Controls Meeting 11-Aug-2011

From GlueXWiki
Jump to: navigation, search

Agenda

  • Announcements
  • Indiana University HV system
  1. Features of the existing software/hardware system.
  2. A proposal for EPICS interface.
  3. Which parameters needs to be controlled and monitored from EPICS?
  • AOT


Time/Location

9:30 AM Thurs 11-Aug-2011 CC F326 (Call regular Hall D ESNET # 8542553)


Minutes

The main (and only) topic of the meeting was the IU HV system control.

Hovanes presented a preliminary proposal describing the basics of the EPICS support for FCAL HV controls. According to his proposal this would be a software IOC based system, and we would need to write the EPICS support for it. The driver part of software would scan the bus every fixed time interval and retrieve the parameter values and keep them in a buffer on a host computer. There were a lot of useful discussions with many suggestions to Hovanes. At 10:00am IU group had to leave to join another meeting.

One of the issues that was mentioned by multiple people is the issue with the frequency that the HV parameters can be monitored. With the current speed of readout the full set of parameters for 100 bases would be read in about 5 to 10 seconds (an estimate by Dan). The dead time (350 ms per parameter according to Hovanes) is dominated by the time the base prepares the response.

  • According to Hovanes, other HV systems (CAEN in Hall D and Hall B, Lecroy in Hal B) are monitored at a rate of ~1Hz.
  • Matt mentioned that such a rate may not be a major problem for FCAL, the original plan was to set to voltages to the bases and set it to a lower power mode and read them out once a minute or so. It all depends on the stability of the system.
  • Matt also suggested that there can be separate buffers for reading and writing. Read would have a priority.
  • Dave suggested that there needs to be some sort of priority for parameters that have been set "recently". The regular synchronization scan gets interrupted and the modified parameters are now monitored with "a higher priority". After "some time" the regular scanning resumes.
  • There might be a possibility to reduce the response time if there are multiple sampling/averaging of ADCs. IU will check if such a possibility exists.

Matt said that the hardware of the bases is fixed, but firmware can be modified. JLab folks needs to send specific request for adding features that can be accommodated in firmware.

  • Elton explicitly asked for trip current control.
  • Beni and Elton were advocating for including a maximum voltage setting in the firmware for protection.
  • Matt made it clear that the ramp rate cannot be changed.


The participants reached an understanding at the last minutes of the meeting that the JLab controls group will develop the EPICS interface with support from IU. IU will take care of all the features on the bases.