As you've correctly noted, the source of the C1:GCV-SCX_ETMX_ALS channels is in the c1gcv model. The first 3 letters of the channel name indicate this (GCV).
The destination of this channel is c1scx, the 2nd 3 letters indicate this (SCX). If it passed through the c1rfm model, it would be written like C1:GCV-RFM_ETMX_ALS.
This particular channel doesn't pass through the c1rfm model, because the computers these two run on (c1ioo and c1scx) are directly connected via our old VMIC 5565 RFM cards, and don't need to pass through the c1sus computer. This is in contrast to all communications going to or from the c1lsc machine, since that is only connected the c1sus machine by the Dolphin RFM. The c1rfm also handles a bunch of RFM reads from the mode cleaner WFS, since each eats up 3-4 microseconds and I didn't want to slow the c1mcs model by 24 microseconds (and ~50 microseconds before the c1sus/c1scx computer switch).
So basically c1rfm is only used for LSC communications and for some RFM reads for local suspensions on c1sus.
As for the reason we have no transmission, that looks to be a problem on c1ioo's end. I'm also noticing that MCL is not updating on the MC2 suspension screen as well as no changes to MC PIT and YAW channels, which suggests we're not transmitting properly.
I rebooted the c1ioo machine and then did a burt restore of the c1ioo and c1gcv models. These are now up and running, and I'm seeing both MCL and ALS data being transmitted now.
Its possible that when we were working on the c1gfd (green frequency divider model) on c1ioo machine we disturbed the RFM communication somehow. Although what exactly, I'm not sure.
No signal is transmitted from C1:GCV-SCX_ETMX_ALS (on c1gcv) to C1:GCV-SCX_ETMX_ALS (on c1scx)
I can't find RFM definition for ALS channels in c1rfm. Where are they???
I've had a go at trying to estimate the frequency noise of the digital frequency discriminator (DFD). I input a 234.5Hz (0.5Vpp) signal from a 30MHz function generator into the ADC. The LP output of the DFD measured 234.5Hz. However, this signal is clearly modulated by roughly +/- 0.2Hz at harmonics of 234.5Hz (as you can see in the top plot in the dataviewer screenshot below). So the frequency noise can be estimated as rms of approximately 0.2Hz.
This is supported by taking the spectra of the LP output and looking at the RMS. Most of the power in the RMS frequency noise (above the minimum frequency) comes from the harmonics of the input signal and the RMS is approximately 0.2Hz.
I believe this stems from the rather basic LP filter (three or four poles around 10Hz?) that is used in the LP filter to remove the higher frequency components that exist after the mixing stage. (The currently loaded LPF filter is not the same as the saved one in Foton - and that one won't load at the moment, so I'm forced to remember the shape of the current filter).
The attached screen capture from data viewer shows the LP_OUT hovering around 234.5Hz.
A prototype freq divider has been made which works up to ~40MHz.
74HC4060 (14bit binary ripple counter) divides the freq of the input signal, which is comverted by the comparator LT1016
into the rectangular signal. The division rate is 2^14.
Attachment1: Circuit diagram
Attachment2: Photo, the prototype bread board
Attachment3: Photo, the spectrum of the freq divided output. The 40MHz input has been divided into 2.4k.
There are the 3rd and 5th harmonics seen. The peak was pretty sharp but the phase noise was not evaluated yet.
The circuit was made on the prototype bread board which is apparently unsuitable for RF purposes.
Indeed, it was surprising to see its working up to 40MHz...
In order to increase the maximum freq of the system we need the following considerations
Here is the spectrum of the input into the DFD (a 234.5Hz sine wave, 0.5 Vpp) and the spectrum and RMS of the LP output. The linewidth of the input signal is clearly much less than 0.1Hz, where as the RMS noise (above 2mHz) is approximately 0.2Hz and the main contributions are clearly the harmonics of the 234.5Hz signal.
I added an MEDM screen for the DFD to the GREEN screen. It is displayed in the attached screen shot.
This screen is located in: /cvs/cds/rtcds/caltech/c1/medm/c1gfd/C1GFD_DFD.adl
This is a plot showing the old filters and the new ones we added this morning.
The new ones have a Cheby for AC coupling below 10 Hz and then a 500 Hz LP after the mixer. The LP frequency has been increased so that we can use this signal in a feedback loop to the ETM with a ~100 Hz UGF.
