I have realigned the steering mirrors for PMC because the transmitted light had been at ~ 0.741

After the alignment it went back to ~ 0.850.

I added an ALS feedback path on the MC2 suspension and this path will enable us to stabilise the MC length using the ALS scheme.

The high frequency noise, which has been a dominant noise above 30 Hz in the Y arm ALS (#6133), decreased by a factor of 5.

This reduction was done by increasing the modulation depth at the Y end PDH locking. Now the noise floor at 100 Hz went to 0.2 pm/sqrtHz.

However the noise source is not yet identified and hence it needs a further investigation.

 (Increasing the modulation depth)

Now a power normalization is doable for the LSC error signals.

It is working fine, but at some point we may want to have some kind of a saturation filter or limiter to avoid dividing a signal by a small number.

 (How to set the normalization)

•   Click a small matrix panel on the LSC OVERVIEW window (shown in the attached screen shot below).
  •     This will give you a pop-up-window, which shows a matrix to route the normalization signals

•   Choose a numerator channel, which you want to divide, and choose denominator channels, which you want to use as a power normalization factor.
•   Put some number in the corresponding matrix elements.
•   Once you put a non-zero element in the matrix, the corresponding numerator channel will be divided by the specified denominator channels.
  •     Otherwise the static normalization factors (e.g. C1:LSC-AS55_POW_NORM, etc.,) will be used for the denominator.

Is there a reason the framebuilder status light is red for all the front ends?

Also, I reenabled PRM watchdog.

It turned out that the power normalization need a modification.

I will work on it tomorrow and it will take approximately 2 hours to finish the modification.

     Concept of Power Normalization         

1.  Static power normalization, which should be placed before the input matrix.
2.  Dynamic power normalization, which should be placed after the input matrix.

Apparently there is a bug in the timing cards having to do with the new year roll-over that is causing front-end problems.  From Rolf:

For systems using the Spectracom IRIG-B cards for timing information, the code did not properly roll over the time for
2012 (still thinks it is 2011 and get reports from DAQ of timing errors (0x4000)). I have made a temporary fix for this
in the controller.c code in branch-2.3, branch-2.4 and release 2.3.1. 

I was going to check to see if the 40m is suffering from this. I'll be over to see if that's the problem.

I found the PSL laser has been off for four hours. Nobody seemed to know why.

I just turned it on and it is now providing about 10% lower power compared with one before the shutdown.
Let's keep the eyes on the power if it can recover as the housing gets warm.

I updated the design requirements for the Stewart platform.  I weighed the unloaded "dirty" SOS that was sitting on the workbench in the control room; its mass is 11 kg.  Steve suggested that the OSEMs (not installed on this model) would add another 0.5 kg.  From the specs in the final SOS design document, LIGO-T970135, I added 0.25 kg for the optic itself; I am therefore taking the total payload mass to be 11.75 kg.  (Now, the upper stage of the Stewart platform itself will likely add a nontrivial amount, but I am not worrying about this yet.)

I have e-mailed Janeen Romie to obtain the actual center of mass and principal moments of inertia of the platform.  I also cooked up a simple scheme to measure both quantities, should this information not be available.  It would involve rigidly mounting the dirty SOS to a rigid bar hung from a pivot.  By translating the mount point in two dimensions and measuring the period of the pendulum, I ought to be able to find the center of mass and moments of inertia by multilinear regression.  However, this elaborate scheme is not necessary to just compute some ballpark figures; it could wait until a later stage in the design.  For the time being, I just rescaled the moment of inertia proportional to the increase in mass, such that the torque-to-force ratio is unchanged.

As such, the design requirements are now 

• linear travel: 40 microns peak to peak (based on SOS design requirements in LIGO-T950011)
• angular travel: 3 mrad peak to peak (based on SOS design requirements in LIGO-T950011)
• payload mass: 11.75 kg (measured unloaded mass, plus educated guesses about combined mass of OSEMs and optic)
• payload moment of inertia: 0.0232 kg m^2 (wild guess)
• bandwidth: 500 Hz (suggestion of Rana and Koji: ~kHz)

From these assumptions, the revised actuator requirements and dimensions are:

• peak actuator force: 2.04 kN
• minimum radius of top platform: 15 cm
• minimum radius of bottom platform: 30 cm
• minimum height: 26 cm

See the attached PDF document.

