The purpose of this test was to evaluate a spring plate in a custom optic mount design which will be deployed in the reworked and upgraded PMC Spacer Assy per D1600270.
Summary of Results (Fit)
Mechanical inspection of the fabricated showed conformance to manufacturing specifications. Assembly of the Fit Test Interface Assy D1600398 in the nominal configuration was successful.
Some additional consideration (Tooling? Interfacing part?) may be necessary to combat these sources of “slop” and permit repeatable assembly.
Summary of Results (Holding Force)
The Fit Test Interface Assy D1600398 was laid flat on the work bench with the normal axis of the mirror vertical. The force required to overcome the static friction for a given spring plate defection and induce macroscopic translation was evaluated as follows:
In comparison, the weight of the optic is estimated to contribute a force of 1.6E-2 newtons in the nominal orientation. The optic did not sag in the mount when held in the nominal configuration and orientation for 2 hours.
In shipping configuration, the set screws were found to couple rigidly to the barrel of the optic and provide sufficient static friction to withstand jarring motions. The design seems to be sound from a shipping point of view.
Summary of Results (Comparison with COTS)
Two Commercial Off-The-Shelf (COTS) optic mounts were investigated. Barrel contact points, namely two flats and a preloading point-contact plunger, were used in conjunction with a spring plate. The assembly was rigidly coupled with high static friction constraining all degrees of freedom.
Perhaps barrel contact points should be employed in the LIGO design.
Full Description of Experiment
COTS Spring Plate Mounts
Two models of Commercial Off-The-Shelf (COTS) Low Outgassing optic mounts from Newport which provide axial clamping using the COTS 906919-02 Mirror Mount Spring Plate were investigated. Both use the Spring Plate in conjunction with barrel contact points, namely two flats and a preloading point-contact plunger.
An optic is loaded into these mounts by using the barrel to depress the plunger, then tilting the optic into flush face contact with a retaining lip that is integral to the mount. The plunger then pushes the optic into its seat on the two flats (visible contacting the right hand site in the image below). Subsequently, the blade spring is bolted to the exposed face, and the assembly is rigidly coupled with high static friction constraining all degrees of freedom.
Optic Mount models are:
The parts procured for the Fit Test Interface Assy D1600398-v1 were received in good condition. Using micrometers, calipers, and pin gages, Stephen employed partial inspection of mating and critical features to establish that the parts were made to specification by the vendor. The parts consisted of the COTS 906919-02 Mirror Mount Spring Plate (Qty 25, New Focus / Newport), the D1600233-v1 1in. Mirror MT. (Qty 1, ProtoLabs), the D1600397-v1 Base Plate (Qty 1, ProtoLabs) and assorted COTS hardware and fasteners.
The only item worth raising from this inspection is the difficulty of inspecting the 1.010” Diameter pocket in the D1600397-v1 base plate. No accurate measurement could be made due to the nominal .030” high shoulder, the nominal .0075” machine tool radius, and the un-deburred lip, all of which limited the contact area and precision for the calipers. A fit check of the PZT showed that the pocket was large enough to insert the PZT, and this was taken to be sufficient to “Pass” the size dimension on this feature.
To assemble the Fit Test Interface Assy, the following procedure was followed:
Note that the mirror used in this experiment was a flat mirror removed from PMC 10. See item 36 of D1001955-v2 Assembly Drawing.
Using the Fit Test Interface Assy D1600398 and a 36 ounce Jonard Compression/Tension Force Guage, the holding force of the 906919-02 Mirror Mount Spring Plate was tested.
This holding force was found to hold the optic tightly without macroscopic translation in the nominal mount configuration (0.050” axial deflection on spring plate) until acted upon by approximately 6 ounces of force (converted, about 1.7 newtons). This force was applied approximately perpendicular to points on the barrel of the mirror near the face opposite the 3-point contact plane.
SEE 50thou Holding Force.MOV
In contrast to the nominal 0.050” of deflection on the spring plate, under an axial deflection of 0.030” approximately 4 ounces of force (1.1 newtons) were required to overcome the static friction and induce macroscopic translation. This set-up utilized precise shims to reduce the deflection, as shown below.
SEE 30thou Holding Force.MOV
The force required to induce macroscopic translation without the spring plate installed was too small to be measured with the same set-up.
SEE NoSpring Holding Force.MOV
The weight of the optic contributes a force of 1.6E-2 newtons in the nominal orientation, using a calculated volume of 0.78 cm^3 and an estimate density of 2.2 grams per cm^3. The optic was set in the Interface Assy for 2 hours under the nominal configuration, and no gravitational sag was observed (using a pin gage as a probe).