Joe injected a 234.567 etc. Hz sine wave into the excitation channel in the DFD INPUT filter. The spectrum of the output of the LP filter with the new filter is shown below with the RMS calculated from 300Hz down to 1mHz - see first attachment. The RMS is equal to about 2.5Hz. (Incidentally, the RMS is very much higher (slightly less than 400Hz - see second attachment) if you calculate it from 7kHz down to 1mHz).
The freq divider was built and installed in the beat detection path.
Attachment 1: Circuit diagram
Note: I have added 7805/7905 regulators to the circuit as I could not find -5V supply on the 1X1/2 racks.
Attachment 2: Packaging
Attachment 3: Input specification
- The freq divider and Rana's PFD were hooked up to the ADCs. (Attachment 1)
(I leave the analog PFD not explained in this entry.)
For this purpose, the VCO feedback signal has been disconnected and the beat signal was moved from the VCO loop to the analog PFD.
The output level of the splitter was +12dBm and was too high for the freq divider.
So, I had to stupidly add an attenuator of 10dB before the box.
- Gain of the digital PFD LPF
The LPF of the digital PFD had the gain of -4096 to let the output signal indicate the direct frequency reading.
The gain has been changed to -67.108864
such that the output shows the direct reading of the beat freq in the unit of MHz
-4096*2^14/10^6 = -67.108864
- Attachment 2 shows the acquired beat note through the freq divider.
The blue is the beat note between "green locked" and "IR locked only to MC" (i.e. MC vs XARM)
The red is the beat note with the both beam locked to the arm
The freq divider is a bit flaky in some freq region as the divided output sometimes shows freq jumps or the captured at a freq.
I still don't know why it happens. We have to check why this happens.
The freq fluctuation of the beat note has been measured with the following condition
- The output of the freq divider is already calibrated to have the unit of MHz.
- The transfer function between the analog PFD channel and the digital PFD output was measured to be -23dB = 0.7.
The gain of the XARM-FINE channel was changed to 0.7 such that the output is calibrated in MHz.
- I have not checked the analog noise level of the analog PFD path. We may need more whitening gain (by icreasing the gain of SR560).
- The analog PFD is always better than the digital PFD above 20Hz.
- Both the digital and analog PFD showed good agreement below 20Hz.
Note the measurement was not simultaneous.
- When the arm is locked with the ETMX being actuated , the fluctuation of the arm length must be stabilized by a huge factor
(~10^5 according to Kiwamu's entry) However, we only could see the stabilization factor of 30.
As this residual is the difference of the freq noise felt by the IR and the green,
this is a real issue to be tackled.
- The RMS fluctuations of the arm with and without the IR beam being locked are 2MHz and 0.1MHz,
which correcponds to the arm length motion of 250nm and 13nm, respectively.
Ed: I had to use 532nm in stead of 1064nm. The correct numbers are 130nm and 7nm.
- Without the IR locked, The typical peak-to-peak fluctuation of the beat freq was 10MHz.
I found that some flakiness of the beat signals comes from the RF components for the beat detection.
They are touching the racks in an indefinite way. If we move the components the output of the analog PFD
Once Kiwamu is back I will ask him to clean up all of the green setting in an appropriate way.
This measurement pertains to the BL2002 VCO PLL unit.
Our goal is to measure the frequency fluctuations introduced by the VCO.
First the VCO calibration was checked. It is -1.75 MHz per volt. The calibration data is below:
Next we measured the Transfer function between points A and B in the diagram below using the Stanford Research System's SR785. This measurement was done with loop opened just after the 1.9MHz LPF and with the loop closed.
The TF[open] / TF [closed ] gave the total gain in the loop. As shown below:
Green curve is the Transfer Function with the loop open and the red with that of the loop closed.
Gain Shown below is the quotient TF[open]/TF[closed]
c) As can be seen from the graph above the loop gain is >>1 over 0.1 to 300Hz. And hence the frequency noise of the VCO is just the product of the voltage noise and the VCO calibration factor over this range,
d) the noise power at the point B was measured and multiplied by the VCO calibration factor to yield dF(rms)/rtHz:
The green line with dots are the data
The blue line is the rms frequency fluctuation.
This corresponds to a arm length fluctuation of about 20pm.
