Here the design of the periscope for the 90 deg steering version is posted.

straight version http://nodus.ligo.caltech.edu:8080/40m/2709

 EO resonance in the RC path?

I had to restart the dhcpd server on Ottavia that allows us to talk to the camera.  I then also changed the configuration script on the camera so that it no longer thinks ottavia is 131.215.113.97, but correctly 192.168.113.97.  Overall took 5 minutes.

I also looked up services for Centos 5, and set it using the program serviceconf to start the DHCP server  when Ottavia is rebooted now.  That should head off future problems of that nature.  For reference, to start the dhcp server manually, become root and type "service dhcpd start".

Thanks for the great entry!

In order to make this work for higher frequencies, I would add Hartmut's suggestion of a frequency dividing input stage.  If we divide the input down by 100, the overall range will be about 200MHz, and the noise will be about 20Hz/rtHz.  That might be good enough... but we can hope that the commercial device is lower noise!

I just put a script in the /cvs/cds/caltech/scripts/general/netgpibdata/ directory to control the network analyzer called AG4395A_Run.py .   A section has been added to the wiki with the other GPIB script sections (http://lhocds.ligo-wa.caltech.edu:8000/40m/netgpib_package#AG4395A_Run.py)

Steve and Koji

WE started to build 5 TTs. 4 of them are used in the recycling cavities. One is the spare.

We built the structure and are building the cantilever springs.

PRM was released from the fixuture without any trouble. This was the last magnet gluing until ETMs are delivered.

The below is the up-to-date Jenne stat table.

The clean room is getting too narrow. I am thinking that we should install ITMs to the chamber so that we can accommodate SRM/PRM suspensions.

[Rana, Alberto]

This evening we measured the noise spectrum of the reference cavity PD used in the FSS loop. From that we estimated the transimpedance and found that the PD is shot-noise limited. We also found a big AM oscillation in correspondence of the FSS modulation sideband which we later attenuated at least in part.

1. Supposing that the laser beam hitting the PD was shot noise limited, we measured 1.1V at the DC output. That let us estimate the photocurrent at DC of 20mA, for a 50Ohm output impedance.
2. The shot noise for 20mA is 80 pA/rtHz
3. From the nosie spectrum, we measured 3e-7 v/rtHz at 21.5MHz
4. The impedance at RF is then Z_rf = 3e-7 V/rtHz / 80e-12 pA ~ 4000 Ohm
5. Since the RF path inside the PD has a gain of 10, the transimpedance is ~400Ohm, which is about as we (ie Rana) remembered it to be.
6. The PD seems to be working fine.

Some more words about the RFAM: I noticed that there was an excess RFAM by unlocking the RC and just looking at the RF out with the 50 Ohm input of the scope. It was ~100 mVp-p! In the end our method to minimize the AM was not so sensible - we aligned the waveplate before the EOM so as to minimize the p-pol light transmitted by the PBS cube just ahead of the AOM. At first, this did not minimize the RFAM. But after I got angry at the bad plastic mounting of the EOM and re-aligned it, the AM seemed to be small with the polarization aligned to the cube. It was too small to measure on the scope and on the spectrum analyzer, the peak was hopping around by ~10-20 dB on a few second timescale. Further reduction would require some kind of active temperature stabilization of the EOM housing (maybe a good SURF project!).

For the EOM mount we (meaning Steve) should replace the lame 2-post system that's in there with one of the mounts of the type that is used in the Mach-Zucker EOMs. I think we have spare in the cabinet next to one of the arms.

After the RFAM monkeying, I aligned the beam to the RC using the standard, 2-mirror, beam-walking approach. You can see from the attached plot that the transmission went up by ~20% ! And the reflection went down by ~30%. I doubt that I have developed any new alignment technique beyond what Yoichi and I already did last time. Most likely there was some beam shape corruption in the EOM, or the RFAM was causing us to lock far off the fringe. Now the reflected beam from the reference cavity is a nice donut shape and we could even make it better by doing some mode matching! This finally solves the eternal mystery of the bad REFL beam (or at least sweeps it under the rug).