Evaluation of Shipping Configuration
In order to push the limits of the holding force of the mount for shipping purposes, the mount was carefully subjected to a repeated jarring motion (held with one hand and "clapped" firmly into the other) with the set screws driven tightly into the barrels. The contact provided by the set screws and the spring plate in conjunction was sufficient that the optic did not macroscopically move (using a pin gage as a probe). Meanwhile, when the optic was held only by the spring plate and the mount was subjected to the same jarring motion, the optic did macroscopically move. The set screw contact on the barrel is very much necessary for shipping of the optic in place within this mount in order to provide sufficient static friction to rigidly couple the optic to the mount.
I haven't posted the BS transfer functions results since we started to test the BS with 2 BRDs attached. I can now read the data that has been saved in .78D from the spectrum analyzer, which makes data analysis much easier! (see elog 171).
The results are summarized in the tables below and in the figures attached.
B1 Freq. [Hz]
Is the low Q of measurement B7 due to a very good tuning? See elog 164, BRD3 was tuned to 0.01%. The measurement B7 was taken soon after tuning, and the BRDs might have drifted for the folllowing measurement B8. However, we don't see this good performance in the overall scan that was taken just after installing BRD3 on BS (see figure 3).
To do: retake transfer functions of the bounce mode region with and without the dampers to confirm the Qs. Understand the difference in the measurements (drifts?). A tentative summary is presented in figure 1, but the undamped mode is missing from the figure.
The results are so far are cleaner for the roll mode (see figure 2).
R1 Freq. [Hz]
We can see that adding a second BRD helped to reduce the Q of the second resonance from ~250 to ~170. From elog , we know that the undamped Q of BS is around 3000, and that adding the first BRD split the resonance in two, with Q1 below 200 and Q2 around 250. A third resonance/feature can be seen on the glocal scans, but it is outside the frequncy range of our current measurements.
To do: Take measurements over a larger range (on the high frequency side) to check the third resonance. Add a comparison with the model to check if the measured behavior is expected.
Andy retuned BRD1_v4 and BRD3_v4 after they had been left alone for a week in the lab. The tuning is described here: alog 177. It is better than 0.4% for all modes of the BRDs. I reinstalled them on BS right afterwards and took measurements of the bounce and roll modes over the following days:
We can see (figure 1) that the resonances are drifting. The BS bounce mode is 16.69 Hz without BRDs, so it is probably B2 here with the added mass.
The roll mode also experiences drifts, but to a lesser extent. The BS roll mode is 24.34 Hz without the BRDs. It is probably R2 here with the added mass.
Unfortunately, some of the measurements are not readable. I will retry the conversion from .78D to .txt. Files I have to mention that saving the data from the spectrum analyzer is unreliable at times, I already lost few hours of measurements because the files were not saved correctly into .78D in the first place (.78D files were empty).
[Marie, Andy, Agueda]
Using a beam analyzer, we found that the laser beam had a mean width (diameter) of 5438.17 micro meters. Below are the images produced by the beam analyzer.
We then proceeded to record the second point absorber on the test mass with the IR camera. Although we increased the beam's percent power every minute, with the maximum being at 95%, the laser did not show up on the recording.
Below are my calculations to find the millimeter/pixel (mm/pxl) within two images taken by the IR camera: one of the first point absorber and one of the second point absorber that were previously on the test mass. Attached are the pictures used to find the mm/pxl results.
PT Absorber 1
*The pythagorean theorem was used only for the first point absorber's calculations, as the pixel ruler on Preview would not work because the mark is angled. Changing the angle of the picture would change the picture size, thus producing an inaccurate mm/pxl result*
(Marker as a Whole)
- 9.5 mm/ 251 pxls wide = 0.04 mm/pxl
- 6.0 mm/ 153 pxls tall = 0.04 mm/pxl
(Inner Circle of Marker)
- 2.5 mm/ 73 pxls wide = 0.03 mm/pxl
- 1.9 mm/ 51 pxls tall = 0.03 mm/pxl
- 3.0 mm/ 93 pxls thick (left) = 0.03 mm/pxl
- 3.1 mm / 97 pxls thick (right) = 0.03 mm/pxl
- 1.9 mm/ 52 pxls thick (top) = 0.04 mm/pxl
- 1.5 mm/ 45 pxls thick (bottom) = 0.03 mm/pxl
PT Absorber 2
- 4.5 mm/ 136 pxls wide = 0.03mm/pxl
- 5.8 mm/ 170 pxls tall = 0.03 mm/pxl
- 1.5 mm/ 55 pxls wide = 0.03 mm/pxl
- 2.0 mm/ 62 pxls tall = 0.03 mm/pxl
- 1.6 mm/ 52 pxls thick (left) = 0.03 mm/pxl
- 1.0 mm/ 45 pxls (right) = 0.22 mm/pxl
- 1.9 mm/ 60 pxls (bottom) = 0.03 mm/pxl
- 1.5mm/ 58 pxls (top) = 0.03 mm/pxl
BRDs with the version 4 of the blades have been tested on stand-alone version and in the BS suspension during the summer. Here is a summary of our findings with the references to the corresponding elogs.