Here's the reference for the self-reference frequency detection idea. See Figure 2.
I did a quick back of the envelope calculation of the expected green phase change on reflection from the aLIGO ITM.
The phase change per nm, K1 = delta phi/delta Lambda, around 532nm is ~1.5 degrees/nm (from the LMA data) [this number is approximately 100x smaller at 1064nm]
I assumed that very small changes in the thickness of the coating appear equivalent to shifting the spectra for reflection/transmission/phase-change-on-reflection up or down by delta lambda, where
delta Lambda/Lambda = delta h/h
where h is the total thickness of the coating and delta h is the change in the thickness of the coating.
Assume that delta h/h = alpha deltaT, where alpha is the coefficient of thermal expansion and delta T is the change in temperature. (approximately 1K)
Then delta phi = K1* Lambda * alpha * delta T = 1.5 degrees/nm * 532nm * 10^-5 K^-1 * 1.0 K = 8 * 10^-3 degrees.
Assume that 360 degree phase change corresponds to one FSR.
Therefore, the frequency shift due to temperature change in the coating = 8*10^-3/360 * FSR = 2.2 *10^-5 * FSR.
Therefore, the expected frequency shift per degree temperature change = 2.2*10^-5 * FSR [Hz/K]
Yesterday I moved the whole green electronics stuff, which had been sitting on the floor at the X end, into a new electronics rack.
The rack now is placed under the cable rail close to the ETMX chamber.
Using a stray beam that is generated as the transmitted green beam from the Xarm goes through the viewport to the PSL table, I installed a fast lens (because I was constrained for space) and a Thorlabs PDA36 photodiode on the PSL table.
The BNC cable runs along the edge of the PSL table, up the corner hole with the huge bundle of cables, and over to IOO_ADC_0. It's channel 3 on the simulink model, which means that it is plugged into connector #4.
With the green resonating TEM00, I have ~1.4V output from the photodiode, as seen on a voltmeter. This corresponds to ~1500 counts on the MEDM screen.
Note to self: Switch to a ~1cm diode with a boatload of gain (either from the 40m or Bridge), and use transmission through a steering mirror of the actual beat note path, not the jittery viewport pickoff. Want RIN noise level to be about 1e-5, only care about below ~100Hz so don't need broadband.
I repeated the same measurement as that Koji did before (see here) with the mixer-based frequency discriminator.
The frequency fluctuation of the beat note is now 50 kHz in rms integrated down to 0.1 Hz, which is a bit better than before.
However there still is the same undesired structure in the spectrum below 10 Hz.
Fig.1 power spectra of the green beat note fluctuation in terms of frequency fluctuation.
Red curves were taken when the IR was locked to the MC, and the green was locked to the X arm.
Blue curves were taken when both the IR and the green were locked to the X arm.
Black curve was also the one taken when the IR and the green were locked to the X arm, but showing the lower noise level.
I have no idea what exactly was going on when I took the black curve, but this noise level sometimes showed up.
The discrepancy may come from a kind of calibration error although I kept using the same calibration factor to convert the data from count to frequency.
Need more investigations.
Additionally Koji and I took the coherence between the beat fluctuation and the transmitted lights of both the IR and the green.
It showed a strong coherence at 1 Hz, which is one of the dominant noise of the beat note.
This probably indicates that the 1 Hz peak is produced by a coupling from amplitude fluctuation.
For monitoring the green transmitted light, I used the Jenne's PD (see here)
The amounts of the X arm's beam off-centering have been measured by the A2L technique.
So now we are able to start aligning the IR beam axis in a quantitative way.
Since we saw big residual motions at 1 Hz, 16 Hz on both the green beat note signal and the IR PDH signal (see #4268 and #4211),
we are suspecting that these noise come from an angle to length coupling.
In order to minimize the angle to length coupling, one thing we can do is to bring the beam spots to the center of ITMX and ETMX more precisely.
To do it, we have to quantitatively know how well the beam spots are on the center of the optics. Therefore I started measuring the amount of the beam off-centering.
The A2L technique was used to measure the off-centering with the real-time lockin system, which has been recently embedded in the real-time code by Joe (see #4265).
The idea is the same as Yuta did before (see #3863).
But this time the excitation signal from the real-time oscillator was injected directly to the coil matrix on either ITMX or ETMX, at 18.13 Hz with the amplitude of about 400 cnt.
