40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log, Page 51 of 339  Not logged in ELOG logo
ID Dateup Author Type Category Subject
  2523   Mon Jan 18 23:44:19 2010 kiwamuUpdateElectronicstriple resonant circuit for EOM

The first design of the triple resonant EOM circuit has been done.

If only EOM has a loss of 4 Ohm, the gain of the circuit is expected to be 11 at 55MHz

So far I've worked on investigation of the single resonant circuit and accumulated the knowledge about resonant circuits.

Then I started the next step, designing the triple resonant circuit.

Here I report the first design of the circuit and the expected gain.

 


( What I did )

At first in order to determine the parameters, such as inductors and capacitors, I have solved some equations with numerical ways (see the past entry).

In the calculation I put 6 boundary conditions as followers;

(first peak=11MHz, second peak=29.5MHz, third peak=55MHz, first valley=22MHz, second valley=33MHz, Cp=18pF)

The valley frequencies of 22MHz and 33MHz are chosen in order to eliminate the higher harmonics of 11MHz, and Cp of 18pF represents the capacitance of the EOM.

Basically the number of parameters to be determined is 6 ( L1, L2, ...,), therefore it is completely solved under 6 boundary conditions. And in this case, only one solution exists.

The point is calculation does not include losses because the loss does not change the resonant frequency.

 

whole_circuit.png

( results )

As a result I obtained the 6 parameters for each components shown in the table below.

Cp [pF] 18.1
C1 [pF]  45.5
C2 [pF] 10.0
Lp [uH] 2.33
L1 [uH] 1.15
L2 [uH] 2.33

Then I inserted the loss into the EOM to see how the impedance looks like. The loss is 4 Ohm and inserted in series to the EOM. This number is based on the past measurement .

Let us recall that the gain of the impedance-matched circuit with a transformer is proportional to square-root of the peak impedance.

It is represented by G = sqrt(Zres/50) where Zres is the impedance at the resonance.

 As you can see in the figure, Zres = 6.4 kOhm at 55MHz, therefore the gain will be G=11 at 55MHz.

Essentially this gain is the same as that of the single resonant circuit that I've been worked with, because its performance was also limited mainly by the EOM loss.

 An interesting thing is that all three peaks are exactly on the EOM limited line (black dash line), which is represented by Zres = L/CR = 1/ (2pi f Cp)^2 R. Where R = 4 Ohm.

 designed_circuit.png

( next plan )

There are other solutions to create the same peaks and valleys because of the similar solution.

 It is easy to understand if you put Cp' = Cp x N, the solutions must be scaled like L1'=L1/N, C1'=C1 x N, ...,  Finally such scaling gives the same resonant frequencies.

So the next plan is to study the effect of losses in each components while changing the similar solution.

After the study of the loss I will select an optimum similar solution.

  2524   Tue Jan 19 00:10:44 2010 ranaUpdateElectronicstriple resonant circuit for EOM

Very cool.         

  2525   Tue Jan 19 02:39:57 2010 kiwamuUpdateElectronicsdesign complete --- triple resonant circuit for EOM ---

The design of the triple resonant circuit has been fixed.

I found the optimum configuration, whose gain is still 11 at 55MHz even if there are realistic losses.

As I mentioned in the last entry, there are infinite number of the similar solutions to create the same resonant frequencies.

However owing to the effect of the losses, the resultant gain varies if the similar solution changes

The aim of this study is to select the optimum solution which has a maximum gain ( = the highest impedance at the resonance ).

In order to handle the losses in the calculation, I modeled the loss for both inductors and the capacitors.

Then I put them into the circuit, and calculated the impedance while changing the solutions.

 


 

(method)

1). put the scaling parameter as k in order to create the similar solution.

2). scale the all electrical parameters (L1, L2,...) by using k, so that C1'=C1 x k, L1'=L1/k ,...

3). Insert the losses into all the electrical components

4). Draw the impedance curve in frequency domain.

5). See how the height of the impedance at the resonance change

6). Repeat many time this procedure with another k.

7). Find and select the optimum k

scaling.png

There is a trick in the calculation.

I put a capacitor named Cpp in parallel to the EOM in order to scale the capacitance of the EOM (see the schematic).

For example if we choose k=2, this means all the capacitor has to be 2-times larger.

For the EOM, we have to put Cpp with the same capacitance as Cp (EOM). As a result these two capacitors can be dealt together as 2 x Cp.

So that Cpp should be Cpp = (k-1) Cp, and Cpp vanishes when we choose k=1.