Drifts measured on two BRDs before baking over 20 and 40 days (elog 186). The variation of the resonance frequencies are reported in the table below:
Time lapse dt [days]
Drift df [Hz]
Drift df [%]
Drift Rate [mHz/day]
Drift Rate [%/day]
On the dummy BS suspension:
1 - Resonance frequencies
The two BRDs were installed on the dummy BS for a month. Unfortunately we are missing some of the measurements because some data got corrupted (see spreadsheet attached). Therefore the analysis is only performed over 16 days (8 days) for the bounce (roll) mode.
For each mode (bounce and roll), two peaks are observed around the resonance where we expected to resolve three peaks. The three peaks would be the main BS resonance as well as one peak per BRD. We might need to increase the scan resolution (see figure 2 and 3).
The frequency of the peaks do not match the frequency of the BRDs measured alone. The frequency of the minima in between the resonance peaks is close to the resonance frequency of the BS for respectively the bounce and the roll modes. The shift in the BS resonance frequency due to the added mass is negligible.
We observed a steady drift of the resonance frequencies over time. The frequencies are increasing by few hundreds of ppm per day, see below and figures 4 and 5.
Summary of BS Bounce frequencies
Initial mistuning [%]
Final mistuning [%]
We are observing a change of 0.2% in the resonance frequency over the 16 days. This is above the requirements that we set at 0.1% tuning. We didn’t observe a stabilization in the drift.
Summary of BS Roll frequencies
We are observing a change of 0.1% in the resonance frequency over the 8 days. We didn’t observe a stabilization in the drift, so this is likely to exceed our requirements.
When remeasured stand-alone after being uninstalled from the BS suspension, we established that the BRDs frequencies drifted of about 0.5% after being for a month on the suspension (elog 203).
2 - Q factor
The resonance amplitudes, corresponding to the quality factor of the modes, fluctuate over time without a distinguishable pattern. However, it seems that for each mode the two resonances vary together, in particular for the roll modes.
The mean of the bounce mode Q is 140 (168) for the 16.66 Hz (16.69 Hz) resonance. The mean of the bounce mode Q is 115 (152) for the 24 Hz (24.5 Hz) resonance. Q are lower than expected according to the model (we expected Q~200).
We observed a steady drift of the resonance frequencies of the BRDs over time, when stand alone or on the dummy BS. The frequencies are increasing by few hundreds of ppm per day. We see a stabilization in the drifts after about a month in the lab. The drift is slightly lower when the BRDs are mounted on the BS suspension compared to the stand alone BRDs in the lab. This could confirm that the excitation measurements cause some of the drift and we need to revise the method. We have no evidence that the baking process reduces the frequency drifts.
The measurements of the quality factor shows that the peaks may not be quite resolved, and that we are underestimating the Qs of the modes on the suspension. However, the value of the Qs in the stand-alone measurements is already promising.
====== Bounce Mode
Here is a summary of the BS bounce mode survey with BRD5_v5 and BRD6_v6 attached to the suspension for 6 months.
Details of the measurements are attached in the spreadsheet.
The measured transfer functions over time are shown in figure 1. We observed a single peak in the data. This is unexpected from our model.
Therefore it seems that the bounce damping is pertty stable over 6 months. We didn't analyze correlations of the variations with environmental factors in the lab.
Measurements after taking off the BRD from BS:
Recall that pre-installation measurements were:
====== Roll mode
Here is a summary of the BS roll mode survey with BRD5_v5 and BRD6_v6 attached to the suspension for 6 months.
Details of the measurements are attached in the spreadsheet.
The measured transfer functions over time are shown in figure 1. We observed two peaks in the data. This is in agreement with our model if the tuning if the BRD is within 0.1% of the BS roll mode (see T1900846).
Therefore it seems that the bounce damping is pretty stable over 6 months. We didn't analyze correlations of the variations with environmental factors in the lab.
Recall that pre-installation measurements were:
This morning I handed off the parts for the 5 LHO BRDs to Bob Cottingham at LLO for Clean&Bake (https://ics.ligo-la.caltech.edu/JIRA/browse/clean-10214)
The parts for each BRD are in separate containers (see pictures attached).
Here are the masses of each component in order to reassemble the BRDs after C&B:
I didn't observe frequency drifts during the month of assembly and monitoring in the Optics Lab. This is not expected from our experiments in the Modal Lab, but it makes the preparation for the sites BRDs easier.
BOUNCE: In anticipation to the frequency drifts, I had tuned the BRDs on the lower side of the target frequency. But the tuning didn't drift so I changed the masses last week for the bounce mode in order to be in the 1% target. The jumps in the curves below are due to retuning (on November 5th for BRD1, on November 20th for the other ones).
Roll: I didn't retune the Roll modes after assembly on October 20th. In the last days, I was experimenting with different ways to excite them (see pictures attached), so this is probably the cause of the slight drifts that we see.
