Today we worked on the remaining part of the implementation of the Autorelock feature; what was missing was dealing correctly with the self-parking of the PR mirror due to the tripping of the protection for SDB1; we made some changes to the code of the nodes ITF_LOCK and DRMI_LOCK and the results are the following, although obtained with low statistics:
- the automation is not stuck anymore in case of PRM tripping and the full interferometer going to DOWN (i.e. in any state higher then LOCKED_CARM_OFFSET_MID);
- if only the DRMI unlocks but the arms stay locked on the green with 3kHz of CARM offset the automation should not get stuck anymore, but as a drawback PRM and SRM won't be kept aligned before trying again and they will get misaligned and realigned again, which is safe but time-consuming and ultimately annoying; another test is being studied in order to overcome this.
In any case, even if some more statistics would be beneficial, I guess we can try to make a first autorelock attempt during the night.