 

The important point is that the scaling parameter k must be greater than unity, that is k > 1.

This restriction directly comes from Cp, the capacitance of the EOM, because we can not go to less than Cp.

If you want to put k < 1, it means you have to reduce the capacitance of the EOM somehow (like cutting the EO crystal ??)

 

(loss model)

I've modeled the loss for both the inductors and the capacitors in order to calculate the realistic impedance.

The model is based on the past measurements I've performed and the data sheet.

   Loss for Capacitor :  R(C) = 0.5 (C / 10pF)^{-0.3} Ohm

   Loss for Inductor :    R(L) = 0.1 ( L / 1uH) Ohm

Of course this seems to be dirty and rough treatment.

But I think it's enough to express the tendency that the loss  increase / decrease monotonically as  L / C get increased.

These losses are inserted in series to every electrical components.

( Note that: this model depends on both the company and the product model. Here I assume use of Coilcraft inductors and mica capacitors scattered around 40m )

 

( results )

The optimum configuration is found when k=1, there is no scaling. This is the same configuration listed in last entry

Therefore we don't need to insert the parallel capacitor Cpp in order to achieve the optimum gain.

The figure below shows the some examples of the calculated impedance. You can see the peak height decrease by increasing the scale factor k.

realistic.png

The black dash line represents the EOM-loss limit, which only contains the loss of the EOM.

The impedance at the resonance of 55MHz is 6.2 kOhm, which decreased by 3% from the EOM-loss limit. This corresponds to gain of G = 11.

The other two peaks, 11MHz and 29.5MHz dramatically get decreased from EOM-loss limit.

I guess this is because the structure below 50MHz is mainly composed by L1, L2, C1, C2.

In fact these components have big inductance and small capacitance, so that it makes lossy.

 

( next step )

The next step is to choose the appropriate transformer and to solder the circuit.

  2526   Tue Jan 19 02:40:38 2010 KojiUpdateElectronicstriple resonant circuit for EOM

The design looks very good. I have some questions.

1. As far as I remember, you've got the gain of slightly worse than 10 for a 55MHz single resonant case. Why your expectation of the gain (G=11) for the highest resonance better than this?

Supposing the loss exists only on the EOM, the other part of the LC network for the triple work as an inductor at the resonant frequency. This is just equivalent as the single resonant case. So the expected gain at 55MHz should coincides with what we already have. Probably, the resistance of 4 Ohm that is used here had too rough precision???

2. How can you adjust the resonances precisely?

Do we need any variable components for Cs and Ls, that may have worse quality than the fixed one, generally speaking.
I myself has no experience that I had to tune the commercial EOM because of a drift or whatever. I hope if you can adjust the resonance with a fixed component it should be fine.

3. Changing Cp. What does it mean?

Do you put additional cap for Cp?

4. The resonances for the lower two look very narrow. Is that fine?

This will show up in a better shape if we look at the transfer function for the gain. Is this right?

If we have BW>100kHz, it is sufficient.

5. Impedance matching for the lower two resonances.

Yep. You know this problem already.

 

  2527   Tue Jan 19 03:04:14 2010 KojiUpdateElectronicstriple resonant circuit for EOM

Self-follow:

I got the answer of Q3 from the follow-up entry.

For Q4, once you get the impedance of the LC network lower than n^2*50, the EOM gain will be quite low. This means that the resonance is anyway narrow.
I did some simple calculation and it shows that the width of the resonance will be 100kHz~500kHz. So it maybe OK.

Quote:

The design looks very good. I have some questions.

1. As far as I remember, you've got the gain of slightly worse than 10 for a 55MHz single resonant case. Why your expectation of the gain (G=11) for the highest resonance better than this?

Supposing the loss exists only on the EOM, the other part of the LC network for the triple work as an inductor at the resonant frequency. This is just equivalent as the single resonant case. So the expected gain at 55MHz should coincides with what we already have. Probably, the resistance of 4 Ohm that is used here had too rough precision???

2. How can you adjust the resonances precisely?

Do we need any variable components for Cs and Ls, that may have worse quality than the fixed one, generally speaking.
I myself has no experience that I had to tune the commercial EOM because of a drift or whatever. I hope if you can adjust the resonance with a fixed component it should be fine.

3. Changing Cp. What does it mean?

Do you put additional cap for Cp?

4. The resonances for the lower two look very narrow. Is that fine?

This will show up in a better shape if we look at the transfer function for the gain. Is this right?

If we have BW>100kHz, it is sufficient.

5. Impedance matching for the lower two resonances.

Yep. You know this problem already. 

