Issues observed by PCT over the weekend (from H. Sinn entry in the PCT log)

- The interlock of the vacuum valves worked as expected. However, we had two times the gas section around the XGM tripping the beam, because the gate valves around the XGM were closing. The reason for these trips will be investigated (probably noise on one of the involved pressure gauges). In the future it would be good to change the Vacuum-to-MPS interlock such that V0 always closes, if any of the other gate vales closes. Then the accelerator permission would come automatically come back after the trip. However re-booting of the vacuum loop needs to be planned (trips the DESY beam permission and potentially shuts down the turbo pumps). 

- The SR Imager may loose steps when moving larger distances (speed too high?)

- Several general problems of Karabo were observed, which make live difficult: For example, the 'V0 closed' and 'B4C in' indication on our main screen may just lock, even when the state of the device is changing (was observed mostly on GUI server1). It means that the status displayed on a scene can be in general not be trusted. In many cases we observed that connecting freshly opened scenes to devices take unacceptably long times (many minutes or never).  The policy of saving projects and sub-projects leads to overwriting changes, if more people work on projects and projects stay open for longer time.  

- The fact that the PLC does actually not store values through a reboot causes many problems. It basically means that motor positions are lost (set to zero) at every reboot and can be only recovered, if homing or absolute encoders exist. Also that positions of motors cannot be reset to their theoretical values means that in practice the state before the PLC reboot cannot be recovered by resetting the values.

- The imagers, in particular the SR-imager, caused a lot of problems because they were often crashing and require advanced and complicated procedures to make them run again (e.g. setting after re-instantiating the imagers are lost and come up with wrong pre-sets). The SR-imager does auto-scaling on the intensity, also when this is disabled in the software. This make the imager only useful for qualitative studies (e.g. not for gain curve or flux measurements).

General

• Parameters of motors get lost after reboot of the PLC
• Setting trigger delays can currently only be done with special authorisation from DOOCS. 
• Suggested solution from AE: Hardwire the correct defaults in the firmware
• A better procedure is needed to communicate the parameters between hardware experts and AE
• L. Samoylova suggests that CAS person on call contacts BKR once per day to establish contact and get an overall impression
• The support of CAS during the weekend was greatly appreciated

Vacuum

• No calls to experts over the weekend
• Issue with the XGM vacuum (spurious triggers of the interlock) needs to be fixed 
• Interlock logic should be changed so that V0 always closes if one of the gate valves closes

Imagers

• SR Imager crashed frequently and was hard to bring back
• Auto-scaling of SR-imager cannot be switched off. This needs fixing with high priority

XGM

• Commissioning starts today
• See above about vacuum issues

PLC

• No calls over the weekend to the experts
• Firmware for MC2 lib (TECHNOSOFT motors) is prepared and waiting for tests

Karabo

• CAS was called a few times on the weekend
• Running more than one camera at the same time appears  to cause crashes. Possibly network bandwidth is insufficient
• A procedure can be set up to read out the motor parameters at a well defined point in time for storing them in the firmware 
• Setting trigger delays should be made to work from Karabo

Work for next days

• Karabo test with MC2 Lib in the lab. As soon as this is successful it will continue in XTD9. Will be done on the SPB attenuator as N. Gerassimova is not available for PBLM tests due to a beam time
• Fix the Photonics camera driver so that auto-scaling can be turned off