When the IR laser stays locked to the X arm, the LSC feedback signal is demodulated with the oscillator signal.
This demodulated signal gives the amount of the off-centering.
For this purpose I modified Yuta's A2L script such that we can use it also for the X arm.
I obtained the following values:
PIT = -1.61 mm
YAW = -0.918 mm
PIT = -3.76 mm
YAW = -2.24 mm
I used the same calibration factor as that of Koji calculated (see #3020) for MC, in order to convert the results from the coil gain to the off-centering.
These values are consistent with the spots appearing on the CCD monitors.
[Koji and Kiwamu]
We took transfer functions (TF) from the angle excitations at ETMX and ITMX to the green beat note signal (i.e. angle to length TF).
It turned out that the coupling from ETMX_PIT is quite large.
I wonder how f2p of the ETMX changes this coupling. We'll see.
The plot above shows a set of the transfer functions from the angle excitation to the green beat note.
Note that the y-axis has not been calibrated, it is just a unit of counts/counts.
You can see that the TF from ETMX_PIT to the beat (red cruve) is larger than the others by about a factor of 10 over most of the frequency range.
This means that any PIT motions on ETMX can be coupled into the green beat signal somewhat over the wide frequency range.
It looks having a resonance at 1.5 Hz, but we don't exactly know why.
At that time the coil gains on only ITMX were tuned by applying f2p filters, but ETMX wasn't because of a technical reason coming from epics.
- - - - measurement conditions
* PSL laser was locked to X arm by feeding back the IR PDH signal to MC2.
* the green laser was locked to Xarm as usual.
* took the green beat note signal (approximately 0 dBm) into Rana's MFD with the cable length of about 6 m.
* the output from the MFD was connected to XARM_COARSE channel without a whitening filter.
* excitation signal was injected into either ASC_PIT or ASC_YAW. The excitation was Gaussian noise with frequency band of 10 Hz and amplitude of 300 counts.
* only ITMY had the f2p filters, which balance the coil gains all over the frequency.
New noise spectra of the green locking have been updated.
The plot below shows the in-loop noise spectra when the beat signal was fedback to ETMX.
The rms noise integrated from 0.1 Hz to 100 Hz went down to approximately 2 kHz.
The red curve was taken when the beat was controlled only by a combination of some poles sand zeros on the digital filter banks. The UGF was at 40Hz.
This curve is basically the same as that Koji took few weeks ago (see here). Apparently the rms was dominated by the peaks at 16 Hz and 3 Hz.
The blue curve was taken when the same filter plus two resonant gain filters (at 16.5 Hz and 3.15 Hz) were applied. The UGF was also at 40Hz.
Due to the resonant gain filter at 16.5 Hz, the phase margin became less, and it started oscillating at the UGF as shown in the plot.
I did a quick calculation to determine the amount of sideband transmission through the FP cavity.
The modulation frequency of the end PDH is 216kHz. The FSR of the cavity is about 3.9MHz. So the sidebands pick up about 0.17 radians extra phase on one round trip in the cavity compared to the carrier.
The ITM reflectance is r_ITM^2 = 98.5% of power, the ETM reflection is r_ETM^2 = 95%.
So the percentage of sideband power reflected from the cavity is R_SB = r_ITM*r_ETM*(exp(i*0.17) - 1)^2 / (1 - r_ETM*r_ITM exp(i*0.17) )^2 = 0.85 = 85%
So about 15% of the sideband power is transmitted through the cavity. The ratio of the sideband and carrier amplitudes at the ETM is 0.05
So, on the vertex PD, the power of the 80MHz +/-200kHz sidebands should be around sqrt(0.15)*0.05 = 0.02 = 2% of the 80MHz beatnote.
Once we get the green and IR locked to the arm again, we're going to look for the sidebands around the beatnote.
Two different measurement have been performed for a test of the green locking last night.
Everything is getting better. yes. yes.
[ measurement 1 : IR locking]
The X arm was locked by using the IR PDH signal as usual (#4239, #4268) .
The in-loop signal at from the IR path and the out-of-loop signal at from the green beat note path were measured.
Let us look at the purple curve. This is an out-of-loop measurement by looking at the green beat note fluctuation.