 

  2528   Tue Jan 19 03:20:28 2010 KojiUpdateElectronicsdesign complete --- triple resonant circuit for EOM ---

First I was confused, but now I think I understood.

My confusion:
If the k get bigger, L get smaller, C get bigger. This makes R(L) smaller and R(C) smaller. This sounds very nice. But why smaller k is preferable in the Kiwamu's result?

Explanation:
The resultant impedance of the network at a resonance is determined by Zres = L/(R C) or something like that. Here R = R(L)+R(C). (I hope this is right.)

Here larger Zres is preferable. So smaller R is nice.

But If the speed of reduction for R is slower than that of L/C (which is proportional to k^-2), increasing k does not help us to increase of Zres. And that's the case.

This means "if we can put the LC network in the box of EOM, we can do better job!" as we can reduce Cp.

Quote:

scaling.png


   Loss for Capacitor :  R(C) = 0.5 (C / 10pF)^{-0.3} Ohm

   Loss for Inductor :    R(L) = 0.1 ( L / 1uH) Ohm

  2529   Tue Jan 19 03:27:47 2010 kiwamuUpdateElectronicsRe: triple resonant circuit for EOM

1. You are right, the gain for the single resonant circuit was about 9.3 in my measurement.

But the reason why the triple is better than the single resonant circuit comes from the transformer.

The impedance can be degraded by a loss of the transformer, because it got worse after applying the transformer in the past measurement.

Also I definitely confirmed that the circuit had the impedance of 7.2 kOhm at the resonance of 52.9MHz without the transformer.

So it shall give the gain of 12, but did not after applying the transformer.

 

2.  Yes, I think we need some variable components just in case.

 

5.  For the impedance matching, I will select a transformer so that 55MHz is matched. In contrast I will leave two lower resonances as they are.

This is because 11MHz and 29.5MHz usually tend to have higher impedance than 55MHz. In this case, even if the impedance is mismatched, the gain for these can be kept higher than 11.

I will post the detail for this mismatched case tomorrow.

 

Quote:

The design looks very good. I have some questions.

1. As far as I remember, you've got the gain of slightly worse than 10 for a 55MHz single resonant case. Why your expectation of the gain (G=11) for the highest resonance better than this?

Supposing the loss exists only on the EOM, the other part of the LC network for the triple work as an inductor at the resonant frequency. This is just equivalent as the single resonant case. So the expected gain at 55MHz should coincides with what we already have. Probably, the resistance of 4 Ohm that is used here had too rough precision???

2. How can you adjust the resonances precisely?

Do we need any variable components for Cs and Ls, that may have worse quality than the fixed one, generally speaking.
I myself has no experience that I had to tune the commercial EOM because of a drift or whatever. I hope if you can adjust the resonance with a fixed component it should be fine.

3. Changing Cp. What does it mean?

Do you put additional cap for Cp?

4. The resonances for the lower two look very narrow. Is that fine?

This will show up in a better shape if we look at the transfer function for the gain. Is this right?

If we have BW>100kHz, it is sufficient.

5. Impedance matching for the lower two resonances.

Yep. You know this problem already.

 

 

  2530   Tue Jan 19 10:30:29 2010 josephbUpdateComputersBoot fest to restart the computer and c1iscey not responding.

Quote:

Thi afternoon I found that the RFM network in trouble. The frontends sync counters had railed to 16384 counts and some of the computers were not responding. I went for a bootfest, but before I rebooted c1dcu epics. I did it twice. Eventually it worked and I could get the frontends back to green.

Although trying to burtrestore to snapshots taken at any time after last wednesday till today would make the RFM crash again. Weird.

Also, c1iscey seems in a coma and doesn't want to come back. Power cycling it didn't work. I don't know how to be more persuasive with it.

During the testing of Megatron as the controller for ETMY, c1iscey had been disconnected from the ethernet hub.  Apparently we forgot to reconnect it after the test.  This prevented it from mounting the nfs directory from linux1, and thus prevented it from coming up after being shutdown.  It has been reconnected, restarted, and is working properly now.

  2531   Tue Jan 19 12:54:39 2010 AlbertoUpdateABSLMeasurement in progress

A measurement will be running for the next hour. Please be careful.

  2532   Tue Jan 19 16:21:18 2010 AlbertoUpdateABSLWatchdogs not working and then fixed

This afternoon the watchdogs stopped working: they didn't trip when the suspension positions crossed the threshold values.

I rebooted c1susaux (aka c1dscl1epics0 in the 1Y5 rack), which is the computer that runs the watchdog processes.