It appears that the actuator that I had originally nominated, PI's model P-225.80, would very nearly meet the actuator force requirement.  Steve also pointed out the following single-axis shakers that are already in use in the 40m:

• Brüel & Kjær type 4809
• Brüel & Kjær type 4810

I want to find out if either of these would meet the present need, but I'm waiting on a response from the manufacturer to get access to the data sheets.

 I checked on the two single-axis shakers that are present at the 40m that Steve pointed out:

• Brüel & Kjær type 4809, rated for 45 N peak, and
• Brüel & Kjær type 4810, rated for 10 N peak.

Neither of these meet the force requirement of 2.04 kN peak.

I don't know what exactly is going on, but MC became flaky and it's been frequently unlocked.

I have turned off the MC WFS servo to check if the WFSs are doing something bad. But it still tends to be unlocked without the WFS servo.

Right now it doesn't stay locked for more than 10 min.

 Time to lower your expectations!

Do we really need 40 microns at 500 Hz? Or perhaps should there be a frequency dependent displacement requirement?

       Locking plan            

Here is a plan in my mind and these are basically the details of the gantt chart (#6143):

• (1 day task) Measurement of the recycling gains of the RF sidebands with the PRMI and DRMI configuration, using POP22/110 RFPD.
  • I need to have confidence that I am really locking the DRMI with SRC resonating to 55 MHz.
  • Also those values will enable us to estimate losses and mode matching again (maybe ?).
• (3-4 days task) Measurement of the sensing matrix using the multiple-LOCKIN system.
  • Write a script to automatically measure the sensing matrix. This must be easy.
    • The results will enable us to diagonalize the input matrix and therefore it eventually gives more solid lock of the DRMI
    • Also it will give us the optical gains of 3f signals. So this is actually a step toward the 3f signal check.
• (3-4 days task) Noise budgeting on the 3f signals
  • This is a very important part of the DRMI characterization because the results will tell us whether we can hold the DRMI lock with a sufficient SNR or not.
  • If it turns out that they don't have good SNRs, we then have to come up with some ideas to improve the SNRs.
• (Extra fun task depending on schedule) 3f DRMI lock + Y arm ALS
  • If the beat-box electronics are not available by the time when the work above are completed, I will do this fun task.
  • Probably it is better to start preparing the common mode servo electronics because it will be needed anyway.

For systems using the Spectracom IRIG-B cards for timing information, the code did not properly roll over the time for
2012 (still thinks it is 2011 and get reports from DAQ of timing errors (0x4000)). I have made a temporary fix for this
in the controller.c code in branch-2.3, branch-2.4 and release 2.3.1. 

 The problem is the same as yesterday.

The dynamic power normalization system has been modified such that the normalization happen after the LSC input matrix.

Communication between the front end models and the framebuilder has been restored.  I'm not sure exactly what the issue was, but rebuilding the framebuilder daqd executable and restarting seems to have fixed the issue.

I suspect that the problem might have had to do with how I left things after the last attempt to upgrade to RCG 2.4.  Maybe the daqd that was running was linked against some library that I accidentally moved after starting the daqd process.  It would have kept running fine as was, but if the process died and was attempted to be started again, it's broken linking might have kept it from running correctly.  I don't have any other explanation.

It turns out this was not (best I can tell) related to the new year time sync issues that wer seen at the sites.

Dataviewer is recovered after heavy new year party.

RTS/RCG/DAQ UPGRADE TO COMMENCE

I will be attempting (again) to upgrade the RTS, including the RCG and the daqd, to version 2.4 today.  The RTS will be offline until further notice.

After running into more problems with the upgrade, I eventually decided to abort todays upgrade attempt, and revert back to where we were this morning (RTS 2.1).  I'll try to follow this with a fuller report explaining what problems I encountered when attempting the upgrade.

However, when Alex and I were trying to figure out what was going wrong in the upgrade, it appears that the fb symmetricom card lost the ability to sync with the GPS receiver.  When the symmeticom module is loaded, dmesg shows the following:

[  285.591880] Symmetricom GPS card on bus 6; device 0
[  285.591887] PIC BASE 2 address = fc1ff800
[  285.591924] Remapped 0x17e2800
[  285.591932] Current time 947125171s 94264us 800ns 
[  285.591940] Current time 947125171s 94272us 600ns 
[  285.591947] Current time 947125171s 94280us 200ns 
[  285.591955] Current time 947125171s 94287us 700ns 
[  285.591963] Current time 947125171s 94295us 800ns 
[  285.591970] Current time 947125171s 94303us 300ns 
[  285.591978] Current time 947125171s 94310us 800ns 
[  285.591985] Current time 947125171s 94318us 300ns 
[  285.591993] Current time 947125171s 94325us 800ns 
[  285.592001] Current time 947125171s 94333us 900ns 
[  285.592005] Flywheeling, unlocked...

