KAGRA Logbook

VIS (MCF)

tomotada.akutsu - 5:02 Thursday 22 April 2021 (16537)

MCo suspension strange

R. Takahashi, Miyo, K. Tanaka, Akutsu

Abstract

As there are several known strange behaviors for the MCo suspension, Takahashi-san revisited the MCo suspension and tweaked the IM-RM, but no improvement in terms of these behaviors. Strange.

Detail

The MCo suspension has several strange behaviors:

The picomotor sometimes mixes the DoF (but this might be the same for the other IMC suspensions) by observing with oplev.
The picomotor sometimes refuses to move.
Even closing the chamber flange at +X with an aluminum foil tilted the MCo in yaw by 100-150 urad! The sign of the tilt sometimes changes.
Miyo-kun found a strange feature in the TFs from each one of the coil-magnet actuators (H1, 2, 3, orr 4) on MCo to oplev YAW.

So Takahash-san and I revisited the MCo suspension this evening, and that was a beginning of a nightmare.

First, Takahashi-san found the IM was too close to IM-RM in the -X direction, so he tweaked the IM-RM in a brute way (but this is the regular way according to him... I was a bit shoked); he bended the supporting bars for the IM-RM with a monkey wrench so that the IM-RM was re-balanced in a nice atitude. When he finished, I found the oplve beam was totally out of the QPD in yaw (sum is almost zero) and illuminated one of the electric pin of the QPD! I immediately tried to move the yaw picomotor, and the beam spot could move in the first place, but for some reason it won't move any more when it reached at a certain point.

Then Takahashi-san again saw the IM and IM-RM were again very close in the +X direction. He did the same way to relocate the IM-RM. Then I found the oplev beam moved much in the opposite direction! Anyway, again, the beam spot illuminated the other edge of the QPD (but this time the sum is large), and so I again tried to move the spot by picomotor in yaw. Finally it came back to the center of the QPD, although there were issues as already listed above... the picomotor in yaw suddenly become irresponsible or moved in very small amount... Anyway, fortunately, the IMC transmission beam profile also recovered.

Comments to this report:

kouseki.miyo - 7:44 Thursday 22 April 2021 (16538)

I found Stefan's posts about MCo strange behavior. See #5110.
They also faced on the same coupling problem,

satoshi.tanioka - 7:53 Thursday 22 April 2021 (16539)

Just for information

We modified Type-C suspensions for IMMTs about three years ago (klog 5191).
This is because that screws which clamped mirror suspension wires have magnetism and sometimes IMs stuck to the damping magnets.

I don't know whether the clamping screws of MCo are facing the same problem though...

yoichi.aso - 13:56 Thursday 22 April 2021 (16543)

K. Tanaka, Y. Aso

To continue the investigation of the strange MCo behavior, we did the following checks this morning.

Response to picomotor

We moved MCo in Yaw by using the picomotor to see if the OpLev responds accordingly.
On the very first command to the picomotor of the day, the direction of the OpLev change was in the wrong direction.
However, after that, the response of the OpLev to the picomotor actuation was normal. There was no jump nor flip of direction.

Jump of Yaw after a large shake

Another problem observed before was that MCo jumped in Yaw by about 100urad when people pushed the breadboard of MCF.
In this morning, we tested this by intentionally pushing the breadboard.
We did observe the jump of tens of urad. However, it only happened when the OpLev was more than +40urad in Yaw.
If we move the OpLev below this value, using the picomotor, the jump does not happen.
This made us suspect that the IM may still be very close to touching the IRM and depending on the Yaw position, IM actually touches IRM.

Inspection of IM

We went into the clean booth and inspected the IM of the MCo suspension.
It seemed that the IM was almost touching the IRM. So I bent the supporting rods of the IRM to make more room.

After the above work, we did the shaking test again. This time, we did not see any jump even when OpLev is at above +40urad.

So we concluded that the jump was caused by the friction of the IM with IRM.