Experimental Physics and
Industrial Control System

Hi Mark,

 

Ø  1.) The biggest stumbling block was that the shutter never got configured properly. The reason being that
     the driver throws some „parameter not found“ errors, especially with the shutter mode. Needless to say:
    I never messed with ADAndors parameter system, so I am pretty sure that I did not introduce the problem.
    I hacked that by placing a suitable call into startAcquisition().

 

Can you please send the exact error message?  Your attached trace file did not have any “parameter not found” error messages.

 

Ø  2.) All the gain logic is wasted on the Ikon-M. The speed logic seems faulty. 

 

Those not very helpful comments.  Please explain exactly what you mean by each one.

 

Ø  3.) There is a missing parameter though: that is the VSAmplitude. I hard coded that one because I believe we only
     use one setting for that. I am aware that this is no general solution.

 

I just merged a pull request from Peter Heesterman that adds support for VSAmplitude.

 

Ø  4.) What is also missing is the SetShutterEx() support. Some of these cameras  apparently support an external shutter.
     This would require another parameter to control that one.  

 

My reading of the Andor Software Develop Kit.pdf documentation is that SetShutter can control both internal and external shutters.  The only time that SetShutterEx is needed is when the camera has BOTH and internal and an external shutter.  The driver would need to decide how to handle this.

 

Ø  5.) In my quest to get it working, I disabled some settings which I did not need and added what did not work. I attach my
     version of andorCCD.cpp. My (unprofessional) changes are marked up with MK in comments.

 

It would be more useful if you would do the following:

-          Create a Github Issue for each problem you uncover.  Provide documentation of the actual error messages.

-          Create a Github Pull Request for your changes to the driver rather than mailing it to tech-talk.

 

Ø  6.) Another thing which tripped me up was that the shutter opening and closing times were in seconds. Not in milliseconds
     as all the Andor documentation states. I only realized that when I found a 20000 in the trace where I expected a 20. I suggest
     to improve the documentation or at least set the EGU field in the DB.

 

All of the time related parameters in ADCore are in seconds.  This includes exposure time, exposure periods, shutter delays, etc.  Some vendor libraries use milliseconds, some use microseconds, etc.  For consistency we use seconds everywhere.  I agree that the EGU field should be set.

 

Mark