At the end, I also fixed the alignment of the RFPD. It should be set so the incident angle of the beam is ~20-40 deg, but it was instead set to be near normal incidence ?! Its also on flimsy plastic legs. Steve, can you please replace this with the new brass ones?

I'd like to try installing an updated multi-threaded ezca extension later this week, allowing for 64-bit builds of GDS ezca tools, provided by Keith Thorne.  The code can be found in the LDAS CVS under gds, as well as in CDS subversion repository, located at 

https://redoubt.ligo-wa.caltech.edu/websvn/

Its under gds/epics/ in that repository.  The directions are fairly simple:

1) to install ezca with mult-threading in an existing EPICS installation
-copy ezca_2010mt.tar.gz (EPICS_DIR)/extensions/src
-cd (EPICS_DIR)/extensions/src
-tar -C -xzf ezca_2010mt
-modify (EPICS_DIR)/extensions/Makefile to point 'ezca' at 'ezca_2010mt'
-cd ezca_2010mt
-set EPICS_HOST_ARCH appropriately
-make


With a 30mm PPKTP crystal the conversion efficiency from 1064nm to 532nm is expected to 3.7 %/W.

Therefore we will have a green beam of more than 20mW by putting 700mW NPRO.

Last a couple of weeks I performed a numerical simulation for calculating the conversion efficiency of PPKTP crystal which we will have.

Here I try to mention about just the result. The detail will be followed later as another entry.

The attached figure is a result of the calculation.

The horizontal axis is the waist of an input Gaussian beam, and the vertical axis is the conversion efficiency.

You can see three curves in the figure, this is because I want to double check my calculation by comparing  analytical solutions.

The curve named (A) is one of the simplest solution, which assumes that the incident beam is a cylindrical plane wave.

The other curve (B) is also analytic solution, but it assumes different condition; the power profile of incident beam is a Gaussian beam but propagates as a plane wave.

The last curve (C) is the result of my numerical simulation. In this calculation a focused Gaussian beam is injected into the crystal.

The numerical result seems to be reasonable because the shape and the number doesn't much differ from those analytical solutions.

Question:

Why does the small spot size for the case (A) have small efficiency as the others? I thought the efficiency goes diverged to infinity as the radius of the cylinder gets smaller.

Rana found that we had a frequency counter SR620 which might be helpful for lock acquisition of the green phase lock.

It has a response of 100MHz/V up to 350MHz which is wide range and good for our purpose. And it has a noise level of 200Hz/rtHz @ 10Hz which is 1000 times worse than that Matt made (see the entry).

The attached figure is the noise curve measured while I injected a signal of several 100kHz. In fact I made sure that the noise level doesn't depends on the frequency of an input signal.

The black curve represents the noise of the circuit Matt has made, the red curve represents that of SR620.

I took data for the 2 NPRO PLL using the script I wrote for the AG4395, but it is very noisy above about 1 MHz.  I assume this is something to do with the script (since I am fairly confident we don't have 600 dB response in the PZT), so I will go in tomorrow to more carefully understand what is going on, I may need to include a bit more latency in the script to allow the NA to settle a bit more.  I am attaching the spectrum just to show the incredibly high noise level, 

When I went to use the elog this morning, it wasn't responding.  I killed the process on nodus, and then restarted, per the 40m wiki instructions.

Good point. There is a trick  to avoid a divergence.

Actually the radius of the cylindrical wave is set to the spot size at the surface of the crystal instead of an actual beam waist. This is the idea Dmass told me before.

So that the radius is expressed by w=w₀(1+(L/2Z_R)²)^1/2, where w₀ is beam waist, L is the length of the crystal and Z_R is the rayleigh range.

In this case the radius can't go smaller than w₀/2 and the solution can not diverge to infinity.