Because of this, the daqd doesn't get the proper timing signal, and consequently is out of sync with the timing from the models.

It's completely unclear what caused this to happen.  The card seemed to be working all day today, then Alex and I were trying to debug some other(maybe?) timing issues and the symmetricom card all of a sudden stopped syncing to the GPS.  We tried rebooting the frame builder and even tried pulling all the power to the machine, but it never came back up.  We checked the GPS signal itself and to the extend that we know what that signal is supposed to look like it looked ok.

I speculate that this is also the cause of the problems were were seeing earlier in the week.  Maybe the symmetricom card has just been acting flaky, and something we did pushed it over the edge.

Anyway, we will try to replace it tomorrow, but Alex is skeptical that we have a replacement of this same card.  There may be a newer Spectracom card we can use, but there may be problems using it on the old sun hardware that the fb is currently running on.  We'll see.

In the mean time, the daqd is running rogue, off of it's own timing.  Surprisingly all of the models are currently showing 0x0 status, which means no problems.  It doesn't seem to be recording any data, though.  Hopefully we'll get it all sorted out tomorrow.

Both the c1scx and its IOP realtime processes became out of sync.

Initially I found that the c1scx didn't show any ADC signals, though the sync sign was green.

Then I software-rebooted the c1iscex machine and then it became out of sync.

For tonight this is fine because I am concentrating on the central part anyway.

Alex and I have taken the framebuilder offline to try to see what's wrong with the symmetricom card.  We have removed the card from the chassis and Alex has taken it back to downs to do some more debugging.

We have been formulating some alternate methods to get timing to the fb in case we can't end up getting the card working.

The framebuilder is back online now, minus it's symmetricom GPS card.  The card seems to have failed entirely, and was not able to be made to work at downs either.  It has been entirely removed from fb.

As a fall back, the system has been made to work off of the system NTP-based time synchronization.  The latest symmetricom driver, which is part of the RCG 2.4 branch, will fall back to using local time if the GPS synchronization fails.  The new driver was compiled from our local checkout of the 2.4 source in the new to-be-used-in-the-future rtscore directory:

controls@fb ~ 0$ diff {/opt/rtcds/rtscore/branches/branch-2.4/src/drv/symmetricom,/lib/modules/2.6.34.1/kernel/drivers/symmetricom}/symmetricom.ko
controls@fb ~ 0$ 

The driver was reloaded.   daqd was also linked against the last running stable version and restarted:

controls@fb ~ 0$ ls -al $(readlink -f /opt/rtcds/caltech/c1/target/fb/daqd)
-rwxr-xr-x 1 controls controls 6592694 Dec 15 21:09 /opt/rtcds/caltech/c1/target/fb/daqd.20120104
controls@fb ~ 0$ 

We'll have to keep an eye on the system, to see that it continues to record data properly, and that the fb and the front-ends remain in sync.

The question now is what do we do moving forward.  CDS is not supporting the symmetricom cards anymore, and have moved to using Spectracom GPS/IRIG-B cards.  However, Downs has neither at the moment.  Even if we get a new Spectracom card, it might not work in this older Sun hardware, in which case we might need to consider upgrading the framebuilder to a new machine (one supported by CDS).

I mounted the new projector to the pipe where the old projector was attached. The mounting hardware wasn't designed for attaching to a pipe, but with Steve's help I mounted the projector. The projector is angled away from the area of the wall designated as the screen, and I am going to meet with Steve Monday to fix this.

 You (Jamie) should talk with Rolf and Alex to find out what framebuilder they will support for > 3 years. Then we should buy that along with the adapter card which allows us to use the same RAID we now have for the frames.

ETMX sus damping restored

No we can't do that because the c1scx model is not working properly.

If you look into the real time controller screen you will find what I mean.

[John / Kiwamu]

 We found that some of the suspensions channels (for example C1:SUS-BS_POS_IN1 and etc) were not accessible from dataviewer for some reasons.

So far it seems none of the channels associated with c1sus are accessible from dataviewer.

