We used 4 channels
K1:PEM_MIC-OMC_TABLE_AS_Z
K1:PEM_MIC-BS_TABLE_POS_Z
K1:PEM_MIC-BS_BOOTH_BS_Z
K1:PEM_MIC-SR_BOOTH_SR_Z
to calibrate the microphone of Virgo in this morning.
You can't acquire the appropriate data of these channels in this morning.
sorry.
Lucia kindly helped to set up the diagonalization matrix for SR2 geophones and implemented blending filters (which I will attempt to optimize later). I designed new damping filters and placed inside the IDAMP filter block. With inertial damping, we get to control at higher frequencies without introducing too much noise. In attachment, please see a comparision between undamped (black), normal damping with LVDT signals (yellow) and inertial damping (blue).
Here are the fitting results of three ring-down measurement when 33K on Friday:
- GPS 1242116740 Finesse 121
- GPS 1242116820 Finesse 129
- GPS 1242116800 Finesse124
Stefan's simple fitting script was used.
Report for last Friday 5/17
Conclusion: Thermal paste is of little use, but low current seems good enough to suppress solenoid heat.
the thermal conductivity of thermal paste is 16 W·m−1·K−1
the thermal conductivity of steel is around 46 W·m−1·K−1
Because last Thursday we loose the spring of B2 solenoid and use the thermal paste at the same time, the experiment condition is not good, so we recover the spring of B2 and test agaiin.
B2 with thermal paste,see Fig,1 :
During this test, first we increase the current to 240mA/18V, then down to 110mA.
The start temperature is 29.1 deg (time: 12:25), end temperature is 31.7 deg (time: 14:32) . The temperature just rise 3 deg, so we test A2 solenoid to confirm it.
From here we fix the FLIR camera on the top.
A2 without thermal paste,see Fig.2: start current 210mA/15.7V, then down to 70mA/5.32
start point: time:15:30, 27.1 deg.
end point: time: 16:47, 27.4 deg (16:30, 27.7deg).
Then we increase the current up to 280mA/22V for 30mins (similar as the shutter drive.16:47-17:18), the temperature up to 42.1 deg, (the temperature is 37.8 when time is 16:56, 9mins after increase current.Fig.6).
After that we down the current to 70mA/5V, it takes 45min to cool down to 27.7 deg (17:18-18:04). see Fig.3.
A2 wit thermal paste,see Fig.4: start current 210mA/16V, then down to 70mA/5.2V.
start point: time:18:50, 26.2 deg.
end point: time: 19:52, 26.5 deg.
Then we increase the current up to 280mA/22V for 9mins (similar as the shutter drive,19:52-20:08), the temperature up to 38.8 deg.see Fig.5.
Low current test (keep open at 70mA)
| A2/1hour |
without glue (15:30-16:30) |
with glue (18:50-19:52) |
| start temperature | 27.1 | 26.2 |
| end temperature | 27.7 | 26.5 |
glue is thermal paste, temperature is deg C.
High current to imitate shutter driver (keep open at 280mA)
|
A2/9mins high current |
without glue (16:47-16:56) |
with glue (19:52-20:01) |
| start temperature | 27.7 | 26.5 |
| end temperature | 37.8 |
38.8 |
So the thermal paste is almost useless, might because its thermal conductivity is much smaller than shutter base (it is stainlesssteel). But the heat at low current is much lower than the heat at current provided by the shutter driver.
If we can make a new driver which can provide low current, it seems worthy to test the shutter in PSL.
Just now, I tried to lock the Yarm, I found the pitch control of the EY suspension has some strange behavior.
When I tried to move to plus pitch direction (monitored by the TM oplev) by moving mass,
TM didn’t move from certain point.
Could someone check it immediately?
Yokozawa
I measured the Yarm finesse by fitting the rindwon.
ITMY TM : 25K
ETMY TM : 51K
maximum count of the TMSY IR PD : ~650count
finesse : ~146
I found large shaking of the ETMY TM.
That was started from 7 p.m. yesterday.
(If my memory was correct, this oscillation is caused by some control loop)
Could someone tell me how to damp this mode? (Or just I have to wait?)
