Controls Meeting 5-Apr-2012
- Minutes from previous meeting 22-Mar-2012
- Security and access from outside - Elliott, Hovanes
- PXI system and channel counts - Yi, Mark, Elliott
- EPICS - Hovanes, Vanik
- Alarm system, PV naming scheme, stepper motors, etc.
- New systems - Hovanes, Elliott
- Goniometer, UConn boards, other power supplies, etc.
- JInventory database - Elliott
- Counting house installation - Hovanes, Mark
- Equipment orders - Mark, Hovanes
- Solenoid_Controls_Redesign - Mark, Josh, Scot, Yi, Elliott
- UPS power to Danfysik, coil taps and PXI, SOE, QD, shunt replacement, ground fault detector, etc.
- Test stand - Hovanes, Josh
- PLC programming - Josh
- Elogs and databases - Elliott
- Midas elog
- Operator elog
- Experiment Control Supervisor - Elliott
- Mantis task tracker
9:30 AM Thurs 5-Apr-2012 CC L207
9:30am 19-Apr-2012 CC L207
Present: Elliott W, Hovanes E, Vanik K, Scot S, Josh B, Mark S.
Andy Kowalski from the Computer Center told us at the Online meeting that all read/write access to computers in the counting house must be via two-factor authentication, i.e. to log in to those computers from home or your office you'll need to have one of those devices that gives you a number when you press a button. That number plus your password constitutes the two factors. The devices cost arount $100 each. Note that read-only access to web pages, etc. will not require two-factor authentication.
Yi ordered the PXI fast/slow DAQ system based on channel counts supplied by Mark and others. With luck the system will be here in a few weeks.
- Hovanes is testing the latest EPICS CSS-based alarm system. He plans to use it in the BCAL test.
- We had a brief discussion concerning PV naming schemes and the relation to PLC tag names. Some characters are forbidden in one or the other.
- Josh can auto-generate a complete list of tag names that Hovanes can use to create the EPICS database. Tags that shouldn't be exposed as PV's will be identified by a suitable naming scheme that Hovanes' scripts can key on.
- Vanik continues to work on stepper motor control. Most likely the Goniometer will use the controller he's working on. The driver came from Mark Rivers (BNL?).
- It would be useful to get some stages so that Vanik can test the driver and board more thoroughly.
- We'll have five magnets and power supplies, and four of them are coming with Danfysik supplies. It seems likely we'll get a Danfysik for the Tagger as well.
- With luck Josh's revised code for controlling the solenoid Danfysik supply will work for the others.
- We'll have additional motion control applications, including target sticks, harps, collimator stages, etc.
- We hope to turn some EPICS applications completely over to Accelerator, e.g. the tagger magnet, as they will be controlling these devices. We'll remain responsible for maintenance in most cases.
The JInventory system is running in production mode. Elliott plans minor improvements in the gui. CLAS is working on a major upgrade that includes a cable database. Currently the basic buildings and rooms are entered so it's ready for the real stuff.
Harris has the contract for running the UPS power from the Utility Room to the Rack Room. Tom move the UPS into the Counting House but it is not in its final location yet due to the electrical work. With luck the UPS will be working in a week or two. After that we can run the under-floor wiring and install the network rack and switch.
The group meets every week. George has made preliminary decisions concerning the FMEA. We'll install a UPS on the Danfysik control boards and contactor, and the PLC is adequate as a backup quench detector. Otherwise we will not make an major changes to the system. Josh is working on revising the PLC program and has had great success using Structured Text programming to replace Ladder-logic for string manipulations. Much work remains to be done!
At this point we got booted out of the room...