Its a good measurement - you should adjust the input range of the 620 using the front panel 'scale' buttons to see how the noise compares to Matt's circuit when the range is reduced to 1 MHz. In any case, we would use it in the 350 MHz range mode. What about the noise of the frequency discriminator from MITEQ?

 Teflon feet removed and heavy brass-delrin pd base installed. Ref-cavity reflected light remains to be beautiful doughnut shape on camera.

I went over to Downs today and was able to secure two 8 core machines, along with mounting rails.  These are very thin looking 1U chassis computers. I was told by Rolf the big black box computers might be done tomorrow afternoon.  Alex also kept one of the 8 core machines since he needed to replace a hard drive on it, and also wanted to keep for some further testing, although he didn't specify how long.

I also put in a request with Alex and Rolf for the RCG system to produce code which includes memory location hooks for plant models automatically, along with a switch to flip from the real to simulated inputs/outputs.

Recorded transfer functions for the 1cm Si-PD as described on p. 2708

for different biases. I put the plots in there, to keep the info in one place,

where the label on the PD case (which Steve made without asking him) points

to.

I talked to some people recently about the fact that the responsivity (A/W) of the PD

changes even at DC for different biases. I tested this again and should be more precise about this:

The first time I observed this was in the transfer functions as shown on p. 2708.

With 'DC' I meant 'low frequency' there, as you can still see an effect of the bias as low as 100kHz.

Then at one point I saw the responsivity changing with bias also at true DC.

However, it turned out that this is only the case if the photocurrent is too high.

If the photocurrent is 4mA, you need 400mV bias to get the max. responsivity.

For 2mA photocurrent, the responsivity is already maximal for 0V bias.

An effect for relative low frequencies remains however.

The DC check of responsivity was done with white light from a bulb.

Kiwamu and I measure the PZT response of the Innolight this evening from 24 kHz to 2MHz.  

We locked the PLL at ~50 MHz offset using the Lightwave NPRO and and swept the Innolight with the network analyzer (using the script I made; it has one peculiar property, but it does work correctly).  

We will post the plot of the Lightwave PZT response tomorrow morning.

**EDIT**: As Koji pointed out, the calibration factor on this plot is WRONG.  See my more recent update for the correctly calibrated plot.

The shape of the TF looks nice but the calibration must be wrong.

Suppose 1/f slope with 10^-4 rad/V at 100kHz. i.e. m_pm = 10/f rad/V
This means m_fm = 10 Hz/V. This is 10^6 times smaller than that of LWE NPRO.

(Edit: Corrected some numbers but it is not significant)

 Koji is absolutely right.  I just double checked my matlab code, and saw that I divided when I should have multiplied.  The correctly calibrated plots are attached here for the Innolight and the lightwave.  Kiwamu and I will measure the amplitude and the jitter today.

Innolight: 100rad/V @ 100kHz  => 1e7/f rad/V => 10MHz/V

LWE: 500rad/V @ 100kHz =>  5e7/f rad/V => 50MHz/V

They sound little bit too big, aren't they?

The Lightwave NPRO should be around 5 MHz/V. 

The Innolight PZT coefficient is ~1.1 MHz/V.

(both are from some Rick Savage LHO elog entries)

just a few infos on Silicon PDs I looked up.

If you want to go beyond the 100MHz achievable with the device I worked on,

the one thing to improve is the opamp, where Steve is trying to find OPA657.

This is a FET with 1.6GHz BWP, minimum stable gain of 7, and 4.8nV/rt(Hz) noise.

Should be ok with 750-1000 Ohm transimpedance.

The other thing you might want to change is the PD

(although it might be the 1cm PD with high bias is as fast as smaller ones with lower bias).

There are two types of other Si diodes at the 40m right now (~3mm):

-Rana and I found a Centronic OSD 15-5T in the old equipment

-Frank gave me a Hamamatsu S1223-01 on a Thorlabs pre-amp device (could be taken out).

