CAMAC Progress Report

                      Week Ending: July 26, 2002
                               Week: 78


Tom Folkers:

Tom Sargent:
     1. We had some network or NFS problems communicating with the 
        computers at the SMT (smtocont, smtots).   It hangs up waiting 
        for disk I/O.  It may be that the NFS daemon is crashed. We 
        didn't have any trouble last week.
     2. Went to SMT and fixed a number of problems. I altered the DSP 
        code and servo to change the corrections report command. It 
        now reports only on one encoder at a time. I also changed the 
        transmissions of speed change messages from servo to DSP in 
        simulation mode to not send 2 consecutive messages unless 
        separated by at least 10ms.  I think these 2 changes will make 
        the simulation mode stop crashing.
     3. I started having some problems with servo deciding that the 
        polynomials from track are too old. I made some changes to the 
        code that determines this. It still fails occasionally wen the 
        program is first started, but recovers gracefully.
     4. The 'hatches' command was tested under different conditions 
        (sun up and down) and seems to be working.
     5. The setdir command works (disables sun avoidance mode).
     6. I fixed the 'stop' command's tendency to stop the azimuth 
        motion too abruptly.
     7. The validate() fcn which controls the ON_SOURCE SAMBUS bit 
        seems to work now. It was working when a new object was 
        selected, but not when an offset was commanded.
     8. Twice we saw the encoder signals (the digitized versions) 
        become chaotic and/or noisy.  This is probably responsible for 
        servo going haywire in at least one case.  I wrote a command 
        to capture more data in these conditions including 3 seconds 
        of raw encoder data.  I didn't get a chance to use it yet.  In 
        one case the A/D boards had to be power cycled in order to get 
        AZ cosine signal to be normal. 
     9. Added the 'manlog' command which allows the operator to type 
        in text directly to the servo.log file. 
     10.  Added the 'purge_logs' command.
     11.  Helped Bill check out the SAMBUS 'latch' option. Apparently 
        it works on some SAMBUS signals but not others.
     12.  Wrote the suspend() command.

Pete Hubbard:

David Forbes: