NAOJ GW Elog Logbook 3.2
participaint: Yuhang and Matteo
After the pre-alignment of crystal, I did final algnment yesterday. The method is to move Newport 9071, vertically and herizontally.
The experience is to move two of those four adjustment screws the OPO hole where the beam comes in. The requirement of this adjustment is to make the reflection beam overlap with incident beam. However, at the same time, pay attention that the whole OPO housing should be bascially looking like not tilt. Because we know the beam is bascially aligned pretty well.
During the alignment, I found the reflection looks like the attached figure 1(for the good alignment).
Then we can move the two another side sctews. At the same time, look at the transmission, at a certain point, we can see the trasmission is bright and looks round.
We still need to do aligment of the combination of two horizental screws to make it look better. While doing this, I found the beam is cutted by the crystal after move the incident tilt screw. This means the beam we are looking at is what we want to find. While moving the outgoing tilt screw, we can see the beam is cutted and then many small blocks of lights. This looks like many reflectivity of beam inside the crystal.
In the end, I got the transmission as in attached figure 2 and 3.
Participants: Yuhang, Eleonora, Matteo
The transmission found before seems to come from a beam passing through a gap beside the crystal (it was too bright and the reflection was very small and with a strange shape)
Therefore we have restarted the alignment from the beginning. We moved the the 9071 newport Four-Axis Tilt Aligner under the OPO following the usual procedure:
1. Make the reflection superpose the incoming beam
2. Look for a good transmission, always taking care to keep the reflection superposed to the incoming beam.
After may tries we could find a resonably good shaped transmitted beam with a power of 0.047mW. The input power was 210mW. This means the reflectivity is 99.977%, which is in agreement with what we expect.
We put a dichroic mirror after the OPO. The reflection is focused on a photodiode while the trasmission is sent to a camera. Before installing the OPO we have marked the position of the transmitted beam on the screen in order to use it as a reference for the alignment.
Unfortunaltely the tramission after the dichroic mirror is too low and we cannot see any light on the camera, therefore we were not able to use the reference we took for the transmitted beam.
Participants: Yuhang, Eleonora
We have tried to recover the cavity alignment after TAMA blackout.
1) We have reset all the local controls. For the input and end mirror we have reset the SR560 which filters the error signals (2nd order lowpass filter with cut-off frequency 100 Hz and amplification factor 100)
2) We observed that the reference out of BS chamber, accunting for PR position, was very far from the good position, especially in pitch. Since the local control range was too small, we used picomotors to recover it. Pic 2 shows the reference after moving PR.
3) We had to move also the BS picomotors to recover the good position of the beam on the two in vacuum targets and make the beam to reach the screen after the end mirror, used as a reference for the alignement.
4) At this point we aligned the input mirror in order to have the reflected beam superposed on the input beam.
5) The last step is to align the end mirror. The standard procedure, when no flashes are visible, consists in raising the second target and making the beam to pass through its hole. Than the end mirorr is aligned by making the reflected beam to hit the hole of the target from the back. The reflection can be observed by monitoring the rear side of the target. Unfortunatly the range of the local control was not enough and the end mirror picomotor got stuck in yaw so we were not able to align the end mirror and to see any flash. Pic 2 shows the rear side of the second target, with the reflected beam offset in yaw with respect to the hole of few cm.
SOME DETAIL ON END MIRROR PICOMOTOR: We used both the computer and the joystick to move it. Pitch was always fine. Yaw was moving at the beginning but after a while it got stuck. We tried to switch the driver on and off many times, to swap the channels but it didn't solve the problem.
Since we moved the end picomotor in pitch we observed a higher movement of the mirror with typical peaks in the spectrum which are not normal mode of the suspension (see pic 3). It made me suspect that the intermediate mass is touching the frame of the damping magnets. We already observed something similar in PR suspension (see entry 239 and confront the spectra).
Even if the damping loop is effective in reducing the mirror motion, this should be possibliy solved.
HOW TO GO ON: according to our experience, picomotors which are stuck are likey to start working again after a while. So we will try to move it again in the next days, before considering to open the chamber.
Yesterday 27 Aug we made another try to move end mirror picomotors. We easily recovered the condition when the beam passing through the hole of the second target was reflected by the end mirror on the back of the target, but displaced few cm in yaw from the center. Unfortunately the yaw picomotor is still stuck and we could not recover the alignement. We are probabably left with no other option that opening the chamber.
SOUTH tunnel:
- Dehumidifier: all ON, all green light
- Light: all working
SOUTH end:
- Air conditioning: working
- light: all working
- phone: working (3472)
WEST tunnel:
- Dehumidifier: all ON, 9/10 green light, N.5 (last from the central area before the mid-arm) is ON but red light.
- Light: all working, 3 fluorescent tube were substituted today.
WEST end:
- Air conditioning: NOT working (further inspection needed)
- light: all working
- phone: NOT working
Participaint: Marc, Matteo and Yuhang
Since we have finished the lens installation for OPO, we arrived the alignment of OPO.
Last week, me and Marc did a similar work, including the preparation of periscope, camera, iris, OD filter and the alignment of the beam(make it hight of 140mm). However, the elevation of beam costs too much space. It makes the beam waist locate just after the periscope. After that we decide to alignment the OPO without the translation/tilt stage. However, the work seems impossible without that stage.
Yesterday, we changed the set of stage(make it hight of 111mm). This saves us a lot of space. We also did the beam measurement and find the beam is desirable.(as in the attached figure) We set up the periscope and did the pre-aligment of OPO. The pre-alignment is to make the reflection of OPO roughly overlap with incident beam. The fine-alignment will be done today by using translation/tilt stage. For the components after OPO, we set up a dichroic mirror instead of OD filter to damp the beam and also get the reflection information. For the transmission of dichroic, we use a camera to monitor. We also align it before putting OPO. And made a circle to mark it on the screen.(as in the attached figure). We also set up a photo diode for the reflection of dichroic mirror.
The video of check the reflection is here. https://drive.google.com/file/d/1-ZfhVS0iZDxF7kWpSCrn2QO5ORTTX4xU/view?usp=sharing
The whole set up is attached as well.
Therefore I decided to put a larger breadboard and realign everything.
To do that I maximized the power transmitted by the pinhole in two positions at 2deg from the center (start and end). When the pinhole is at the start position I aligned moving the steering mirror on the optical table. When the pinhole was at the end position I aligned moving the mirror on the periscope (usual procedure to align a beam to make it pass through two points).
When we went to the end room with Takahashi-san, we saw that the south end room (the one we're currently using) was hot and humid.
The power was only off since one stormy day so it can't explain this.
It seems that every now and then the air conditioning system stop in both end rooms.
Seeing the damages in the west end room it could be useful to go from time to time to check if the air conditioning is on in both end rooms.
We slightly moved the positions of the last lens of the OPO telescope as well as the one of the EOM.
We obtain the 2 characterizations attached to this entry.
The nominal value of the beam inside the OPO should be 36 um so we are now within +/- 1.3 um of the nominal value.
The OPO telescope was adjusted so to have the mean beam radius of 36 um.
The x-axis "0" is the last hole of the rail.
The 2 beams are still superposed as we could see on the beam profiler.
All the four pumping station are now back in operation and the vacuum level reached the 10^(-8) mbar level.
As a consequence of the blackout, all the three big gate valves along the arm were closed. After switching on the air compressor, the three gate valves opened.
Note: the penning at the South End station, tunnel side, measures 1.0e-7 TORR even if the other instrument (tower side) is deep in the 10^(-9) region. I tried to close the arm gate valve and the turbo pump gate valve on that side to check if the value was real or stuck and it seems that that penning cannot go lower than 1.0e-7 TORR.
The output fiber of the laser is single mode, polarization maintaining.
The reason of the jumps could be related to the polarization or some modes hops. So I tried to rotate the polarization.
Until now the polarization of the beam exiting the fiber was P (I didn't consider this parameter when I assembled it).
I rotated the fiber collimator in its mount by 90deg, in order to have S-polarization.
In this configuration, the loop works better. See the plot
Since rotating the collimator caused the misalignment, I ordered beam splitters and a half wave plate to optimize the polarization without loosing the alignment every time (they should arrive within one week.)
The out-of-loop PD was on the waist of the probe (at the sample position). The final test will be to put the out-of-loop PD on the imaging unit, where it will sense only the central part of the beam.
If the jumps we see in the intensity are related to the modes, and therefore to the shape of the beam, then the loop will not work, because the in-loop PD will see the total power, and doesn't see the shape.
On behalf of Takahashi-san whose presence and help today was crucial !!
Participants: Takahashi, Marc, Yuhang, Manuel
Today from around 15h there was a really strong storm in Mitaka.
Issue 1: power shut down.
There was a power shut down (caused by a lightning) while we were doing experiment for few minutes.
We turned off everything but many parameters have to been checked as they have been reset. We didn't yet turn on laser or others electronic devices (We heard some strange sounds we need to investigate).
Takahashi-san was there to help with the compressed air leakage.
All the pumps were also turned off!
He closed the valves of the chambers and of the pipe.
When we arrived at the south end room it was quite hot and humid (similar situation as the other one). Takahashi-san turned the air conditioning on but this might show that it turned off by itself from time to time.
Issue 2: water in central room
Coming back from the end room we found out that a lot of water was leaking inside Tama on an electrical board (probably the general power).