The reboot fixed the problem.

  2533   Tue Jan 19 23:26:07 2010 kiwamuUpdateElectronicsRe:Re: triple resonant circuit for EOM

Quote:

5.  For the impedance matching, I will select a transformer so that 55MHz is matched. In contrast I will leave two lower resonances as they are.

This is because 11MHz and 29.5MHz usually tend to have higher impedance than 55MHz. In this case, even if the impedance is mismatched, the gain for these can be kept higher than 11.

I will post the detail for this mismatched case tomorrow.

 

Here the technique of the impedance matching for the triple resonant circuit are explained.

In our case, the transformer should be chosen so that the impedance of the resonance at 55MHz is matched.

We are going to use the transformer to step up the voltage applied onto the EOM.

To obtain the maximum step-up-gain, it is better to think about the behavior of the transformer.

When using the transformer there are two different cases practically. And each case requires different optimization technique. This is the key point.

By considering these two cases, we can finally select the most appropriate transformer and obtain the maximum gain.

 

 


( how to maximize the gain ?)

Let us consider about the transformer. The gain of the circuit by using the transformer is represented by

eq1.png         (1)

Where ZL is the impedance of the load (i.e. impedance of the circuit without the transformer ) and n is the turn ratio.

It is apparent that G is the function of two parameters, ZL and n.  This leads to two different solutions for maximizing the gain practically. 

 

matching_edit.png

 

  - case.1 : The turn ratio n is fixed.

In this case, the tunable parameter is the impedance ZL.  The gain as a function of ZL is shown in the left figure above.

In order to maximize the gain, Z must be as high as possible.  The gain G get close to 2n when the impedance ZL goes to infinity.

There also is another important thing; If the impedance ZL is bigger than the matched impedance (i.e. ZL = 50 * n^2 ), the gain can get higher than n.

 

  - case.2 : The impedance ZL is fixed.

In contrast to case1, once the impedance ZL is fixed, the tunable parameter is n. The gain as a function of n is shown in the right figure above.

In this case the impedance matched condition is the best solution, where ZL=50*n^2. ( indicated as red arrow in the figure )

The gain can not go higher than n somehow. This is clearly different from case1.
 

 

( Application to the triple resonant circuit )

Here we can define the goal as "all three resonances have gain of more than n, while n is set to be as high as possible"

According to consideration of case1, if each resonance has an impedance of greater than 50*n^2 (matched condition) it looks fine, but not enough in fact.

For example if we choose n=2, it corresponds to the matched impedance of 50*n^2 = 200 Ohm. Typically every three resonance has several kOhm which is clearly bigger than the matched impedance successfully.

However no matter how big impedance we try to make,  the gains can not be greater than G=2n=4 for all the three resonance. This is ridiculous.

What we have to do is to choose n so that it matches the impedance of the resonance which has the smallest impedance.

In our case, usually the resonance at 55MHz tends to have the smallest impedance in those three. According to this if we choose n correctly, the other two is mismatched.

However they can still have the gain of more than n, because their impedance is bigger than the matching impedance. This can be easily understand by recalling the case1.

 

(expected optimum gain of designed circuit)

 By using the equation (1), the expected gain of the triple resonant circuit including the losses is calculated. The parameters can be found in last entry.

designed_response.png

The turn ratio is set as n=11, which matches the impedance of the resonance at 55MHz. Therefore 55MHz has the gain of 11.

The gain at 11MHz is bigger than n=11, this corresponds to the case1. Thus the impedance at 11MHz can go close to gain of 22, if we can make the impedance much big.

 

  2534   Wed Jan 20 20:11:56 2010 AlbertoUpdateABSLSome preliminary results from measuring PRC's transmissivity for an amplitude modulated beam
Here I'm posting a plot showing a set of measurements that I made in the past few days to determine the absolute length of the PRC cavity.
As in my other AbsL measurements, I inject an auxiliary laser beam into the cavity and look at the transmission. In the PRC case, the beam is injected through the dark port and I look at a pick-off of the REFL beam. The aux laser is phase locked to the PSL beam and I control the differential frequency between the two. The PSL and the aux beam interfere and beat at their differential frequency.
 
The attached plot shows the transmitted power as a function of the beat frequency.
 
Fitting the data with the model would let me determine the cavity length. 
By now I can estimate the length of the PRC at about 2.257m, but it's still a rather approximate value.
I can't provide accurate error bars yet. I need to optimize the measurements to get a more precise value.
 