This is image of the 4 sectioned green lantern on the mock up quad at cit.
1) BA to complete cabling for use in chamber
2) CT to clean & bake
3) KG to practice imaging. unit is in 318 Downs on quad mock-up.
Harrison has put together a first look at what a revised HSTS with blades at middle mass might look like. This is a work in progress, some parts missing (magnet /flag assemblies etc.)
This week was spent mainly finishing up designs for the new HSTS middle mass. The main tasks were:
The top plate was basically done at this point, all that remained was to make sure all of the holes were the correct tap size and such. What required the most attention was clamps holding the wires from above. I made an initial design and added holes to the top plate such that the clamp could be attached from below. However, once these clamps were in place, I realized the wire separation was too small. I made a change that created an asymmetrical piece with a handedness but, after a discussion with Calum, I modified the top plate so that we could go back to the symmetric design and maintain the correct wire seperation.
As for the clamps on the blades holding the bottom mass, I adapted a previous design to not only maintain the wire separation, but also to account for the new angle of the wires (previously the wires from the middle mass down to the bottom mass were vertical, the addition of the blades doesn't allow for this)
The main goal for this week was to start getting engineering drawings in order for fabrication. With Eddie's help, I drew up all of the parts in the new middle mass. After a couple rounds of red lines, I got the drawings to a near-finished state. I also got DCC numbers for all of these parts.
While doing this, I realized that I had not figured a way to attach the two bottom pieces (pictured below).
We decided brackets were the way to go, so I drew some up in solidworks, got another DCC number for them, and added them to the assembly.
I also began a Bill of Materials for the new middle mass
I had a few things left over from last week to finish up early this week, like finishing a few engineering drawings and fleshing out the Bill of Materials (I have attached its state so far to this entry). I also made a drawing of the new blades for this mass using the parameters Norna found.
In addition, I added in a roll adjuster similar to the pitch adjuster and added tapped holes for set screws to hold the adjusters in place.
Norna, Calum, and I met on Thursday to discuss what I'd be doing for the next couple of weeks. The tasks we settled on are as follows:
1) Finish Bill of Materials (BOM). Finish getting DCC numbers. Put onto Vault(?)
2) Write up what you have done in the design and explain why. Could add to existing document.
3) Blade design/ drawing . Also FEA of blade. See Calum's reference.
4) Work with Eddie to get quotes for parts. Write statement of work. Could also get quotes for rapid prototype version.
5) Put together rendering of whole suspension including bottom wires going round mass - use a glass mass.
6) Adjust length of bottom wires to get the orignal overall lenght of HSTS. Note new length
7) Look into redesign of wire jig for producing clamp/wire clamp assembly to take your new clamp design. Draw up modification.
8) FEA of new mass - basic modes in free / free case cf top mass also look at adding loads to represent blades
9) Get set-up going to measure mode frequencies of triple suspension in lab Could use B&k or spectrum analyzer - need to think about non-contact probe or accelerometer ( this item in conjunction with Norna /Calum).
10) deep fallback - resurrect Kristen's middle mas for HLTS and apply design considerations used in HSTs to continue that design.
I forgot to make drawings of the T-section and I-section, I need to do that as soon as I return after the holiday weekend
This week I tackled items 1-3 on the to do list I got last week.
All of the parts I have drawn for the new HSTS middle mass have been added to a sandbox file on the vault. I have also summarized all of the work done so far on the new middle mass in a review document on the DCC (found here). Finally, I began running FEA on the new blade design to A) find its normal modes and B) make sure the deflection we calculated was accurate. The findings of that analysis are in a report on the DCC (found here). At first, the deflection was not looking correct, but after consulting with Calum and reviewing an FEA he ran on some other blades, I realized I had entered my loading conditions wrong. I corrected them and got the following deformation.
I'll begin next week with doing FEA on the entire mass and adding all the FEA results to the overall report. Hopefully we will also begin getting quotes for parts
I forgot to enter this on the 18th, so here is what I did the week ending on the 18th.
I completed FEA of the overall middle mass and found the "wing mode." It occurred at 217.03 Hz.
I have begun on a rendering of the HSTS with the new middle mass. In doing so, I have realized that the clamps on the middle mass do not quite line up with the wires. I have done some tweaking and we will see when Norna returns if the fix is acceptable.
We also began work on the mini-project Calum presented to us. I have some preliminary designs, but I'd like to get my hands on some of the mesh I want to use to get a feel for it.
Pumping down on Quad Nov 25th.
With Vac at 2.15e-5 Torr ran B&K system for 256s with 1 average. df=3.9 m Hz. Re-ran Graphical Setup in Trigger and changed hold off to 100 from 1000s.
Goal: thermal conductivity test depending on pressure
Setup: no gasket at the ring heater; no thermal pad on the thermocouple, thermocoupel at the center of the back plate
Current: 200 mA
Expected max temperature: 130 C
Pressure: 2 10-6 torr - 8 10-8 torr
I measured the temperature during 3 days at constant current just to if there is any influence of the pressure on the result.
