CAMAC Progress Report

                     Week Ending: August 24, 2001
                               Week: 30


Tom Folkers:
     1. Other then frequent talks with Tom Sargent about control 
        system philosophy, not much don this week.

Tom Sargent:
     1. Wrote the analog.c  program which reads in the 16 channel A/D 
        card twice per second and stores the various readings (motor 
        currents, tachometer readings, etc.) in global memory. Also 
        computes average, RMS and peak values.  I altered the respond 
        program to copy that information across to the SMT control 
        computer. 
     2. Examined the CPU utilization problem that we noticed several 
        weeks ago. When running 'top' it occasionally showed the servo 
        program CPU utilization going up to nearly 100% for a few 
        seconds every 5 or 10 minutes.  This appears now to be a red 
        herring. It is an artifact of running the top program.  I was 
        able to have it detect and print out when it was slowing down.  
        This showed that it always occurred synchronized with a top 
        refresh operation.  I was then able to show that when the top 
        program is not running, servo does not show this effect.  
     3. I made a significant change to the encoder reset logic so that 
        it can operate on azimuth and elevation separately.  The first 
        version did both axes simultaneously.  Since we probably won't 
        be able to move the telescope in Azimuth on our next trip, 
        this change seemed to be necessary.
     4. Started working on having servo handle the 48 bits of parallel 
        inputs and 32 bits of parallel outputs. Bill is supplying me 
        with descriptions of how these various control and status 
        lines work.

Pete Hubbard:
     1. The move to new lab's down stairs and helping with the setup 
        of benches in the lab.
     2. Made the cables to allow testing VCO Mux chassis with the 
        counter interface chassis.
     3. Starting to make a KRUPP rack simulator for software testing.

Dave Ashby:
     1. No Report.