The rms down to 0.1 Hz used to be something like 60 kHz (see here), but now it went down to approximately 2 kHz. Good.
This rms corresponds to displacement of about 260 pm of the X arm. This is barely within the line width. The line width is about 1 nm.
[ measurement 2 : green locking]
The motion of the X arm was suppressed by using the green beat signal and feeding it back to ETMX.
After engaging the ALS servo, I brought the cavity length to the resonance by changing the feedback offset from epics.
Then took the spectra of the in-loop signal at the beat path and the out-of-loop signal at the IR PDH path.
Here is a time series of TRX after I brought it to the resonance.
TRX was hovering around at the maximum power, which is 144 counts.
Since I put one more 10:1 filter to suppress the noise around 3 Hz, the rms of the in-loop beat spectrum went to about 1 kHz, which used to be 2 kHz (see #4341).
But the out-of-loop (IR PDH signal) showed bigger noise by a factor of 2 approximately over frequency range of from 2 Hz to 2 Hz. The resultant rms is 2.7 kHz.
The rms is primarily dominated by a peak at 22 Hz (roll mode ?).
I calibrated the IR PDH signal by taking the peak to peak signal assuming the finesse of the cavity is 450 for IR. May need a cooler calibration.
I forgot to mention about the whitening filter for the ALS digital control system.
As usual I used a whitening filter to have a good SNR against ADC noise, but this time our whitening scheme is little bit different from the usual our systems.
I used two ADC channels for one signal and put a digital summing point and digital filters to keep good SNR over the frequency range of interest.
It's been working fine but it's still primitive, so I will study more about how to optimize this scheme.
The diagram above shows our scheme for the signal whitening.
Basically the SNR at DC is bad when we use only a whitening filter as shown on the bottom part of the diagram because the signal is quite tiny at DC.
On the other hand if we take raw signal into ADC as 'DC path' shown above, the SNR is better at DC but not good at intermediate frequencies (30 mHz - 1kHz).
So the idea to keep the good SNR over the frequency range of interest is to combine these 'DC path' and 'AC path' in a clever way.
In our case, the 'DC path' signal is not as good as the 'AC path' signal above 30 mHz, so we cut off those high frequency signals by using a digital low pass filter.
In addition to it, I put a gain of 1000 in order to match the relative gain difference between 'DC path' and 'AC path'.
Then the resultant signal after the summing point keeps the good SNR with a flat transfer function up to 1 kHz.
In this past weekend I replaced a summing amplifier for the end green PDH locking by a home-made summing circuit box in order to increase the control range.
It's been working well so far.
However due to this circuit box, the demodulation phase of the PDH locking is now somewhat different from the past, so we have to readjust it at some point.
However due to this circuit box, the demodulation phase of the PDH locking is now somewhat different from the past, so we have to readjust it at some point.
At the X end station, the voltage going to the NPRO PZT had been limited up +/- 4 V because of the summing amplifier : SR560.
Therefore the laser was following the cavity motion only up to ~ +/- 4 MHz, which is not wide enough. (it's okay for night time)
So we decided to put a passive circuit instead of SR560 to have a wider range.
We made a passive summing circuit and put it into a Pomona box.
The circuit diagram is shown below. Note that we assume the capacitance of the 1W Innolight has the same capacitance as that of the PSL Innolight (see #3640).
The feedback signal from a PDH box goes into the feedback input of the circuit.
Then the signal will be low passed with the corner frequency of 200 kHz because of the combination of RC (where R is 681 Ohm and C is capacitance of the PZT).
Because of this low pass filter, we don't drive the PZT unnecessarily at high frequency.
On the other hand the modulation signal from a function generator goes into the other input and will be high passed by 50 pF mica capacitor with the corner frequency of 200 kHz.
This high pass filter will cut off noise coming from the function generator at low frequency.
In addition to it, the 50 pF capacitor gives a sufficient amount of attenuation for the modulation because we don't want have too big modulation depth.
Here is a plot for the expected transfer functions.
You can see that the modulation transfer function (blue curve) has non-zero phase at 216 kHz, which is our modulation frequency.
The power ratio of the beatnote signal vs. the 216kHz sideband has been measured.
The measured ratio was -55 dB, which is smaller by about 20 dB than Aidan's estimation.
To confirm this fact we should check the modulation depth of the end PDH somehow.