Yesterday, Kazuhiro Yamamoto took 2nd loop Servo from KAGRA mine to Toyama University in order to fix the connection of D-sub connection.
I will return it to the KAGRA mine next week.
[Keiko, Takaaki, Stefan, Masayuki]
The ETMY started to be warmed up today. We did ring-down measurement again, and found that the finesse got increased even at 33 K. The estimated finesse is 120, corresponds to the mirror loss of 1.1 %. We did several measurement and all of the fittings shows consistent value.
The low finesse might be because of the contamination on the mirror...??
We made a first incarnation of an LSC_LOCK guardian - at this point mostly as a way to record the settings that worked.
Until we link it somewhere, it can be opened by typing
guardmedm LSC_LOCK
So far it has four simple states: DOWN, IR_ARM_LOCK, ALS_X_SETUP and ALS_X_IR_FEEDBACK.
DISCLAIMER: So far the guardian does NOT do any of the following:
- error checking (i.e. is the IMC locked?, did we loose the arm and need to go back to DOWN? etc)
- aligning the optics into the desired position
For now this has to be done by hand.
Attached plot is long term trend of the IR (right) and GR (left) transmitted power with the the temperature of ETMY (upper). The transmitted IR power was around 3000 and the GR power was 1500 about one month ago, and seems to get smaller with the ETMY temperature.
Mmm, does it indicate anything?
[yutaro, masayuki]
The attached picture is four resonant peaks with free swinging cavity. I picked at random, and they don't seem to have double peaks.
[Masayuki, Yutaro]
This is the result of cavity scan of Y arm for green.
From this, finess is estimated to be 17, while the design is something like 50.
Today we test B2 solenoid with shutter base and thermal paste,see Fig.1.
First we test the B2 solenoid without heat sink, we rise the current up to 240mA/18V to move the solenoid, then decrease the current down to 110mA/8.26V
The start temperature is 27.7 deg C (13:29), 90 mins later, the temperature rise to 34.6 (15:13),see Fig.2, this temperature is lower than yesterday's test which is without shutter base, the temperature of the other side of the solenoid is 36.9 deg C, see Fig.3
Then we put the thermal paste on the surface between the soleniod and solenoid base, and the surface between solenoid base and shutter base, in this process, we found that the spring of the solenoid can be adjust manually, see FIg.4, we loose the spring of A1 and test the current to move it, the current is about 200mA, and the minimum current is around 60mA, it is lower than before, actually it make sense, the needed force is smaller because the spring is loosed. Then we loose the spring of solenoid B2, it turn out that the it only need 180mA/13/5V to move the solenoid and the minimum current to keep it open is down to 85mA/6.38V, it was 240mA/110mA.
Finally we the the solenoid at lower minimum current (85mA/6.38V).
The start temperature is 31.2deg C (16:03), 2 hours later, the temperature rise to 34.1 (17:58),see Fig.5, The temperature increase by only 3 deg, it might because of the lower current. And the hot part is on the bottom of the solenoid, the top part is not so hot, we believe that this is due to thermal paste.
The temperature of the other side of the solenoid is differ as the position of the camera change see Fig.6-8,
Fig.6-36.5 deg C,
Fig.7-34.9 deg C,
Fig.6-31.8 deg C,
Next we need figure out the calibration of the FLIR camera.
After aligning the IR to both arms, we did a arm finesse ring-down measurement. The fit uses the nominal ITM transmissions (0.44% ITMX, 0.48% ITMY).
X-arm:
======
Finesse = 1399
Power build-up factor = 891
Ring-down half-life = 3.11msec
REFL POWER locked/unlocked = 0.96
Y-arm:
======
Finesse = 28
Power build-up factor = 18
Ring-down half-life = 0.42msec
REFL POWER locked/unlocked = 0.50 (agrees with experiment)
We also tried to move the alignment of the y-arm around a bit, in the hope that we could find a better spot. There was only very little dependence on spot position.
Finally, we also did a beam scan of the y-arm, confirming a 30-ish finesse. That picture is also attached. Note the wide, almost double-peaked resonance structure.
Attached are plots of the ring-down data (blue) and fit (orange), as well as the plotting function and raw data.
