current value of ADGain is 1. Setting this in the Andor SDK corresponds to x2 gain. So we were using x2 before. 

My current value of AndorPreAmpGain is 0, which corresponds to x1 in the Andor SDK. 

So I guess that might explain it. It seems like that commit didn’t have to break the existing ADGain function, in order to get the AndorPreAmpGain record working?

Cheers,
Matt

Data Acquisition and Controls Engineer
Spallation Neutron Source
Oak Ridge National Lab

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System