The Centronic OSD 15-5T has up to 80pF with 12 V bias according to the datasheet.

The Hamamatsu S1223-01 is stated with 20pF only, but stated to have a max. frequency resp. of 20MHz ('-3db point').

I dont know what this means, as the corner freq. of 10pF into 50Ohm is still 160MHz.

In any case there are faster 3mm types to start with, as for example Hamamatsu S3399 (~ 90$),

which is stated to have the corner at 100MHz with 50 Ohm load.

For this type the stated capacity (20pF) looks consistent with ~100MHz corner into 50 Ohm.

So probably you can get higher BW with this one using much smaller load, as in transimpedance stage.

Steve and Koji

- We removed old ITMX/Y from the chambers. Now they are temporarily placed on the flow table at the end. Steve is looking for nice storages for the 5inch optics.

- We wiped new ITMX/Y by isopropanol as they were dusty.

- We put them into the corresponding towers. Checked the balancing and magnet arrangements with the OSEMs. They were totally fine.

- We clamped the mirrors by the EQ stops. Wrapped the towers by Al foils.

Tomorrow we will put them into the chambers.

 We realized that we had measured the wrong calibration value; we were using the free-running error signal with the marconi far from the beat frequency, which was very small.  When we put the Marconi right at the beat, the signal increased by a factor of ~12 (turning our original calibration of 10 mV/rad into 120 mV/rad).  The re-calibrated plots are attached. 

Innolight 10 rad/V @ 100kHz => 1e6/f rad/V => 1MHz/V

LWE 30 rad/V @ 100kHz => 3e6/f rad/V => 3MHz/V

---------

BTW, don't let me calculate the actuator response everytime.

The elog (=report) should be somewhat composed by the following sections

Motivation - Method - Result (raw results) - Discussion (of the results)

 We measured the Amplitude Modulation response of the PZTs, to find regions with large phase modulation but small amplitude modulation.

We did this by blocking 1 arm of the PLL, feeding the source output of the Network Analyzer into the PZT input of the laser in question, and reading the output of the PD on the NA.  We calibrated by dividing by the DC voltage of the PD (scaled by the ratio of the AC gain to DC gain of the New Focus PD).

The AM response of the Innolight looks fairly smooth up to ~1MHz, and it is significantly below the PM response for most of its range.  The region between 20 and 30 kHz shows very good separation of about 10^3 rad/RIN (and up to 10^5 rad/RIN at ~21.88 kHz, where there is the negative spike in the AM response). The region between 1.5 MHz and 2MHz also looks viable if it is desirable to actuate at higher frequencies.

The Lightwave offers very good AM/PM separation up to about 500 kHz, but becomes quite noisy about 1MHz.

I made a simple PD test circuit which may allow to test PD response up to few 100MHz.

Its not for low noise, only for characterising PD response.

Here is the circuit:

The 2 capacitor values (for bypassing) are kind of arbitrary, just what I found around

(one medium, one small capacity). Could be improved by better RF types (e.g. Mica).

The PD type has no meaning. I put in the Centronic 15-T5 for a start.

The bias can be up to 20V for this diode.

The signal appears across R1. It is small, to make a large bandwidth.

R2 is just for slightly decoupling the signal from the following RF amplifier.

The wire into the RF amplifier is short (~cm). And the amplifier is supposed to have 50 Ohm

input impedance.

I use a mini circuits ZFL 500 here.

power supply for this is 15V.

i just restarted the elog for the third time in the past 12 hours.

I checked the elog.log file to debug the problem. It doesn't contain eveidence of any particular cause, except for png/jpg file uploads happened last night.

I'm not sure we can blame Image Magic again because the last crash seems to be occurred just after an entry with e jpg picture was included in the body of the message. I think Image Magic is used only for previews of attachments like pdfs or ps.

Maybe we should totally disable image magic.

