The following report has been submitted to the On-call interface.

On-call events -> Injection System

Title: SL cristal controller failure

Author(s): gosselin, gherardini

Details:

This morning I have been called by Fabio because the slave laser switched off. After investigation it seemed that the culprit was one of the temperature controller.
I came on site and went in EER with Fabio. It appeared that the controller for the crystal 1 was indeed not working. We started to swap it with a spare one. Setting the good parameters is not an easy task. The procedure is explained in the log entry 62780.
Some additional info about this procedure for the next time it occurs :

- To access the first menu one need to switch the PID ON and the 5 first seconds push simultaneously the two buttons on the right.
- When in the first or second menu, DO NOT push simultaneously the two buttons on the right. It makes the PID enter in a default mode with the display not properly working. To access the third menu it is better to switch it off, wait more than 10 s and then push on the simultaneously the two buttons on the right.
- LOC = 10 to deblock
- Not sure about the adress parameter but should be 1 2 3 or 27.

PID :
#Z131010... 008... => faulty one that was installed
#562 => display in default mode.
#560 => Installed now

When we finished with the PID controllers and checked that they were correctly working, we noticed that even if they were sending corrections the temperature of the pumping diodes and the crystals were not changing.
We suspected that the power supply of the peltiers was not working properly. With the help of Walid from remote we did some test and finally noticed that it was off... When arriving in the morning it was ON and working properly. Maybe the security chain somehow switched it off at some point of our intervention.

With the power supply ON again I started to increase the pumping diodes of the SL but the security triggered again. Most likely because the temperature of crystal reached 25 degrees. There might be an alarm not correctly set in the crystal 2 PID. I set the set point to 24 degrees (as it was this morning) and managed to recover the SL at full power.

I also recovered the neovan and slightly after the PMC relocked. We had to wait about 2 hours for everything to thermalized before going on with the lock acquisition (see plot).

To be noticed that, with the same parameters, the power at the output of the neovan is lower than this morning. 0.345 instead of 0.36 (see plot 2). This would need to be investigated later.

* Note that any files attached to this report are available in the On-call interface.