[John / Kiwamu]

 We tried to figure out what is causing spikes in the REFL33 signal, which is used to lock PRCL.

No useful information was obtained tonight and it is still under investigation.

(Background)

 One thing preventing us from doing smooth measurements of the noise budget and the sensing matrix is some sharp spikes in the LSC error signals.

For example when we lock PRMI with REFL33 and AS55 fedback to PRCL and MICH respectively, both the REFL33 and AS55 signals show some spikes in time series.

Those spikes then bring the noise spectra higher than how they should be.

So for the reason, taking the noise budget doesn't give us much information about the interferometer rather than there are spikes.

Also the sensing matrix measurement has been suffered from those spikes, which excite the impulse responses of the low pass filters in the LOCKIN detection systems a lot.

(What we did)

 We looked into the actual analog signals to see if there are indeed spikes or not before they are acquired to the ADCs.

But we didn't find any corresponding spikes in the signals that are after the mixers.

It maybe because the signals we looked into didn't have high enough SNR because they were coming out from the monitor lemo outputs on the demod boards.

 Then we thought the spikes are from the whitening circuits, due to some kind of saturation.

We decreased the gain of the whitening filters by a factor of 10, but it didn't help and the spikes were still there.

I finished mounting the new projector. The projector and computer monitor now display information.

[John / Valera / Kiwamu]

 We installed a new weapon, an optical spectrum analyzer in the AS port.

Like we used to do in the old days, two BNC cables were newly laid down and they bring the output of the OSA to the control room to monitor the spectrum with an oscilloscope.

(Some notes)

The photo diode of the OSA was replaced by a Thorlab PDA100A to amplify the signals.

The carrier peak is at about 6.9 V and the f1 and f2 sidebands peaks are at about 40 mV when the beam is in straight shot (everything is misaligned except ITMY and BS).

According to a rough calculation, those numbers correspond to a modulation depth of about 0.16 or so.

The depth agree with what Mirko measured before (#5519)

On Tue, Dec 20, 2011 at 11:37 PM, Rana Adhikari wrote:
Steve,

We are thinking of replacing a few metal parts in the suspension with sapphire/ruby. We want to replace the standoff (which is Al) with sapphire and also the top plate where the wire is clamped.

For the top, we should make a cutout for a little plate in the existing crossbar. In the cutout would go a little mounting plate. Then the clamping plate (which is now steel) we would also replace with a Sapphire one.

I don't think it matters whether its sapphire or ruby, just has to be very hard.

Can you please get some quotes on the parts?

rana

 The present plan to go with clear cast acrylic plexiglass 1" thick side wall  and  two  clear  1/2 thick top cover.

The inside would be lined with light braun YAG safety window sheets 0.14"  VLT ~25%  OD 4 @ 532nm & OD 5 @ 1064nm

I am trying to stitch together spectra from seismometers and accelerometers to produce a ground motion spectrum from Hz to 100's of Hz.  I was able to retrieve data from two seismometers, GUR1 and STS_1, but not from any of the accelerometers.  The GUR1 spectrum is qualitatively similar to other plots that I have seen, but the STS_1 spectrum looks strange: the X axis spectrum is falling off as ~1/f, but the Y and Z spectra are pretty flat.  All three axes have a few lines that they may share in common and that they may share with GUR1.

See attached plot.

c1iscex is working as before and the optic is damped.

What I checked

1. I went to the X-end rack. I found the io-chassis was turned off.

2. I shutdown c1iscex, turned off, and turned on everything. Again, we did not have any signal from the ADC into c1scx model.

However, I found that c1x01 indicates healthy ADC signals.
This means that the connection between the IOP and the c1scx model was wrong ==> Simulated Plant

3. Burtrestored X'mas eve snapshot. This restored the gains and matrices as well as C1:SCX-SIM_SWITCH
which switches the input between the real ADCs and simulated plant.

4. The signals came back to c1scx.

Below are revised design parameters for the Stewart platform based on ground motion measurements.

The goal is that the actuators should be able to exceed ground motion by a healthy factor (say, two decades in amplitude) across a range from about .1 Hz to 500 Hz.  I would like to stitch together data from at least two seismometers, an accelerometer, and (if one is available) a microphone, but since today this week I was only able to retrieve data from one of the Guralps, I will use just that for now.