We removed as much water as possible but this shows that there is a leak coming from there. The electric board seem not having any damages.
Issue 3: air conditioning in elec shop
Going inside the electrical shop we saw that one air conditioning system has fallen off the wall. The humidity didn't drain outside so a lot of water was on the floor. We removed 4 buckets of water.
Isse 4: air leakage
Takahashi-san also had the time to fix the air leakage by tightening the joints (without replacing them). When we were there, also other 2 joints in the central room started leaking air. We tightened them as well.
Finally, electricity seems to be back everywhere.
A careful check will be made tomorrow.
Takahashi-san said he will turn on the pumps / open the valves tomorrow.
This entry will summarize past working days (sorry for the delay)
New Optical scheme and wiki sections :
The new optical scheme (v 31) ".svg" has been updated in the wiki. The ".png" is attached to this entry.
The available lenses have also been added to the "Optics" section of the wiki.
EOM alignment and p-pol power :
The 2 steerings mirrors before the EOM have been remplaced by PBSW-1064 which should only transmitt 5% of p-pol each.
When doing this, we saw that the lens before the EOM was a bit tilted so we aligned this EOM path again.
We now have
| before EOM | 10.0 mW |
| after EOM | 8.7mW |
| p-pol after EOM | 100 um |
Note that the power values where chosen to compute easily the % transmission.. We can increase a bit the power if needed. However to lock the OPO, it is planned to use a similar photodiode to the one used to lock the FC where the power is ~200 uW.
It seems that p-pol inside a modulator can lead to amplitude fluctuations. Is this power low enough to avoid them or should install an additional PBS before the EOM?
P-Pol and CC beam characterizations :
In order to characterize this 2 beams, we did as for the FC green reflection.
We added one f=100mm lens and check the beam profiles after it.
This 2 characterizations are attached to this entry.
The lens after the EOM was then moved a little in order to match more precisely the beam waists.
The 2 beams were aligned before the first sterring mirrors of the OPO telescope using 2 irises and checking the power transmitted.
The 2 beams are now quite well overlapped on the rail as we can see on the beam profiler.
OPO telescope :
Using the CC beam as our reference, we designed a telescope that is now placed between the 2 last steering mirrors before the OPO on a rail.
Our first simulation didn't match the data. It seems that we need to understand it a bit better. Please note that the value of the FC relfected green beam parameter before the lens is wrong!
Using Jammt for the moment we could recover the beam parameters before the lens and design the telescope.
It consists of 2 lenses as attached to this entry (sol2).
Before the storm we found a mean beam waist of 36.3 um quite close to the nominal one!
A better characterization of the 2 beams will be performed tomorrow.
The noise may come from the fiber.
I checked with a sound generator on my phone and put the speaker near the fiber. I see the peak at 4690Hz (random choice, but the same thing happen for other frequencies) on the spectrum (see pictures). Some acoustic noise is coupled to the fiber. So I fixed the fiber with some tape.
I put the OD2 filter to reduce the power, then a non-polarizing BS to send half of the beam on the in-loop PD.
I put the in-loop PD on a separate optical board standing on a periscope post because there was no space on the breadboard.
Before the in-loop PD, I put a f=50mm lens. The in-loop PD is mounted on an XY lens mount for a fine centering of the beam on the detector.
I put the out-of-loop PD on the translation stage and moves it to center the beam inside the PD.
Both PDs have a load of 7.5kOhm.
On the oscilloscope, the two signals (AC coupling) are very similar. Also, the coherence is high.
I closed the loop but the noise doesn't reduce on the out-of-loop PD.
I took the spectra with 400 peakhold averages, not linear averages, because I'm interested in reducing the nose peaks, the spikes due to the jumps.
I really need to understand where these "jumps" come from.
I made a measurement of the pump beam profile to make sure of the parameters without the last converging lens.
Result:
waist = 500um
waist position = -1m
the axis origin is the converging lens mount.
Then I updated the Jammt design using these parameters for the initial beam.
The current converging lens is f=150mm, so Jammt gives a resulting waist of 67um at 165mm from the lens. See figure.
I also made a measurement and fit after putting back the lens. the fit gives a waist of 70um at 171mm. Quite comparable with the jammt result.
then I updated the configuration to have a pump size of 70 - 75um (diameter).
See the diagram.
I need a diverging lens f=-125mm and a converging lens f=100mm
I reduced the probe size as well, from 180um to 120um, to be 3 times larger than the pump (which is 40um), but the signal doesn't change much.
I order to reduce the size of the pump and keep the waist on the same position we should add a diverging lens before the converging one.
We designed a configuration on Jammt using a -100 lens and a +100 lens. See the attached image.
The new pump size will be 37.6um (radius).
A more precise design will be after a better characterization of the pump beam profile before the lenses
I made a measurement of the pump beam profile to make sure of the parameters without the last converging lens.