I will go more through the details of the measurement technique and of the fitting function as soon as I have more definitive results.
 
The data points shown here were taken at different times and not always in optimal alignment condition of the PRC. 
To get a good fit of the data I should have fewer frequency segments, taken in a shorter period of time, and in which the power circulating inside of the cavity (ie SPOB) fluctuates as little as possible.
 
For what regards the time needed for a measurements, I already significantly sped up the measurements (i.e. optimizing the scanning and acquisition GPIB scripts, and fixing a couple of problems with the PDH box used in the PLL), and finally now I can scan several tens of MHz in few minutes.
About the frequency segments, so far they have been determined by two factors
1) Tthe frequency generator in the PLL: the Marconi works as a continuous wave generator only in limited ranges. Switching from one to another brakes the wave in a way that causes the PLL to lose lock.
2) Getting below 18 MHz a series of other beats appear on the PLL photodiode and make the PLL lose lock.
 
For the first problem, I'm thinking of using two Marconis and to mix their signals. I would keep one at 300MHz and I would scan the other from 300MHz to 500MHz. In fat, in that frequency range the Marconi has not discontinuity.
 
To try to avoid the other beats at low frequency, I'm not entirely sure about what to do yet. 
 
To be continued...
Attachment 1: 2010-01-19_PRCtransmissivityVsModel.png
2010-01-19_PRCtransmissivityVsModel.png
  2535   Thu Jan 21 10:09:27 2010 KojiSummaryIOOPhotos of the optical tables

I made a wiki page dedicated for the photos of the optical tables.
The current layouts were uploaded.

http://lhocds.ligo-wa.caltech.edu:8000/40m/Optical_Tables

  2536   Thu Jan 21 10:31:13 2010 KojiUpdateABSLSome preliminary results from measuring PRC's transmissivity for an amplitude modulated beam

Nice and interesting plot.

I suppose slight decrease of the Schnupp asymmetry (in your model) adjusts the discrepancy in the high freq region.
At the same time, it will make the resonance narrower. So you need to put some loss at the recombination (=on the BS)?

...of course these depends on the flatness of the calibration.

  2538   Thu Jan 21 11:08:30 2010 kiwamu, steveUpdateVACDry Pump replacement

 This morning I and Steve replaced  the dry fore pump of TP3, which is located under the y-arm.

After replacing it we confirmed vacuum normal condition. The fore line pressure of TP3 went down to 11 mTorr from 750 mTorr

  Attached picture is new pump after setup.

Attachment 1: DSCN0428.JPG
DSCN0428.JPG
  2539   Thu Jan 21 15:16:16 2010 josephb, kojiSummaryComputersMegatron used to lock Y arm

We succeeded in having a stable single arm (Y) lock using Megatron to replace c1iscey.

Now the lock with megatron is pretty easy. Really. It's very cool.

As we saw the oscillation of the YARM servo, we temporalily increased the gain of TRY filter by a factor of 2 (0.003->0.006). Also decreased the gain of YARM servo by the factor of  2 (1->0.5). This makes the servo gain reduced by a factor of 4 in total. This change seemed to come from the change of the ADC/DAC range.

We finally fixed the hi-gain pd transmission communications from Megatron to the c1lsc by tracking down the correct RFM memory location (which is unhelpfully labeled as a qpd channel in both losLinux and lsc40.m).  The memory location is 0x11a1e0, and is refered to as qpdData[3].

  2540   Thu Jan 21 17:23:30 2010 josephb,alex,kojiHowToComputersRCG code fixes

In order to see the Contec DO-32L-PE Digital output PCIe card with the new controls, we had to add the CDO32 part to the CDS_PARTS.mdl file in control /cds/advLigo/src/epics/simLink/ directory on megatron, as well as create the actual model mdl file (based on cdsDio.mdl) in the control/cds/advLigo/src/epics/simLink/lib directory. 

The CDO32.pm file (in /home/controls/cds/advLigo/src/epics/util/lib) has existed for some time, it was just missing the associated pieces in simlink.  However, Alex also checked out a newer version Dio.pm in the process.  As we are not using this part at this time, it should be fine.

The code now compiles and sees the digital output card.

You need a special care on this block as it turned out that the code does not compiled if the "constant" block is connected to the input. You must use appropriate block such as bitwise operator, as shown below.

Attachment 1: CDO32.png
CDO32.png
  2541   Fri Jan 22 02:54:06 2010 AlbertoUpdateABSLOvernight measurement

I'm leaving a measurement running overnight. It should be done in about one hour.