[Masayuki, Yutaro]
This is the result of cavity scan of Y arm for green.
From this, finess is estimated to be 17, while the design is something like 50.
[yutaro, masayuki]
The attached picture is four resonant peaks with free swinging cavity. I picked at random, and they don't seem to have double peaks.
Attached plot is long term trend of the IR (right) and GR (left) transmitted power with the the temperature of ETMY (upper). The transmitted IR power was around 3000 and the GR power was 1500 about one month ago, and seems to get smaller with the ETMY temperature.
Mmm, does it indicate anything?
[Keiko, Takaaki, Stefan, Masayuki]
The ETMY started to be warmed up today. We did ring-down measurement again, and found that the finesse got increased even at 33 K. The estimated finesse is 120, corresponds to the mirror loss of 1.1 %. We did several measurement and all of the fittings shows consistent value.
The low finesse might be because of the contamination on the mirror...??
Here are the fitting results of three ring-down measurement when 33K on Friday:
- GPS 1242116740 Finesse 121
- GPS 1242116820 Finesse 129
- GPS 1242116800 Finesse124
Stefan's simple fitting script was used.
= I am happy with the current beam spot positions on both ITMs =
[Stefan, Yutaro]
As entitled. After the height adjustment work by VIS team, we aligned the main beam to X arm and Y arm.
The attachment shows the spot position on ITMX (1st) and ITMY (2nd). I would say it's acceptable for now.
(3rd one is ITMX spot with green)
With Mark,
After 8885, from the MEDM screen, we discovered that the optical lever signals for SR3, SRM and the BS are not fluctuating with amplitude as high as that in SR2 length optical lever. So, we decided to measure the residual motion of SR3 as a comparision.
Figure 1 and figure 2 show the residual displacement and the residual velocity with all controls on except TM damp. From the graphs, we can see that the residual displacements of longitudinal, pitch and yaw are all below 1 rms, while the residual velocities are slightly above 1, which means we are very close to the 1 µm/s requirement (if this is the correct requirement). I haven't had the chance to measure the residual motion with the TM damped.
More on last night's TM feed-back work:
=======================================
- We were able to increase the DAMR1 gain to -60000, which led to the reported suppression of some length peaks.
- For this we had a simple control filter in DARM1 FM8:p40^2,z1 : zpk([1],[40;40],1,"n")
- For the MN stage, we initially used a simple 1/f integrator and notch at .63Hz. That worked fine for a very low frequency bleed-off.
But naturally we tried to increase the gain to the MN stage.
- This rang up the fundamental Pendulum resonance at 0.63Hz, so we started designing the plant compensation filter. Details are below.
- We successively tested the first two pole compensations, i.e. with the 0.63Hz compensation, the 1.53Hz rang up, with the 0.63Hz and 1.53Hz compensations, the 2.49Hz rang up.
- We haven't tested the full compensation filter yet.
- given the almost 90dB of drive enhancement needed for the MN drive, we will also try the IM stage as an intermediate actuator.
- Filter design strategy
- We know the MN to TM feed-back has three main resonances. We found them by increasing the MN stage gain and ringing them up:
0.63Hz, 1.53Hz, 2.49Hz
- In other words, the MN->TM transfer functions has 3 high-Q pole pairs at those frequencies (and no zeros, since it has to be 1/f^6 at high frequencies).
- Since these resonances have high, not well known Q, there is no chance of exactly compensating them. That's especially true since additional damping will change the effective Q.
- In addition, high-Q digital filters are never a good idea from an operational point of view.
- So we need to compensate these high-Q poles with lower-Q zeros. But putting the compensating low-Q zero-pair at the same frequency as the pole results in a huge phase-drop just above the pole frequency.
- Therefore we need to move the compensating low-Q zeros to a slightly lower frequency, effectively generating a complex pole-zero lead filter, recovering some of the phase.
- However we can't be too aggressive with that strategy, as the multiple lead filters will increase the gain above resonances.
- This design strategy leads to filters that are relatively robust against changes in the Q of the mechanical plant.
- The attached bode plot of the compensation filter illustrates that.