The spectra below, spanning GPS times 1010311450--1010321450, show the x, y, and z axes of one of the Guralps.  Since the Guralp's sensitivity cuts off at 50 Hz or so, I assumed that the ground velocity continues to fall as f^-1, but eventually flattens at acoustic frequencies.  The black line shows a very coarse, visual, piecewise linear fit to these spectra.  The corner frequencies are at 0.1, 0.4, 10, 100, and 500 Hz.  From 0.1 to 0.4 Hz, the dependence is f^-2, covering the upper edge of what I presume is the microseismic peak.  From 0.4 to 10 Hz, the fit is flat at 2x10^-7 m/s/sqrt(Hz).  Then, the fit is f^-1 up to 100 Hz.  Finally, the fit remains flat out to 500 Hz.

Outside this band of interest, I chose the velocity requirement based on practical considerations.  At high frequencies, the force requirement should go to zero, so the velocity requirement should go as f^--2 or faster at high frequencies.  At low frequencies, the displacement requirement should be finite, so the velocity requirement should go as f or faster.

The figure below shows the velocity spectrum extended to DC and infinite frequency using these considerations, and the derived acceleration and displacement requirements.

As a starting point for the design of the platform and the selection of the actuators, let's assume a payload of ~12 kg.  Let's multiply this by 1.5 as a guess for the additional mass of the top platform itself, to make 18 kg.  For the acceleration, let's take the maximum value at any frequency for the acceleration requirement, ~6x10^-5 m/s², which occurs at 500 Hz.  From my previous investigations, I know that for the optimal Stewart platform geometry the actuator force requirement is (2+sqrt(3))/(3 sqrt(2))~0.88 of the net force requirement.  Finally, let's throw in as factor of 100 so that the platform beats ground motion by a factor of 100.  Altogether, the actuator force requirement, which is always of the same order of magnitude as the force requirement, is

(12)(1.5)(6x10^-5)(0.88)(100) ~ 10 mN.

Next, the torque requirement.  According to <http://www.iris.edu/hq/instrumentation_meeting/files/pdfs/rotation_iris_igel.pdf>, for a plane shear wave traveling in a medium with phase velocity c, the acceleration a(x, t) is related to the angular rate W'(x, t) through

a(x, t) / W'(x, t) = -2 c.

This implies that |W''(f)| = |a(f)| pi f / c,

where W''(f) is the amplitude spectral density of the angular acceleration and a(f) of the transverse linear acceleration.  I assume that the medium is cement, which according to Wolfram Alpha has a shear modulus of mu = 2.2 GPa and about the density of water: rho ~ 1000 kg/m³.  The shear wave speed in concrete is c = sqrt(mu / rho) ~ 1500 m/s.

The maximum of the acceleration requirement graph is, again, 6x10^-5 m/s² at 500 Hz.^.  According to Janeen's SolidWorks drawing, the largest principal moment of inertia of the SOS is about 0.26 kg m².  Including the same fudge factor of (1.5)(100), the net torque requirement is

(0.26) (1.5) (6x10^-5) (100) pi (500) / (1500) N m ~ 2.5x10^-3 N m.

The quotient of the torque and force requirements is about 0.25 m, so, using some of my previous results, the dimensions of the platform should be as follows:

radius of top plate = 0.25 m,

radius of bottom plate = 2 * 0.25 m = 0.5 m, and

plate separation in home position = sqrt(3) * 0.25 m = 0.43 m.

One last thing: perhaps the static load should be taken up directly by the piezos.  If this is the case, then we might rather take the force requirement as being

(10 m/s²)(1.5)(12 kg) = 180 N.

An actuator that can exert a dynamic force of 180 N would easily meet the ground motion requirements by a huge margin.  The dimensions of the platform could also be reduced.  The alternative, I suppose, would be for each piezo to be mechanically in parallel with some sort of passive component to take up some of the static load.

I had been thinking of using this flexure for the bearings for the leg joints, but I finally realized that it was not the right type of bearing.  The joints for the Stewart platform need to be free to both yaw and pitch, but this bearing actually constrains yaw (while leaving out-of-plane translation free).

I stole the foam EOM box and the temperature controller circuit from the PSL table so that I could continue with the RAM measurements here at the ATF.

That is all.