The units on this plot are completely bogus - we know that the thermal noise from the resonant part of the circuit is just V = sqrt(4*k*T*Z) ~ 3nV/rHz. Then the gain of the MAX4107 stage is 10. The output resistor is 50 Ohms, which forms a divide by 2 with the input impedance of the spectrum analyzer and so the bump in the dark noise should only be 15 nV/rHz and not microVolts.

 I was aware of a problem on those units since I acquired the data. Then it wasn't totally clear to me which were the units of the data as downloaded from the Agilent 4395A, and, in part, still isn't.

It's clear that the data was in units of spectrum, an not spectral density: in between the two there is a division by the bandwidth (100KHz, in this case). Correcting for that, one gets the following plot for the FSS PD:

Although the reason why I was hesitating to elog this other plot is that it looks like there's still a discrepancy of about 0.5dBm between what one reads on the display of the spectrum analyzer and the data values downloaded from it.

However I well know that, I should have just posted it, including my reserves about that possible offset (as I'm doing now).

These are the dark noise spectrum that I measured on the 11MHz and 55MHz PD prototypes I modified.

The plots take into account the 50Ohm input impedance of the spectrum analyzer (that is, the nosie is divided by 2).

With an estimated transimpedance of about 300Ohm, I would expect to have 2-3nV/rtHz at all frequencies except for the resonant frequencies of each PD. At those resonances I would expect to have ~15nV/rtHz (cfr elog entry 2760).

Problems:

1. For the 55MHz PD the resonance peak is too small
2. In the 55 MHz: noise is present at about 7MHz
3. In the 11MHz PD there's a lot of noise below 10 MHz.

I have to figure out what are the sources of such noises.

Suggestions?

We tested out the functionality of the EG&G 113 preamp that I found in one of the cabinets. This is one of the ancestors of the SR560 preamp that we are all used to.

It turns out that it works just fine (in fact, its better than the SR560). The noise is below 3nV/rHz everywhere above 30 Hz. The filter settings from the front panel all seem to work well. And the red knob on the front panel allows for continuous (i.e. not steps) gain adjustment. In the high-bandwidth mode (low pass filter at 300 kHz), there is ~35 deg of phase lag at 100 kHz. So the box is pretty fast.

I would easily recommend this above the SR560 for use in all applications where you don't need to drive a 50 Ohm load. Also the battery is still working after 17 years!

There's several more of the this vintage in one of the last cabinets down the new Y-arm.

 Hold on, did the arms get re-baptized?

I was checking into the Guralp situation today. I put the rubber balls underneath the granite block (the Q is too high), but found unfortunately that Jenne's styrofoam box is too short to cover the Guralps on top of the granite. If the box was skinny enough to fit on the block or taller by ~6 inches, it would be perfect. We need some new Seismo boxes.

Here's the story of the Gur2 noise so far. We need to pull out and repair the breakout box.

1) At some point we noticed that the Guralp2 X channel was behaving badly.

2) Steve tried recentering with just a +12V supply - this didn't work. Jenne then centered it using the +/- 12V supply. This was OK.

3) Around noon on March 24, the channel 'goes bad' again.

4) On the afternoon of the 25th, most of the channels go to zero, but the GUR2X channel stays bad. There's NO ENTRY in the elog about this. This is UNACCEPTABLE. Apparently, the seismometers were disconnected without shutting off the power to the box. You MUST elog everything - otherwise, go home and sit on your hands.

5) On the evening of the 31st, Steve turns off the Guralp breakout box. From the trend, you can see that the signals all go to zero at that time.

6) From then until today, there is no noise in the GUR2X channel. From these tests we can guess that the problem is in the GUR2X channel of the breakout box, but not in the AA Chassis or the ADC, since those showed no excess noise with the box turned OFF. Its hard to be sure without elog entries, but I assume that 3/25-3/31 was a 'seismometer disconnected', but 'box on' state.

Large earthquake shakes Baja California, Mexico and 6 over Magnitude 5 aftershakes follow.  The frontend computers are still down since Friday.