Result: The temperature measurement in future can be started at low 10-6 torr range (an overnight pump down for Thomas vacuum chamber)
Goal: cheking repetability of the tempreture elevation
Current: 250 mA
Expected max temperature: 200 C
Pressure: 2 10-7 torr
Increasing resistance with temperature (decreasing current):
This is a new log started on Sept 30 2016. For the older logs and procedures please see https://dcc.ligo.org/LIGO-T1600205
The modified Viewport cover was installed using the new adapter ring. No leaks. Pump down looked Ok.
Goal: cheking contact at thermo couple is ok ore needs improovement
Setup: no gasket at the ring heater; indium thermal pad on the thermocouple, thermocoupel at the center of the back plate
Current: 189(207), 250(258), 300 mA - 7:30 AM, 11:30 AM, 3:30 PM
Expected max temperature: 220 C
Results: The temperature measured with the thermocouple almost did not change by adding indium foil between the washer and the gold plated plate (red curve). Hovewer the temperature at the ring heater changed. A second run with no indium (after reassembling the setup) shown a consistent temperature at the gold plated surface and again completely different temperature at the ring heater. Pressure was the same during all three measurement.
Goal: comparison of the temperature at the center of the gold plated plate during two runs: without gasket and with In gasket
Setup: thermocouple at the center
Current: 190, 250 and 300 mA
Pressure: 2 10-6 torr
In pad under the thermocouple washer after the first run :
Mounting the indium gasket
Results: This two runs did not demonstrate the same dramatic improvement by adding the gasket as previously. The most probable reason is the decreased compression because of adding a thermocouple with a washer to the assembly. Another confirmation of the bad compression is higher ring heater temperature comparing to previous runs.
Goal: comparison of the temperature at the center of the gold plated plate during two runs: with 4 and with 3 washers (part #4 per D1500387)
Setup: thermocouple at the center, In gasket
Current: 190(207), 250(258) and 300(308) mA
Pressure: 4 10-7 torr
Using 3 washers instead of 4 improves the thermal contact. The temperature at the thermocouple did not change but the ring heater doesn't warm up as hot as before.
Goal: Position dependent measurement of the temperature at the plate.
Setup: thermocouple next to the center with capton tape
Pressure: 5 10-6 torr
I moved the thermocouple from the center because it's washer changes the amount of compression and the measurement it difficult to compare with other once. Also the bump at the washer makes holes in the plating if everything is tighten. The measurement shows that more compression is needed (less washers)
Measured lasaer power at the end of the layout with defould pyrex viewport, fused silica AR coated viewport and no viewport. See https://dcc.ligo.org/T1700003-v1 for more info.
Laser power measured using a “PM100USB” power meter and an S314C sensor (±3% measurement uncertainty at 1064 nm). Gray dots represent the power indicated at the laser display which can be inaccurate. There are additional mirror losses at the beam transport via RTS setup. That is why power at LDF setup location will differ from the values shown on the laser display.
Black curve on figure1 is the power measured at the end of the LDF laser layout (including a window from the viewport cover) with no viewport (figure 2). Blue and red curves represent the power measured with a fused silica and pyrex glass correspondently. In both cases the glass was added at the end of the laser layout (see figure 3).
Figure 1: 1064 laser power measured as a function of the input value
Figure 2: Power-meter installed at the end of the LDF laser layout (correspond to the black curve on figure 1)
Figure 3: Power-meter installed at the end of the LDF laser layout plus the fused silica glass (correspond to the blue curve on figure 1)
Almost 100% transmission trough the fused silica viewport has been observed. Losses in fused silica were less that the resolution of the power meter. In contrast, losses up to about 15% were observed in the pyrex glass. An integrating sphere may be used for more accurate measurement.