We can account for delays in the oaf system by compensating it in the adaptive path of the filter. But using only this procedure is not enough. Parameters mu and tau should be chosen accurately:

w = (1 - tau) * w;

w += mu * dw / norm;

NLMS algorithm without considering delays works well for mode cleaner length and gur1 seismometer signals, significantly reducing MC_F  with parameters mu=1, tau=0. These parameters are considered because nlms algorithm should converge with the highest speed when mu=1. However, if the system has a delay so at time moment n:

error_signal [n] = desired_signal [n] - filter_output [n-delay];

then the adaptive filter diverges for the same parameters mu=1 and tau=0 even for delay=1. For that reason we make the same calculations with tau = 1e-4 and tau = 1e-2 without reducing mu conserving the adaptation rate and get the same result as nlms algorithm without delays. Next figure shows MC_F signal, error after applying e-nlms filter with tau=1e-4 and tau=1e-2. "e-" is added to show that  a small number (epsilon) is added to the norm of the signal in order to prevent the filter from diverging in the beginning of the process when the norm is not well-determined yet.

The test was done offline with the sampling frequency 2048 Hz, without downsampling and any filters. We can see that tau=1e-4 is still not enough, tau=1e-3 or tau=1e-2 is as good as nlms without delays, tau=1e-1 and high are also bad.

Correctly choosing tau we have some freedom for delay compensation in the adaptation path. This is important as we do not know exactly what is the delay in the real system. We can measure it approximately. In order to figure out the range of reasonable delay errors we make a test with delay = 1, but to the adaptation path we give delays from 0 to 10. It turns out that adaptation path delays greater then 5 make the filter diverge, delays in the range 0-3 produce a reasonable error. In the figure below errors with adaptation path delays = 1 (correct) and 3 are presented.

Here for the downsampling process we use a low-pass Bessel digital filter of order 6, normalized cut-off frequency = 0.1. In the plot presented below we compare the results with downsampling ratio = 1, 2, 4.

We can see that increasing the downsampling ratio, we increase the error of the filter. Moreover, the error at some particular frequency f seems to depend on the ratio f/Fs, where Fs - sampling frequency (2048 Hz) devided by downsampling ratio. Error is the same for all curves below 1 Hz but then begins to increase as we increase the sownsampling ratio. In order to figure out what the problem is - mistake in the filter code, inaccurate upsample algorithm or this is NLMS particularity, I've changed sampling frequency in the chans/daq/C1PEM.ini and C1IOO.ini files from 2048 Hz to 512 Hz for corresponding channels. Now, we compare the error from the filter working with 2048 Hz frequency, downsampling ratio = 4, low-pass filter = Bessel of order 6, normalized cut-off = 0.1 and filter working with 512 Hz sampling frequency, without downsampling and with corresponding Bessel low-pass filter with normalized frequency 0.4.

MC_F measurement at 2048 Hz was done during the day, for that reason red curved is slightly higher then green in the resonance frequencies. But still we can see that these two cases are very much alike. For this reason, it seems that NLMS filter works better with higher sampling frequencies.

 In order to figure out what downsampling ratio we can take, we need to determine the running time of the fxlms_filter() function. If the filter length is equal to 5000, downsampling ratio is equal to 1, number of witness channels is 1 then with ordinary compilation without speed optimization one call runs for 0.054 ms (milli seconds). The test was done on the 3 GHz Intel processor. With speed optimization flags the situation is better

-O1 0.014 ms, -O3 0.012 ms

However, Alex said that speed optimization is not supported at RCG because it produce unstable execution time for some reason. However, by default the kernel should optimize for size -Os. With this flag the running time is also 0.012 ms. We should check if the front-end machine compilers indeed use -Os flag during the compilation and also play with speed optimization flags. Flags -O3 and -Os together might also give some speed improvement.

But for now we have time value = 0.012 ms as running time for 5000 coefficient filter, 1 witness channel and downsample ratio = 1. Now, we need to check how this time is scaled if we change the parameters.

5000 cofficients - 0.012 ms

10000 coefficients - 0.024 ms

15000 coefficients - 0.036 ms

20000 coefficients - 0.048 ms

We can see that filter length scaling is linear. Now we check downsampling ratio

ratio=1 - 0.048 ms

ratio=2 - 0.024 ms

ratio=4 - 0.012 ms

Running time on the dependance of downsample ratio is also linear as well as on the dependence of the number of witness channels and degrees of freedom. 

If we want to filter 8 DOF with approximately 10 witness channels for each DOF, then 5000 length filter will make 1 cycle for ~1 ms, that is good enough to make the downsample ratio equal to 4.

