aLIGO LHO Logbook

H1 SEI (ISC)

stefan.ballmer@LIGO.ORG - posted 08:03, Thursday 27 February 2014 - last comment - 14:35, Thursday 27 February 2014(10370)

Question to the ISI team

I just spent another hour turning on ISI ETMX. (see previous elog for ITMX.)

This seems to me that most of the turn-on trouble comes from our attempt to turn on loops in an equilibrium position, and afterwards move to the target position. This move then leads to DC saturations in various sensors.

Question:
Why don't we put in fixed drive offsets - even after the DACkill - that allow us to make the OFF position equal to the locked position. This way the ISI never has to "move", we never have T240 saturations.

Comments related to this report

brian.lantz@LIGO.ORG - 10:22, Thursday 27 February 2014 (10376)

Link

When the watchdog turns things off, they really need to be off. 
Last time a fiber was broken, it was because there was a bogus signal coming from the IOP process.

That said, We agree that it would be really good if turning the ISI on/off didn't mess with the alignment so much.

One clear action item is to trip the watchdog less often - all suggestions on this are welcome. 

Other Suggestions - 
1) Hold most offsets with HPI - we have in mind a way to hold HPI offsets longer during watchdog trips.

2) Stop going back to the target locations for most DOFs on the ISI. One can not see how recovering the old X, Y, Z, and roll DOF (e.g. if the beam direction is along X, the rX direction seems not-very-important at the micro-radian scale).

Clearly yaw is important for all optics, and pitch is important for tables with TMS. 

3) get the guardian for BSC-ISI up so we can turn on ISI whilst ignoring the T240s - in the mean time, this will be put into the BSCISItool scripts available in the next few days (testing is ongoing)

stefan.ballmer@LIGO.ORG - 14:35, Thursday 27 February 2014 (10386)

Link

>When the watchdog turns things off, they really need to be off. 
>Last time a fiber was broken, it was because there was a bogus signal coming from the IOP process.

I don't understand. A constant value is just as "off" as the value 0. Not energy will be transferred to the payload.