Tomorrow morning, If you need to use the interferometer, and you don't want to have the auxiliary beam going onto the dark port, you can turn down the flipping mirror and close the NPRO's mechanical shutter.

  2542   Fri Jan 22 12:33:37 2010 josephb, alexUpdateComputersModified CDS_PARTS for Binary output

Turns out the CDSO32 part (representing the Contec BO-32L-PE binary output) rquires two inputs. One for the first 16 bits, and one for the second set of 16 bits.  So Alex added another input to the part in the library.  Its still a bit strange, as it seems the In1 represents the second set of 16 bits, and the In2 represents the first set of 16 bits.

I added two sliders on the CustomAdls/C1TST_ETMY.adl control screen (upper left), along with a bit readout display, which shows the bitwise and of the two slider channels. For the moment, I still can't see any output voltage on any of the DO pins, no matter what output I set.

 

  2543   Fri Jan 22 14:40:49 2010 AlbertoUpdateABSLOvernight measurement

Quote:

I'm leaving a measurement running overnight. It should be done in about one hour.

Tomorrow morning, If you need to use the interferometer, and you don't want to have the auxiliary beam going onto the dark port, you can turn down the flipping mirror and close the NPRO's mechanical shutter.

 This is what I measured last night:

2010-01-21_PRCtransmissivityVsModel.png

 This is not a fit. It's just a comparison of the model with the data.

  2544   Mon Jan 25 11:42:24 2010 josephbUpdateComputersMegatron and BO board

I was talking with Vladimir on Friday discussing the Binary Output board, we looked at the manual for it (Contec DO-32L-PE) and we realized its an opto-coupler isolated open-collector output.  He mentioned they had the right kind of 50 channel breakout board for testing in Riccardo's lab.

This morning I borrowed the 50 channel breakout board from Riccardo's lab, and along with some resistor loads, test the BO board.  It seems to be working, and I can control the output channel on/off state.

  2545   Mon Jan 25 16:30:37 2010 AlbertoUpdateABSL166 MHz sideband turned off

I turned off the modulation at 166MHZ becasue I don't need it if I'm only locking the PRC.

It was introducing extra amplitude modulations of the beam which interfered with the AbsL's PLL photodiode.

I'm going to turn it back on later on.

  2546   Mon Jan 25 16:46:33 2010 AlbertoUpdateABSL166 MHz sideband turned off

Quote:

I turned off the modulation at 166MHZ becasue I don't need it if I'm only locking the PRC.

It was introducing extra amplitude modulations of the beam which interfered with the AbsL's PLL photodiode.

I'm going to turn it back on later on.

 I turned back on the 166MHz modulation just a bit. I set the slider at 4.156.

When it was totally off the MZ seemd quite unhappy.

  2547   Tue Jan 26 03:28:56 2010 ranaUpdateABSL166 MHz sideband turned off

 

 You can turn the 166 off if you want. MZ is unhappy after its turned off, but that's just the thermal transient from removing the RF heat. After a several minutes, the heat goes away and the MZ can be relocked.

One of these days we should evaluate the beam distortion we get in EOMs because of the RF heat induced dn/dT. Beam steering, beam size, etc.

  2548   Tue Jan 26 19:51:44 2010 Sanjit, ranaUpdateAdaptive FilteringOAF details

We turned on the OAF again to make sure it works. We got it to work well with the Ranger as well as the Guralp channels. The previous problem with the ACC is that Sanjit and Matt were using the "X" channels which are aligned the "Y" arm. Another casualty of our ridiculous and nonsensical coordinate system. Long live the Right Hand Rule!!

The changes that were made are:

  • use of RANGER channel (with ACC_MC1_X and/or ACC_MC2_X)
  • mu = 0.01, tau = 1.0e-6, ntaps = 2000, nDown = 16
  • nDelay = 5 and nDelay = 7 both work (may not be so sensitive on delay at low frequencies)
  • Main changes: filter bank on the PEM channels - ASS_TOP_PEM_## filters: 0.1:0, 1:, Notch24, AA32, gain 1
  • Added the AI800 filter for upsampling in MC1 (should not matter)

Other parameters which were kept at usual setting:

  • CORR: AI32, gain = 1
  • EMPH: 0.001:0, AA32, gain = 1
  • ERR_MCL: no filters, gain = 1
  • SUS_MC1: no filter, gain = 1
  • PEM Matrix: All zero except: (24,1), (15,2), (18,3)
  • ADAPT path filter: union of CORR and EMPH filters, gain 1
  • XYCOM switches # 16-19 (last four on the right) OFF 

Screenshots are attached.