Result:
waist = 500um
waist position = -1m
the axis origin is the converging lens mount.
Then I updated the Jammt design using these parameters for the initial beam.
The current converging lens is f=150mm, so Jammt gives a resulting waist of 67um at 165mm from the lens. See figure.
I also made a measurement and fit after putting back the lens. the fit gives a waist of 70um at 171mm. Quite comparable with the jammt result.
then I updated the configuration to have a pump size of 70 - 75um (diameter).
See the diagram.
I need a diverging lens f=-125mm and a converging lens f=100mm
Simulating the absorption of the surface reference, I optimized the Imagin Unit distances to have the maximum signal in the two cases, pump waist 40um and pump waist 80um. See the first plot, it shows the signal as a function of the distance d2 from the lens and the small sphere.
Using the optimum value of d2 in the two cases, I repeated the simulation of elog entry 905.
In the case of pump size 40micron, the absorption is 14.7ppm/cm, which, compared with the 60ppm/cm gives a material correction factor of 4.09
In the case of pump size 80micron, the absorption is 19.7ppm/cm, which, compared with the 60ppm/cm gives a material correction factor of 3.03
the probe size is still 180um in both cases, next step is to reduce it as well to be 3 times larger than the pump
I used ABCD matrix to calculate the beam parameter before the lens(the lens used to perform better beam measurement). The result is as following:
beam waist position: -3.2m (relative to Faraday Isolator)
beam waist size: 877.20um
I also attached the python code, if you are interested, please have a look.
Today we work on the recombination of the 2 auxiliary laser beams.
- We put the rail after the EOM so that the lens before the EOM is on an usual post. The purpose of this change was to be able to match the CC and P-Pol beam sizes. After trying a lot of lenses before the EOM, we couldn't see any drastic changes on the beam shape so we guessed this EOM doesn't affect the beam shape as much as the green one. We now have 85% transmission of the EOM. The missing few % can come from dust inside the EOM making the alignment a bit more tricky.
- We put 2 steering mirrors before the lens in order to save some space on the bench.
- The second lens is set such as the beam is collimated with diameters 2150 and 2050 um close to the CC beam size (2100 um)
- we put the lambda/2 on the rail in order to save place
- We put a 2' PBS cube and checked that the 2 beams are superposed. It is still the case after the last steering mirror.
Today the 2 beams seem to be overlapped.
One issue is that we have some power loss on the p-pol beam (EOM path) :
| before EOM | After EOM | After 1st steering mirror | before lambda/2 | After | PBS transmission | PBS reflection |
| 13.5 mW | 11.5 mW | 10.1 mW | 9.5 mW | 9.36 mW | 8.07 mW | 1.12 mW |
This might be linked with some polarization troubles. We will look this hypothesis tomorrow.
The new optical path is attached to this entry and the optical scheme will soon be updated.
After some rough simulations, it seems that one f = 125mm lens 5 holes before the OPO can provide a beam waist of 40 um. This is close to the nominal 36 um.
Anyway, we will use this lens tomorrow to make sure the 2 beam parameters are similar.
In previous entry ( #934 ) we reported that the BS pitch was saturating causing a lot of residual motion of the beam.
Yesterday, we found out that the BS pitch correction was not saturating anymore (~ -6V) and the motion was at the order of 10 urad both for pitch and yaw.
This is still yet ~5 times higher than other mirror motions at the exception of EM pitch which has similar motion.
To do the characterization we installed a f = 100 mm lens
In order to be sure that we had a good beam shape we installed a CCD after the periscope.
When the cavity is locked, the beam is shaking a lot preventing to take much more points.
Good news is that the beam doesn't seems astigmatic.
I used ABCD matrix to calculate the beam parameter before the lens(the lens used to perform better beam measurement). The result is as following:
beam waist position: -3.2m (relative to Faraday Isolator)
beam waist size: 877.20um
I also attached the python code, if you are interested, please have a look.
There were few mistakes made on this entry corrected in this one.
Attached to this entry is the proper fit of beam after the lens (previously a wrong wavelength was used for the plot).
The mean profile was used (w0 = 113.37 um 0.6981 m after the f = 100mm lens used for the characterization [lens is 40 cm after the Faraday Isolator])
The beam parameter is the following : w0 = 18.921 um @ 0.1168 m before the lens ie roughly 0.28320m after the faraday isolator.
Since my result is different from Marc's result, I did calculation again. I found a mistake in my calculation of ABCD matrixs.
From the calculation point of view, Marc's result is correct.
I will check in actual case to see if the calculation aggres with calculation or not. As I have already mentioned in the meeting, we can see the reflected beam is shaking while the filter cavity is locking. So if they don't agree with the actual case, I think the discrepancy comes from the beam shaking.