165 card database modification for 40 second FT
- Wally


We might have some problems running a longer (40 second) flat top in the summer time. We may find it necessary to change the length of FT as a function of temperature and cooling. The database entries for the SY 165 card ramped supplies have been changed to allow an easy way to change the FT duty factor. Here's how it should work:


165 TABLE GENERATE is located on "S11" This image was captured from a test page.

1. Ramp tables will be built to span the maximun FT duration (~40 seconds)

2. The power supply parameter (scale factor) with the "P" suffix (i.e. S:V100P, S:MLAM1P) should be considered the "POST" or "2nd" ramp and should always be zero.

3. Ramps will be started on event <31> as before.

4. "Switch to this scale factor" will be event <46> (the End of Flat top event)

5. The delay time in this switch will be set to its minimum value (4 ms)

SY Power Supply waveforms will then track the end of FT. (External Beamline Power Supplies do this already.)

Parameter page parameters should respond as in previous runs.

Other stuff


This info is most likely in the rest of the help files, but it doesn't hurt to have some of it here also.

The 165 cards can have up to 10 waveforms stored onboard. Only 2 are defined in the database for saveing and restoring. We seem to really only need one (1) anyway! So where there's a "Read table" or "Load table" and a "1->2" , this means choose either 1 or 2.


 

Here's how to read the table in the picture.


START LEVEL ( 0) The starting level of the waveform is zero times the scale factor
02 ( 4.8 ) ( 0 ) Take 4.8 seconds to go to the value of zero. (in this case, stays the same.)
03 ( 1 ) ( 1 ) Take 1 second to go to the value of one times the scale factor. (the scale factor is the D/A)
04 ( 43.3 ) ( 1 ) Take 43.3 seconds to go to the value of one times the scale factor. (in this case it stays at the full scale factor value)
05 ( .005 ) ( 0 ) Take 5 milliseconds to go to zero.
If event 46 occurs at any time during the ramp, the card switches to the 'P' suffix device. (should be zero)

 

 


Limitation
There is a limitation in the 165 card that apparently has been there since day one in 165 card history. The 165 card uses signed arithmetic in calculating each table segment. There is a problem if the segment is longer than 32.7 seconds and the segment is not flat (has a desired slope). The 165 card was designed with a slope parameter but it wasn't used 'cause it was easier to change the table break points. Devices like S:H90 and S:HT202 have in the past had slopes to help correct for beam roll due to momentum change during spill. With a longer flattop for this run, we MUST use the flattop slope parameter for devices like S:H90 and S:HT202 if slopes are needed. (Slope parameters have the suffix "Q" for the first ramp and "S" for the second {not used}) - Wally
Saves are Broken

Unfortunatly, S11 is hard coded with the 165 parameter/sibling relationship. The S11 program (PA0185) needs to be changed in order to handle this database change. I've decided to NOT try to fix the S11 save problem, but instead use some of the generic features of the control system to avoid problems with saving and restoring. All the SY 165's are saved during an ARCHIVE (big) and GLOBAL save. Restoring should be easy.

Restore S:165ALL from D1 to restore all the SY 165's

- Wally