The below is a picture showing the sidebands around the beatnote locked at 66.45 MHz.
Other than the +/-216 kHz sidebands, we can see some funny peaks at +/- 50 kHz and +/-150 kHz
I wonder if they come from the servo oscillation of the MC servo whose UGF is at 24 kHz. We can check it by unlocking the MC.
Can we set up a fiber-PD on the end table to look at the beat between the "end laser IR beam" and the "PSL IR beam fiber-transmitted end beam"?
We should see the same thing on that PD that we see on the green PD (plus any fiber noise and I'm not really sure how much that'll be off the top of my head). If we unlock the lasers from the arm cavity then the free-running noise of the lasers wrt to each other will probably swamp the 50kHz and 150kHz signals. Maybe we could lock the end laser to the free-running PSL by demodulating the beat note signal from the fiber-PD and then we could look for the extra sidebands in the IN-LOOP signal. Then we could progressively lock the PSL to the MC and arm cavity and see if the sidebands appear on the fiber-PD at some point in that process.
It's possible that the 216kHz drive of the PZT on the Innolight is somehow driving up some sub-harmonics in the crystal. I think this is unlikely though: if you look at Mott's measurements of the Innolight PZT response, there are no significant PM resonances at 50 or 150kHz.
Other than the +/-216 kHz sidebands, we can see some funny peaks at +/- 50 kHz and +/-150 kHz.
When Koji and I were massaging the MC, we noticed that the oscillations were at 48.5 kHz. They were pretty huge and are probably what you're seeing on the beat. My guess is that they are the PZT resonances of the PSL 2W NPRO; we need to put a notch in the FSS box - it still has the notch from the old NPRO.
I moved old POX shutter from ITMY optical table to the south end. MEDM POX mechanical shutter screen is now closing the green beam injection into the Y arm.
I kluged in a 40m long bnc cable that Alberto left on the floor for control. It is labelled POX-sht This is a temporary set up.
As previously noted, there are multiple beams coming back from the ETM surface onto the PDH PD on the end table. They are spread out in a vertical pattern. All the spots swing together (as the ETM moves?).
I moved the PDH Green PD on the end table so that it was further away from the Faraday and I added an iris in between the Faraday and the PD. Now only the principle reflection from the ETM is incident on the PD. See attached photos. In order to sneak past the neighbouring optics, I had to steer the beam down a bit, so the PD is now lower than it previously was.
Just FYI: the angle between the returning beams is about 5 or 6 mrad at the PD. Before the beams get to the PD they go through a telescope that de-magnifies the beam by about 5 or 6 times. This implies that the angle between adjacent returning beams from the ETM is around 1 mrad at the ETM.
This does make the position of the spot on the PD more susceptible to the alignment of the ETM - we should use a short focal length lens and image the ETM plane onto the PD.
First image - overview of table
Second image - the three returning beams immediately before the IRIS
Third image - a close up of the IRIS and PDH PD.
Kiwamu and I noticed that there is a ghost beam on the green beam going into the ETM. What we see is some interference fringes on the edge of the transmission of the green beam through the dichroic beam splitter (DCBS). If we look at the reflection from the dichroic beam splitter these are much more pronounced.
The spacing of the fringes (about 2 per 10mm) indicates an angle between the fields of around 0.1 mrad.
We were able to cause significant motion of the fringes by pushing on the knobs of the steering mirrors that steer the beam into the DCBS. A rough calculation of the derivative of optical path difference between the ghost and the primary beam as a function of input angle gives about 15 microns per mrad. What filtering the effect the arm cavity will have on the ghost beam is not immediately clear, but the numbers shouldn't be too difficult to determine.
I somehow screwed up the PDH box at the X end station.
Right now it's not working, so I am going to check and fix it today.
In the last evening I found that one of the gain stages on the PDH box wasn't fully functional.
So I started investigating it and I though it was going to finish soon, but actually it wasn't so easy.
The PDH box has several gain stages. So an input signal goes through a buffer, a filter, a boost and an output buffer stages sequentially.
The boost stage is supposed to have gain of 10, but I found it didn't have such gain.
In fact the gain was something like -30dB which is pretty small. Plus this boost stage was imposing an wired bump on the transfer function around 50 kHz.
I checked the voltages on some components around the boost stage and confirmed there were no strange voltage.