Calibration function between the display input and the power in vacuum chamber after passing the viewport (±3% measurement uncertainty in not considered):
PLDF[W]=-4.796812+ 0.549683 Pdisplay input [%]
This log entry reflects on recent efforts to image the 1064nm spot on samples undergoing BRDF measurement within the CIT CASI scatterometer. The cameras used are (reiterating from eLOG ENG_Labs/33)
The set up and measurement is described by the following procedure:
Hardware for the cameras included:
The acquired images may be summarized by the following points:
Image 07: Lights on, scout OpLev GigE Camera Link to casi_test_07_laptop_lights_high_exposure_10_mw_high_gain.bmp
Image 08: Lights off, scout OpLev GigE Camera Link to casi_test_08_no_lights_high_exposure_10_mw_high_gain.bmp
Image 09 Lights off, bandpass filter, scout OpLev GigE Camera Link to casi_test_09_no_lights_high_exposure_10_mw_high_gain_bandpass_filter.bmp
Image 13: Lights on, ace Site GigE Camera Link to casi_test_13_site_gige_foyer_light_auto_exposure_10_mw_auto_gain.bmp
Image 15: Lights off, ace Site GigE Camera Link to casi_test_15_site_gige_foyer_light_auto_exposure_10_mw_100_gain.bmp
Image 19: Lights off, bandpass filter, ace Site GigE Camera Link to casi_test_19_site_gige_no_light_auto_exposure_10_mw_100_gain_bandpass_filter.bmp
PZT`s transfer function measurement with a test v-block was done in tree differen configurations:
- no viton (blue curve)
- viton under PZT (orange curve)
- viton only under the front part of the PZT (green curve)
Conclusion: 3rd configuration is the winner. It provides stiff clamping of the back of the PZT plus dumping. Need to design new v-block type mount with two clamps. Promissing first resonance at 3kHz with the new mount
New mount has been reworked https://dcc.ligo.org/D1700002 in order to use 2 v-lamps. See modefied mount https://dcc.ligo.org/LIGO-D1700002-v5
See T1600060 the transfer function measured with four different configurations: - elliptical mirror, no viton (black curve)
- elliptical mirror, viton under tip of the PZT (blue curve)
- elliptical mirror, viton (full length) (green curve)
- 2" mirror, viton (full length) (red curve)
First resonance appears at about 2 kHz which is very close to the internal resonance frequency of the PZT (3 kHz with no load according to the specs). Adding viton in the grove did not dump the resonance (black curve vs green and blue) however it can be an option. Changing the mirror from 2" to a lighter elliptical is still a significant improvement even with the new mount (red curve vs others).
Worked on a laser layout for in-air fused silica viewport optic laser damage test. Made a temporary enclosure to prevent high power laser scatter and to stop any possible fragments of potentially damaged viewport. A “labyrinth” shape was built instead of making holes for the laser tube. The top of the enclosure will be covered with one large black panel as well. The beam was focused to 2w=100 microns at the position of the target (red line)
Added a magnetic base for the 3" diameter target mount. The mount is angled in order to dump the reflection from uncoated fused silica target. Beam on the target is 127 by 91micron ellipse. Irradiation is planned at 25, 50, 75 and 100% power (100%=50W). Equivalent power densities and irradiation runs see in the dcc doc https://dcc.ligo.org/S1700118-v1
Made series of laser irradiation on a 3” fused silica uncoated optic https://dcc.ligo.org/LIGO-S1700118-v4
The optic is planned to be used as a viewport at LDF. In air laser damage test is required before using the optic as a viewport on the LDF vacuum chamber. Beam on the target is 127 by 91 microns ellipse. See estimated equivalent laser power input table and layout picture in the related doc https://dcc.ligo.org/LIGO-T1700184-v1
No damage was observed using DF microscope. The inspection was done using manual control – no scanning. After each irradiation run a picture was taken with a FLIR camera. Post irradiation pictures taken with DF microscope are posted herte https://dcc.ligo.org/LIGO-S1700118.
Warming up of the beam dumps up to 57 deg C was detected. No other elements of the set up were changing temperature (including the target – 3” fused silica optic and it`s mount)
WIP (photos will be posted, more information will be gathered)
A rough OFI shroud fitcheck was done on a earlier version of the structure. We found out a lot of issues with various types custom hardware. Most common problem - tapped not all the way through where it needs to be (D1700233, D1700244), bad threads on D1800111. The hardware has been re-tapped with clean taps.
Some photos of the assembled shroud are attached (did not use viton and coated hardware for the fit check because some parts were still at the C&B etc.)
I designed a rectangular shaped coil loop with 1A of current building off of an Ansys tutorial and generated a solution which plotted the magnetic field due to the coil. The attached pictures show the magnetic field magnitude and the magnetic field vector from various angles of viewing the coil. The thicker and smaller arrows signifying the magnetic field vector represent the same magnetic field and the thickness of the arrows does not indicate anything about the field; both pictures are included just for ease of viewing and interpretation. As expected for a current carrying loop, the magnetic field lines appear to go through the coil and loop around the sides and the field seems stronger in the center of the coil compared to the outside magnetic field loops.
Next steps with this design include estimating field strength expected from such a coil and verifying those calculations with these simulation results as well as designing a similar project for a circular coil and then also a coil that matches the shape and measurements of those used for the fast shutter. After generating the magnetic field for one coil used for the fast shutter, the field due to two such identical coils placed next to each other can be modeled as well.
Project Name : Coil_practice | Design Name: Rectangular_coil (Magnetostatic)
Similar to the rectangular coil posted earlier, I desiged a circular coil loop with 1A of current and generated a solution which plotted the magnetic field due to the coil. The attached pictures show the magnetic field magnitude and the magnetic field vector from various angles of viewing the coil. As expected for a current carrying loop, the magnetic field lines appear to go through the coil and loop around the sides and the field is stronger towards the center of the coil compared to the outside magnetic field loops, which is consistent with the magnetic field from the rectangular loop simulation. A Boolean union of parts of the geometries used to construct the rectangular and circular coil can be used to design the geometry for the coils part of the fast shutter so that those magnetic fields can be modeled as well.