Steve and Koji (Friday, Apr 02)

Summary

Intsallation of ITMs are going on. Two new ITMs were placed on the optical table in the vacuum chambers. ITM for the south arm was put at the right place in accordance to the CAD drawing. ITM for the east arm is still at a temporaly place.

Tower placement (10:30-11:30)

- Put the tower on the table at a temporary place such that we can easily work on the OSEMs.

ITM (South arm) (14:00-16:30)

- Put the tower on the table at a temporary place such that we can easily work on the OSEMs.

- Leveled the table approximately.

- Released the EQ stops

- Removed anchors for the OSEM cables as it was too short. The wire distribution will be changed later.

- Put the OSEMs. Adjust the insertion to the middle of the OSEM ranges.

- Clamped the EQ stops again

- Placed the tower to the right place according to the CAD drawing.

- Released the EQ stops again.

- Check the OSEM values. The LL sensor showed small value (~0.5). Needs to be adjusted.

ITM (South) damping adjustment

- Found the signs for the facing magnets are reversed.

- Otherwise it damps very well.

After adding an inductor L=100uH and a resistor R=10Ohm in parallel after the OP547A opamp that provide the bias for the photodiode of REFL11, the noise at low frequency that I had observed, was significantly reduced.

See this plot:

A closer inspection of the should at 11MHz in the noise spectrum, showed some harmonics on it, spaced with about 200KHz. Closing the RF cage and the box lid made them disappear. See next plot:

The full noise spectrum looks like this:

A big bump is present at ~275MHz. it could important if it also shows up on the shot noise spectrum.

This morning the pencil soldering iron of our Weller WD2000M Soldering Station suddenly stopped working and got cold after I turned the station on. The unit's display is showing a message that says "TIP". i checked out the manual, but it doesn't say anything about that. I don't know what it means. Perhaps burned tip?

Before asking Steve to buy a new one, I emailed Weller about the problem.

 The cabling on the seismic stack was rerouted so it could reach the south edge of the table: the cables  were removed from the viton padded clamps and repositioned this morning.

ITM-south tower's earthquake screw viton tips could be a little bit larger. They do not stay in their screw hole after a hard clamping action.

4-40 earthquake screws under the test mass:viton tips can fall out without action, the treads are cross threaded so the screws are wobbling

 There should be a supply of extra tips in the Blue Spinny Cabiney (I can never remember it's French name....)  The drawer is something like the top row of one of the bottom sets of drawers.  You can pick the shape of tip you want, and stick it in.

Albeto and Koji

We took the tip replacement from the blue tower.

I am looking at http://www.cooperhandtools.com/brands/weller/ for ordering the tips.

The burnt one seems to be "0054460699: RT6 Round Sloped Tip Cartridge for WMRP Pencil" We will buy one.

The replaced one is "0054460299: RT2 Fine Point Cartridge for WMRP Pencil" We will buy two.

I like to try this: "0054460999: RT9 Chisel Tip Cartridge for WMRP Pencil" We will buy one.

This morning, at about 12 Koji found all the front-ends down.

At 1:45pm rebooted ISCEX, ISCEY, SOSVME, SUSVME1, SUSVME2, LSC, ASC, ISCAUX

Then I burtestored ISCEX, ISCEY, ISCAUX to April 2nd, 23:07.

The front-ends are now up and running again.

 I restored damping to all SUSes except ITM-east. The ITMX OSEMs are being used in the clean assembly room.

Lately I've been trying to sort out the problem of the discrepancy that I noticed between the values read on the spectrum analyzer's display and what we get with the GPIB interface.

It turns out that the discrepancy originates from the two data vector that the display and the GPIB interface acquire. Whereas the display shows data in "RAW" format, the GPIB interface, for the way the netgpibdata script is written, acquires the so called "error-corrected data". That is the GPIB downloaded data is postprocessed and corrected for some internal calibration factors of the instrument.