Burt snapshot is kept as: /cvs/cds/caltech/scripts/OAF/snaps/ass_burt_100126_211330.snap

taken using the script: /cvs/cds/caltech/scripts/OAF/saveOAF

we should put this in ASS screen.


ERROR Detected in filter ASS_TOP_PEM_24 (RANGER): 1: was actually typed as a 1Hz high pass filter!

(Correcting this one seems to spoil the adaptation)

Possibly this makes sense, we may not want to block witness signals in the 0.1-20 Hz range.


  11:40 PM: Leaving the lab with the OAF running on 5 PEM channels (Ranger + Guralp 1&2  Y & Z). There's a terminal open on op440m which will disable the OAF in ~2.8 hours. Feel free to disable sooner if you need the MC/IFO.

Attachment 1: C1ASS_TOP.png
C1ASS_TOP.png
Attachment 2: C1SUS_SRM_XYCOM1.png
C1SUS_SRM_XYCOM1.png
Attachment 3: Untitled.png
Untitled.png
  2549   Tue Jan 26 20:18:32 2010 ranaConfigurationALARMop540m: alarms and BLRMS and StripTool restored

I turned the StripTool and ALARMS and BLRMS back on on op540m. Looks like it has been rebooted 5 days ago and no one turned these back on. Also, there was a bunch of junk strewn around its keyboard which I restrained myself from throwing in the trash.

The BLRMS trends should be active now.

  2550   Wed Jan 27 11:02:30 2010 AlbertoUpdateABSLPRC Cavity Length
 I fitted the data from scanning the PRC by changing the beat frequency of the auxiliary laser beam with the PSL beam.
The data points that I've taken so far over the entire frequency range (0-300 MHz) are not continuous. For several reasons the PLL was unable to maintain lock for such a large range and I had to break it into smaller segments. The measurements to acquire them stretched over a too long period of time during which the status of the PRC changed.
 
Because of that, before I get a continuous set of data points (perhaps normalized by the circulating power inside of the cavity), I restricted the fit to a 55MHz range around 100MHz. I obtained the following numbers for the fit parameters:
Length PRC = 2.169 +/- 0.007 m
Schnupp Asymmetry: 0.471+/- 0.006 m
 
The fit is shown in the attached plot:
2010-01-21_PRCtransmissivityVsFit.png
When I fit over the entire set of data I get this:
 
2010-01-21_PRCtransmissivity_EntireFreqRange_VsFit.png
 
Length PRC = 2.224 +/- 0.005 m
Schnupp Asymmetry: 0.457+/- 0.005 m
 
The results are different. Evidently I have to improve the measurement. I'm working on it.
 
For posterity:
The function I used to fit the transmitted beat power vs. frequency is the following:
 
E_trans = - t_prm * r_itm * exp(1i*2*wb*l_prc/c) .* sin(wb*l_/c) ./ ( 1 + r_prm * r_itm * exp(1i*2*wb*l_prc/c) .* cos(wb*l_/c)
 
Where wb is the angular frequency of the beat, l_prc and l_ are the length of the PRC and the Schnupp asymmetry, respectively; r_itm, t_itm, r_prm, t_prm are reflectances and transmittances of PRM and ITM; c is the speed of light.
 
  2551   Thu Jan 28 09:14:51 2010 AlbertoConfigurationComputersc1iscey, c1iscex, c1lsc, c1asc rebooted

This morning the LSC scripts wheren't running properly. I had to reboot c1iscey, c1iscex, c1lsc, c1asc .

I burtrestored to Monday January 25 at 12:00. 

  2552   Thu Jan 28 09:17:32 2010 AlbertoUpdateLSC166 Modulation turned off

I temporarily turned off the 166 modulation.

  2553   Fri Jan 29 12:06:58 2010 AlbertoUpdateABSLMeasurement running today at lunch time

I just started a measuremtn that will be running for the next hour or so. Please be careful with the interferometer.

  2554   Fri Jan 29 13:14:49 2010 AlbertoUpdateABSLMeasurement running today at lunch time

Quote:

I just started a measuremtn that will be running for the next hour or so. Please be careful with the interferometer.

Done. IFO available

  2555   Mon Feb 1 18:31:00 2010 SanjitUpdateAdaptive FilteringOAF details

  I tried downsampling value 32 (instead of 16), to see if it has any effect on OAF. Last week I encountered some stability issue - adaptation started to work, but the mode cleaner was suddenly unlocked, it could be due to some other effect too.