Then I suspected that the op-amp : LF356 had been broken for some reason. So I replaced it by LT1792 to see if it fixes the issue.
Indeed it did make it functional. However after few minutes of the replacement, it went back to the same bad condition.
I have no idea about what was going on at that time. Anyway it needs more careful investigations.
I temporarily put a jumper cable on the board to skip this stage, but now the PDH lock is not healthy at all.
I made a noise budget for the ALS noise measurement that I did a week ago (see #4352).
I am going to post some details about this plot later because I am now too sleepy.
Here I explain how I estimate the contribution from the differential noise shown in the plot on my last entry (#4376) .
According to the measurement done about a week ago, there is a broadband noise in the green beatnote path when both Green and IR are locked to the X arm.
The noise can be found on the first plot on this entry (#4352) drawn in purple. We call it differential noise.
However, remember, the thing we care is the noise appearing in the IR PDH port when the ALS standard configuration is applied (i.e. taking the beatnote and feeding it back to ETMX).
So we have to somehow convert the noise to that in terms of the ALS configuration.
In the ALS configuration, since the loop topology is slightly different from that when the differential noise was measured, we have to apply a transfer function to properly estimate the contribution.
(How to estimate)
It's not so difficult to calculate the contribution from the differential noise under some reasonable assumptions.
Let us assume that the MC servo and the end PDH servo have a higher UGF than the ALS, and assume their gains are sufficiently big.
Then those assumptions allow us to simplify the control loop to like the diagram below:
Since we saw the differential noise from the beatnote path, I inject the noise after the frequency comparison in this model.
Eventually the noise is going to propagate to the f_IR_PDH port by multiplying by G/(1+G), where G is the open loop transfer function of the ALS.
The plot below shows the open loop transfer function which I used and the resultant G/(1+G).
In the curve of G/(1+G), you can see there is a broad bump with the gain of more than 1, approximately from 20 Hz to 60 Hz.
Because of this bump, the resultant contribution from the differential noise at this region is now prominent as shown in the plot on the last entry (#4376).
I am going to post some details about this plot later
[Jenne, Chris, Kiwamu]
A photo diode and an AOM driver have been newly setup on the PSL table to measure the intensity noise coupling to the beat note signal.
We tried taking a transfer function from the PD to the beat, but the SNR wasn't sufficient on the PD. So we didn't get reasonable data.
(what we did)
- put a DCPD after the doubling crystal on the PSL table. The PD is sitting after the Y1 mirror, which has been used for picking off the undesired IR beam.
- installed the AOM driver (the AOM itself had been already in place)
- injected some signals onto the AOM to see if we can see an intensity fluctuation on the PD as well as the beat signal
In order to have better SNR for the intensity measurement, we put an AC coupled SR560 with the gain of 100 just before the ADCs.
When a single frequency signal was applied from a Stanford Research's function generator to the AOM, we could clearly see a peak at the doubled frequency of the injected signal.
Also a peak at the same frequency was found on the beat note signal as well.
But when random noise was injected from the same function generator, the random noise looked below the ADC noise.
Jenne adjusted the output voltage from the PD to about 1 V to avoid a saturation in the analog path, but later we realized that the ADC counts was marely ~ 20 counts.
So we will check the ADC tomorrow. Hopefully we will get a good SNR.
Noise below 10 Hz became larger again compared with the data before (see here #4352)
Note that the Y-axis is in MHz.
Here is a diagram for our intensity noise coupling measurement.
The below is a plot for the intensity noise on the DCPD. (I forgot to take a spectra of the PD dark noise)
For some reason, the RIN spectrum becomes sometimes noisier and sometimes quieter. Note that after 10 pm it's been in the quiet state for most of the time.
An interesting thing is that the structure below 3 Hz looks like excited by motion of the MC when it's in the louder state.
A photo diode and an AOM driver have been newly setup on the PSL table to measure the intensity noise coupling to the beat note signal.
Here is a new plot for the differential noise measurement. I plot a noise contribution from the intensity noise (brown curve).
If we believe this data, the differential noise is NOT dominated by the intensity noise of the PSL.
(intensity noise coupling measurement)
Here is a plot for the transfer functions (TFs) from the intensity noise DCPD to the beat signal.
In principle these TFs tell us how much intensity noise are contributed into the differential noise.
