Reports 1-1 of 1 Clear search Modify search
fabian.arellano - 21:26 Tuesday 15 December 2020 (15651) Print this report
Ballast mass adjustment

With Washimi-san.

Preparation and system condition:

  • Temperature: 23.6 degrees.
  • The polarity of the F1 fishig rod is such that a negative number of steps moves the keystone up.
  • IP and F0 are locked.
  • I moved the BF keystone from 300 um to approximately zero using the fishing rod.
  • I checked the F1 fishing rod was at the lowest limit of its range at approximately 557 um, then I moved the carriage up to midrange by 852,000 steps:
    • At lowest point: the carriage is at 5,515 stepsandthe keystone at 557 um.
    • At highest point: the carriage is at -846,390 steps; the keystone is at 978 um.
  • The displacement due to buyancy is 212 um; this means we aim to set the keystone to such a height in air.
  • The position of the system before adding ballast mass can be seen in the attached picture.


We added 66 grams in opposite sides of the BF. We had to leave the tunnel  because of too much  snow outside and there was no chance to assess the result.

I'll add another entry tomorrow.

Images attached to this report
Comments to this report:
fabian.arellano - 19:20 Thursday 17 December 2020 (15658) Print this report

Adding ballast mass to the top of the BF sounds like a very easy task. However, the challenge is to add mass and, at the same time, keep the existing balance, in pitch and roll, of the BF. Becuase the IRM with the OSEMs hangs from the BF base with three thick (2 mm) maraging steel rods, any change in the orientation of the BF would affect the OSEM alignment. In the worse case the OSEM bodies would touch the OSEM flags.

On Tuesday afternoon we added ballast mass on opposite sides of the BF top:

  • We put 66 grams at each side.
  • At the -Y side we put mass in an unused space and the 66 grams includes the mass sof the screw.
  • At the +Y side we had to put the ballast mass on top of other masses and we had to change the existing screw by a longer one:
    • New ballast mass (66 g) + old screw M6 L35 mm (9 g ) = 75 g
    • New ballast mass (63 6) + new screw M6 L50 mm (12 g) = 75 g
  • We fastened the screw by hand because we required to test the result before commiting.
  • We released the BF, which had been locked to the side security structure. 
  • Right after putting the masses we had to leave the site due to heavy snow outside.

Today I measured transfer functions and also I compared the new position with the old one. The results are good:

  • The IM transfer functions show a heathly system. I didn't measured the one for IM-V because F0 is locked.
  • The addition of the ballast masses to the BF changed the orientation of the IRM slightly but this does not create any problems. The largest change was -42 µrad in IM-P. See the atteched picture.
  • The changes in the readout of the IM OSEMs are due to a change in the orientatio of the IRM, not of the IM, which remains in place.
  • The only chnage in the oplev QPD readout that is significant is in TILT QPD H, which indicates the optic moved in yaw. I'm not sure yet why this happened.
  • Current temperature: 23.6 degrees.
  • F1 keystone moved down by -560 µm to 423 µm. Because the goal is 212 µm, we will need to add more mass later once the temperature is suitable.
Difference 15 µm -10 µm 8 µm -22 µrad -41 µrad 24 µrad 3 ndu 5 ndu 12 ndu 89 ndu -10 µm -560 µm
  • The acronym ndu means normalized displacement unit [-100,100].
  • The DTT files with the transfer function measurements are in TypeBData/TF/Measurements/20201217/

As a conclusion we can say that it's not so difficult to add ballast mass on top of the BF without disturbing the alignment of the IRM. However, I should check whether there is an obvious reason for the yaw of the mirror to change.

Images attached to this comment
fabian.arellano - 14:23 Saturday 19 December 2020 (15665) Print this report

Displacement in yaw, measured by the oplev, after a GAS vertical displacement, has been observed many times before and it's common. This type of coupling is very likely produced by cables. 

Search Help