Project Name : Coil_practice | Design Name: Circle_coil (Magnetostatic)
Magnetic field magnitude and vector solution generated for a coil with the dimensions of a coil part of the Fast Shutter with 1 turn and 1A of current. As we saw for the circular and rectangular coil models in Ansys, the magnetic field lines for this model also go through the coil and loop around the sides and the magnetic field is stronger in the center of the coil compared to the outside magnetic field loops for this coil as well. Next steps include exploring how to add more turns to the model as well as placing 2 of these coils next to each other to more accurately model the current Fast Shutter setup.
Project Name: Coil_Practice | Design Name: Fast_shutter_one_coil (Magnetostatic)
Little late on the update, but modeled the magnetic field of the fast shutter coil in Comsol (it previously had been modeled in Ansys). Both coils were modeled as single conducors with 1A of current with the appropriate dimensions and attached is the picture of the magnetic field density/field. As we would expect, the magnetic field and magnetic flux density as shown in the picture are strongest in between the two coils and drop off on the outside sides of both coils and the picture fits with intuition. In future posts, I will be uploading pictures of the coils modeled as homogenized multi-turn coils with the ability to adjust number of turns as well as the addition of a magnet moving in between the two coils to model our current fast shutter system setup.
The Boltable Structure assemby (D1101986) has been modified to accept Test Plates (D1800232) for vibration testing.
The new "Boltable Structure with Test Plates" assembly (D1800237) has been assembled and all bolts have been torqued to spec.
The assembly is in the Modal Lab and is ready for testing. See attached picture.
Added in the NdFeB magnet used in the lab into the Comsol simulation model in order to get a reasonable stationary model with the magnet and two coils. The two coils have been updated to be homogenized multi-turn coils with 500 turns as in our current physical setup and currently have 1A of current each.
The flux density and field for the coils and the magnet are plotted in the below pictures and I am currently in the process of checking with Comsol about how the magnet's magnetic field arrows are not drawn as we would expect (looping from one end to another) and whether that's a limitation of how the arrows are drawn or of the model. With this model, the force has been calculated as a function of the magnet's location in the z-direction (moving up and down between the coils) and pictures of that will be posted soon as the plotting settings are sorted out. For now, a picture of the force calculated on a rectangular model from the Comsol training Rich went to is attached.
Next steps with this model are to verify whether the lifting force for the magnet from the simulation compares with the gravitational force calculated and magnetic force determined from lab with the physical setup of the first generation fast shutter (by measuring the current required) so that we have an indication of whether the simulation accurately represents what has been measured with the physical setup in lab. After that confirmation, next steps are to understand the impact of changing the magnet geometry on the force required to lift the magnet and in particular see what effect changing the thickness and length of the magnet have and whether there is a limiting point at which the addition of mass of a magnet as we increase a dimension outweighs any advantage we may gain by having more of the magnet to be affected by the coils' magnetic fields. The biggest change to introduce to this stationary model in the coming weeks is the modeling of the response when current is initially applied to the coils which will have some rise time till it reaches its intended value due to the inductance of the coils. An external circuit modeling this inductance will need to be added to this model so that we can determine the force necessary to overcome this inductance and the optimal magnet geometry for that. Thus, the three main upcoming goals are 1) verify lifting force for magnet in simulation matches what was measured in lab, 2) understand optimal magnet geometry, and 3) understanding the transient model before we reach steady-state.
Contributors: Craig, Alexei, Stephen
Project: VMD Testing (BK Connect)
Summary: Non-excitation measurements by laser vibrometer were successful. We will work on curve fitting the spectra to extract Qs, but this technique seems to work well.
The following measurements have been made:
Remember, our bolted-base excitation measurement in the longitudinal direction was poor, so we did not feel like the non-excitation measurement would be informative.
We made measurements of a field calibrator reference, which oscillates at 159.2 Hz at a velocity of 10 m/s. We measured the velocity of this peak using a separate setup within our VMD Testing project, using 2 vibrometers. We adjusted the sensitivity of each vibrometer in our hardware table using the proportional relationship:
[Old Sensitivity * (Measured Velocity / Nominal Velocity)] = [New Sensitivity]
We then collected non-excitation measurements in two directions at once using the two calibrated vibrometers, and then swapped the response direction of each vibrometer to make a second measurement. There was good agreement between the two vibrometers, and the technique seemed very promising for resolving the peaks fully.
We investigated comparison with the autospectrum of the excitation measurements, and we'll tie up lose ends on that - seems like a visual check of the shape and the prominence of the resonance peak will be a good way to confirm that this technique is working.
None to share this time - oops!