When I measured the spectra shown above, the frequency offset of the beatnote was at about 8 MHz from the zero cross point.
Keeping the same lock, I measured the transfer function (red curve) by using the swept sine technique on DTT. The setup for this measurement is depicted on the last entry (#4389).
Then I made the spectra above by multiplying the intensity spectrum by this TF.
Later I measured another transfer function when the beatnote was at about 2 MHz from the zero cross point.
According to this measurement, our MFD gets insensitive to the intensity noise as the beat offset goes close to the zero cross point. This is consistent with what we expected.
We are limited by the intensity noise of the X arm transmitted green light.
Since the intensity noise from the PSL wasn't big enough to explain the differential noise (#4392), so this time I measured the noise contribution from the X arm transmitted light.
I performed the same intensity noise coupling measurement, but this time between the DC signal of the beatnote RFPD and the beatnote signal.
While measuring it, I excited the intensity of the PSL laser by using the same AOM like I did yesterday. This AM cause the observable intensity noise on the beatnote RFPD.
With the excited AM, we can pretend to have an excited AM on the green transmitted light from the X arm, of course assuming the intensity noise coupling from the PSL is less.
The next steps we should do are :
We can modify the freq divider circuit to make it a comparator.
There are 3 standard techniques to reduce this effect:
1) Stabilize the end laser by sensing the green light coming into the PSL before recombination and feeding back with SR560 (this is the only one that you should try at first).
2) Moving to the center of the MFD fringe via ETM steps.
3) Auto-alignment of the beam to the arm.
Aidan: Joe and I have added a channel that takes the DC output from the vertex beatnote PD and sends it, via RFM, to a DAC at the ETMX end. Immediately before the output is a filter C1GCX_AMP_CTRL. The output of the DAC is connected to the CURRENT LASER DIODE modulation input on the back of the Innolight driver. This will modulate the current by 0.1A/V input.
We should be able to modulate the green laser on the end now and stabilize the intensity of the amplitude on the beatnote PD at the vertex. (In this configuration, the ampltiude noise of the PSL laser will be injected onto the end laser - we may want to revisit that).
Joe's comments on model change:
I added a RFM connection at the output of the C1:GCV-XARM_BEAT_DC filter in the c1gcv model. The RFM connection is called: C1:GCV-SCX_ETMX_AMP_CTRL.
This RFM connection goes to the c1scx model and into Kiwamu's GCX box, which uses top_names. There's a filter inside called AMP_CTRL, so the full filter name is C1:GCX-AMP_CTRL. The output then goes to the 7th DAC output.
The reason that I chose this PD is that, apparently, the green light coming from the cavity is clipped when it is picked off for its DC PD.
Ridiculous and hacky. Digital stabilization removed as well as the old "leave a pile of equipment on a stool" strategy.
We used a a BNC cable to send a pickoff of the beam before the recombination to the end via an SR560.
Prior to the works on the Y end setup I propose to perform the temperature scan business like Koji and Suresh did before (see this entry).
This business will allow us to easily find a beatnote at 532nm after the installation on the Y end.
I guess the right persons for this work are Bryan and Suresh.
Bryan will have a safety guidance from Steve in this after noon. So after that they can start working on it.
/* - - - coarse plan - - - */
* remove Alberto's laser from the AS table
* setup Alberto's laser on the PSL table
* put some stuff such as lenses, mirrors and etc. (Use the IR beam picked off after the doubling crystal for the main laser source)
* mode matching
Which laser are we going to use, Alberto's laser or MOPA laser ?
We use Alberto's laser for the Y end Green Locking.
Which laser are we going to use, Alberto's laser or MOPA laser ?
The reason for using Alberto's laser is that some amount of work has already gone into characterising its phase noise. Ref elog entry 2788
A rough time-table and the various tasks are given below:
Note: 700mW NPRO sitting on AP table (Model No: 126-1064-700, Sl No. 415) = Alberto's laser
Temperature dependence of frequency of Alberto's laser:
a) Shifting Alberto's Laser (AL) to the PSL table and setting up a beat frequency measurement between AL and PSL
b) Determining the frequency vs Temperature curve for the AL
Repositioning the optics on the Y-end table and relocating Alberto's laser ( at this point it will be rechiristened as Y-End-NPRO )