Filter details:
- Fundamental (0th) Pendulum (0.63Hz) inversion:
zero pair: z=0.55Hz, Q=5 (pendulum compensation); pole pair: z=4Hz, Q=3 (roll-off)
zpk([0.055+i*0.547243;0.055-i*0.547243],[0.666667+i*3.94405;0.666667-i*3.94405],1,"n")
- 1st Pendulum (1.53Hz) inversion:
zero pair: z=1.46Hz, Q=5 (pendulum compensation); pole pair: z=4Hz, Q=3 (roll-off)
zpk([0.146+i*1.45268;0.146-i*1.45268],[0.666667+i*3.94405;0.666667-i*3.94405],1,"n")
- 2nd Pendulum (2.49Hz) inversion:
zero pair: z=2.31Hz, Q=5 (pendulum compensation); pole pair: z=5Hz, Q=4 (roll-off)
zpk([0.231+i*2.29842;0.231-i*2.29842],[0.625+i*4.96078;0.625-i*4.96078],1,"n")
With Mark,
A measurement of the residual motion SR2 optic shows that it has ~1 µm or urad of longitudinal/tilt rms displacement and ~10 µm/s rms velocity in longitudinal, ~10 µrad/s rms angular velocity in pitch and ~1 µrad/s rms angular velocity in yaw. The measurement was across 0 - 15Hz and was done with all local controls on. Since the oplev is known to be noisy, I turned off the TM damping control and did the same measurement again and it didn't seem to work and in fact there is no change in the result at all. I came to conclusion that the optical lever signals are just too noisy because the measured rms velocity drastically reduced to ~1 µm/s if the measurement was only across 0 - 1 Hz (the pendulum mode is at <1 Hz so I expect any motion in L should happen at below 1 Hz). Same measurement was done for the BS and the result is about the same.
To rule out electronic noise we measure the QPD vertical and horizontal signal before the division by the QPD sum. In figure 1, PIT_IN1 is the vertical channel, YAW_IN1 is the horizontal channel and the SUM_IN1 is the sum. If the noise is electronic noise, we should expect the noise level in both horizontal channel and vertical channel to be at around the same level. But, as shown in figure 1, the horizontal and vertical noise level is clearly separated from the sum as if there the beamspot has real horizontal and vertical displacements. We tweaked the gain of the whittening filter and we also tried enabling the built-in gain of 5 of the QPD circuit board and no obvious improvements were observed, as in, there is still a huge gap between the horizontal/vertical noise level and the sum noise level. Then, we blocked the beam from entering the lower viewport and did the same measurement again. This time the horizontal and vertical noise level became similar to that of the sum noise level (see figure 2).
Today, the air conditioners and HEPA fans were turned off for a short window for a separate measurement so we used this opportunity to do the same measurement again. Figure 3 shows the same measurement, with dotted lines as old measurements. As can be seen from the figure, the noise levels have obvious improvements at ~>1 Hz and it actually reveals some mechanical resonances at >10 Hz. So, I decided to measure the residual motion under this circumstance. Figure 4 and figure 5 show the residual rms displacement and rms velocity up to 5 Hz. dotted lines are the measurements with fans off while solid lines are the measurements with fans back on. As shown in the figure, turning off the fans drastically improve the longitudinal and pitch noise levels (But mysteriously worsen yaw).
However, the noise level is still too high for us to measure the real residual motion of the optic. We can only make a pessimistic conclusion about the residual motion. In terms of displacement, we can say that the residual displacement/angular displacements are < 1 µm or µrad. But, we can't say for sure that the suspension meets the type B residual velocity requirement. The best result we get is ~3 µm/s and I am quite sure that a huge portion of this is noise.
In the long term, we need to suppress the air flow around the optical levels so to reduce the displacement noise induced by the fans. There is also a possibility that the noise is caused by beam jitters. But, we still need further investigations to confirm.
As far as I know, the optical lever signal is also highly susceptible to any force that is acted on the suspension's white outer frame where the optical lever setup is physically based on. A push on the frame is sufficient to cause the beam to move out of QPD linear range. I discovered this half a year ago but I am not sure if this is relevant.