Contributors: Alexei, Stephen
Project: PMC Bounce Mode Testing
Summary: PMC body mode testing performed with kinematic mounts and no kinematic mounts (bolted directly to breadboard with mounting blocks).
Description: A variety of vibrometer modal measurements were taken to test the PMC bounce modes.
Relevant Links: https://dcc.ligo.org/DocDB/0126/D1600227/011/D1600227-v11.PDF
Contributors: Stephen, Rich
Project: Optical Cavity Eddy Current Damping (see 120)
Summary: Mechanical design shared with Rich (Dean unavailable). Quotes obtained from Proto Labs for machined parts. Surface Field of Ni-plated ND42 magnets was measured and compared with magnets from prototype.
Characterization of Magnets
The following questions I came up with after listening in to the VMD design review on 2/13/2019
1) One method of tightening the VMD assembly is the allen key pattern in the top of the copper rod. If this process of tightening the assembly needs to be repeatable, how will the distortion of this pattern be taken into account? Just in the lab, the during re-assembly there has become noticeable distortion of this key pattern making it difficult to obtain the correct torque when tightening the assembly.
2) I believe I heard there can be a +/- 1 Hz difference in resonance frequencies between Air/Vacuum, but how will epoxy react to this difference? Also, is there concerns of outgassing of the epoxy?
I hope to have those questions clarified by Stephen/Calum next week
Work in Progress
Contributors: Alexei, Stephen
Project: Optical Cavity Eddy Current Damping Mode Testing (BK Connect)
Summary: Non-excitation, transverse, and longitudinal excitation measurements by laser vibrometer (OMC and Ometron). Recordings taken and fitted with fft spectrum to compare damping effects of different magnet configurations used for eddy current damping.
The pendulum mode of the optical cavity assembly was recorded when the assembly was undamped, damped by 4 magnets oriented at 45 degrees from the horizontal and with 2 magnets oriented along the centerline. Distance of the magnets from the copper sheeting of the cavity was also tested (magnets kept at distance of 1.5mm and 3mm).
Note: OMC laser vibrometer has issue of signal LED having gone out. Prompted huddle tests.
Looking at this work, the next step will be to re-generate plots with (1) improved resolution [Q should be resolved with (3dB width) > (10*resolution), if possible!] and (2) on longitudinal scales (either physical units or dB) to enable qualitative visual comparison. (also, I think the longitudinal movies you took didn't get posted?)
The following step would be to make a quantitative comparison of Q - see below table, intended as a mockup of the important information to summarize.
The key outcome of this effort is assessing whether the prototype damping scheme is working, and if so, what degrees of freedom should be incorporated into the final design of the damper. Hope this helps clarify why these requests are being made!
Best damping parameters
Today we moved the shelves containing the electronics for the BOSEM control and acquisition from Thomas Lab to the Modal Lab. The shelves are now near the dummy BS. We are planning to measure the bounce and roll modes of the BS suspension with and without BRDs. Please find attached some pictures for visual reference.
Luis fixed the satellite box for us, thanks! It is now in the lab, we still have to incorporate it back in the setup.
We received the 12 new springs made with a sandwich layer of Pyralux damping today (see dcc D1700188-v3).
According to T1700176, the bounce and roll modes of the dummy BS in the modal lab are respectively 16.70 Hz and 24.34 Hz. Aiming for a Q~100 for the damped BS, the relative mass of the damper and the modes has to be > 2.0e-4. It means the masses have to be >= 5.556 g for the bounce mode and >= 2.580 g for the roll mode, taking into account the weight of the spring (0.168 g) and the screws (~0.49g each). I got: mb = 0.91 + 4.23 + screw = 5.622 g for bounce and mr = 2*1.06 + screw = 2.620 g for roll.
Calum helped me taking the first measurement of the new BRD with the B&K software and the vibrometer. We found that the roll resonance is at 27.750 Hz (i.e. 14% too high), with a measured BRD Q of 59 (TBC).
[Calum, Chub, Marie]
We installed a new support for the BOSEM (see first picture attached). The bosem translation stage seats on an aluminium plate, above a bridge in aluminium that is clamped to the BS cage.
Then we reconnected the satellite box in the setup, and we were able to restart the bosem excitation and the data acquisition. We removed the BRDs that had been previously installed on the pum. Then we measured the BS bounce and roll modes:
We will scan around the resonances to check if the frequency is the sameas it was measured previously.
On the side, I measured the BRD that was assembled yesterday (brd_v4_n1) with the vibrometer:
There is quite a large dispersion in the measurement of the Q, I have to improve (or pratice) the technique to excite only the main mode and avoid saturations (only 9 out of my 30 measurements are clean for bounce and 5 out of 30 for roll).
The Q's are lower than what had been measured with the version 3 of the blades, so we are probably going in the right direction (bounce mode Q ~ 135, roll mode Q~125 in T1700176-v7, section 6.2)