Things get a little bit complicated when the code is called for the first time. Some time is taken to initialize variables and check input data. As a result the running time of the first cycle is ~0.1 ms for 1 DOF that is ~10 times more then running time of an ordinary cycle. This effect takes place at this moment when one presses reset button in the c1oaf model - the filter becomes suspended for a while. To solve this problem the initialization should be divided by several (~10) parts.

A very strange thing is going on.

The REFL11 and REFL55 demod signals show high frequency noise depending on how big signals go to the POS actuator of PRM.

This noise shows up even when the beam is single-bounced back from PRM ( the rest of the suspensions are misaligned) and it's very repeatable.

Any idea ?? Am I crazy ?? Is PRM making some fringes with some other optics ??

(background)

(Glitch is related to the PRM POS actuation)

(Possible scenario)

I did a quick test to check a hypothesis that PRM is interfering with some other optics in the single bounce configuration.

I shook all the suspensions (except the MC mirrors) at 3 Hz in POS, PIT and YAW with an amplitude of 1000 counts.

No effects were found in the REFL demod signals.

So it is NOT a fringe effect caused by the other suspended mirrors.

AWG is not working. This needs to be fixed.

I could set the channel and the parameters in the AWGGUI screen, but it never inject signals to the realtime system.

 Last night I found that there were many dust particles on the second steering mirror in the REFL path on the AS table.

Looking at it through an IR viewer, I saw the REFL beam hitting one of the biggest dust particles on that mirror.

This dust particle maybe causing the glitches or maybe not.

Anyway because it's always better to have clean mirrors, I will wipe the steering mirror in this evening and check the presence of the glitches again.

Actually awg works fine without any problems when the excitation channels belong to the c1lsc machine.

It seems that the awg doesn't inject signals on the channels of the c1sus machine, for example C1:SUS-BS_LSC_EXC and so on.

 Another possibility is that there is some beam clipping of the REFL beam before it gets to the PD. Then there could be a partial reflection from that creating a spurious interference. Then it would only show the fringe wrapping if you excite the scatterer or the PRM.

I wiped both surfaces of the REFL second steering mirror.

However no improvements. The glitches still remain.

(Pic.1 before wiping, Pic.2 after wiping)

Transparent- clear plexyglass from tree different sources were measured in 1064 and 532 nm light.

Samples: a, clear Acrylic-GP 0F00  from Ridout Plastics in thickness 0.7" ,  made by  Evonic Ind

                   b, clear cast acrylic from Mc Master Carr in thickness 0.94" , likely  made by Reynolds-Cast

                   c, clear cell cast  plexyglass from Delvie's Plastics - Utah in thickness  0.93" , maker not known

PMC reflected beam was used at 92 mW and 6 mm diameter at incident angle 0-25 degrees.

All tree samples agreed on Transmittance of ~90%, Reflectivity ~3-4% and calculated Adsorption ~6-7%

Transparent Colored Acrylic orange-amber #2422   from www.eplastics.com in 0.12" thickness gave  T 96%,  R 1% and  Ab-calc ~3% in the beam of 92 mW 1064  nm at 6 mm diameter.

Transparent , colored   Light Red #26 thin film filter   policarbonate-polyester   0.002" thick   from Roscolux measured T 81% of 115 mW 1064 nm

Now I changed power meter FieldMate to Ophir and the light source to laser pointer 2.2 mW  ~532 nm  with 1-2 mm beam diameter.

Orange - amber #2422  sample, 0.12" thick,  T 1% ,  R 4%  and  Ab-calculated ~95%, estimated visibility  ~50% It does cut out the green at this low power level.

 Light red #26  sample  T 0.5%  at 2.5 mW of 532 nm . The transparent green is not visible.  The softening point of this sandwiched polycarbonate-polyecter filter is 160C. Estimated VLT of this film ~40%

SUMMERY:

Clear and colored acrylics'  @ 1064 nm  transmittance 90% or higher  regardless of thickness. Softenig point 115 degrees C

Colored acrylic and colored policarbonate film are adsorbing the low power green and they  transmit the 1064nm beam.

Options to consider: a, acrylic laser safety shield liner of  0.125" thick inside of 1" thick clear acrylic  box, OD +5 @1064 and OD +4 @ 532nm,  amber color VLT 27%,  150$/sqft

                                      b, thin metal liner for 1" wall acrylic box, VLT 0%