One point to note is that different downsampling did not have any effect on the CPU meter (I tried clicking the "RESET" button few times, but no change).

  2556   Mon Feb 1 18:33:10 2010 steveUpdateMOPAVe half the lazer!

The 2W NPRO from Valera arrived today and I haf hidden it somewere in the 40m lab!

 

Rana was so kind to make this entry for me

Attachment 1: inno2w.JPG
inno2w.JPG
Attachment 2: inno2Wb.JPG
inno2Wb.JPG
  2557   Mon Feb 1 21:51:12 2010 SanjitUpdateAdaptive FilteringOAF details

 I tried some combination of PEM channels and filters to improve OAF performance at other frequencies, where we do not have any improvement so far. There is progress, but still no success.

Here are the main things I tried:

For the ACC channels replaced the 0.1 Hz high pass filters by 3Hz high pass and turned off the 1: filter.

Then I tried to incorporate the Z ACC/GUR channels, with some reasonable combination of the others.

The Z axis Guralp and Accelerometers were making OAF unstable, so I put a 0.1 gain in all four of those.

Following the PEM  noise curves Rana has put up, we should probably use

  • two ACC_Y channels (3:0, Notch24, AA32)
  • two GUR_Z channels (filters: 0.1:0, 1:, AA32, gain 0,1)
  • one RANGER_Y, just because it works (0.1:0, 1:, Notch24, AA32)

In the end I tried this combination, it was stable after I reduced the GUR_Z gain, but looked very similar to what we got before, no improvement at 5Hz or 0.5Hz. But there was a stable hint of better performance at > 40Hz.

Possibly we need to increase the GUR_Z gain (but not 1) and try to use ACC_Z channels also. Since we can not handle many channels, possibly using one GUR_Z and one ACC_Z would be worth checking.

  2558   Tue Feb 2 10:38:30 2010 josephbUpdateComputersMegatron BO test

Last night, I connected megatron's BO board to the analog dewhitening board.  However, I was unable to lock the y arm (although once I disconnected the cable and reconnected it back the xy220 the yarm did lock).

So either A) I've got the wrong cable, or B) I've got the wrong logic being sent to the analog dewhitening filters.

During testing, I ran into an odd continuous on/off cycle on one of the side filer modules (on megatron).  Turns out I had forgotten to use a ground input to the control filer bank (which allows you to both set switches as well as read them out), and it was reading a random variable.  Grounding it in the model fixed the problem (after re-making).

 

 

  2559   Tue Feb 2 13:14:09 2010 KojiHowToIOOAnatomy of New Focus Resonant EOM

Joe let me use the resonant EOM for GigE phase camera for a while.
Then, I immediately started to open it :)

it uses the MiniCIrcuits T5-1T transformer and a TOKO RCL variable inductor.

The photos are on the Picasa 40m album.

http://lhocds.ligo-wa.caltech.edu:8000/40m/40m_Pictures

  2560   Tue Feb 2 15:30:03 2010 AlbertoFrogsTreasureWild Oats
FYI. Sitting on the top shelf of George I found an opened jar of raspberry jam and an opened jar of creamy peanut butter. Both are branded Wild Oats Market.
 
Wikipedia:
"Wild Oats Markets was an operator of natural foods stores and farmers markets in North America... Whole Foods officially completed their buyout of Wild Oats on August 27, 2007 [...]"
  2561   Tue Feb 2 16:10:09 2010 steveUpdatePEMconstruction progress next door

CES construction is progressing. The 40m suspensions are bearing well.

atm1, PEM  vs sus plots of 120 days

atm2, big pool walls are in place, ~10 ft east of south arm

atm3, 10 ft east of ITMY

atm4, ~60 ft east of ITMY

atm5, cold weather effect of N2 evaporator tower

Attachment 1: pem120d.jpg
pem120d.jpg
Attachment 2: 02022010.JPG
02022010.JPG
Attachment 3: 02022010b.JPG
02022010b.JPG
Attachment 4: 02022010c.JPG
02022010c.JPG
Attachment 5: icewall.JPG
icewall.JPG
  2562   Tue Feb 2 18:15:47 2010 AlbertoUpdateelogElog restarted it

 Zach made me notice that the elog had crashed earlier on this afternoon. 

I just restarted it with the restarting script.

Instructions on how to run the last one are now in the wiki page. Look on the "How To" section, under "How to restart the elog".

  2563   Tue Feb 2 22:39:12 2010 JenneUpdatePSLIFO isn't playing nice tonight

[Jenne, Kiwamu]

It's been an iffy last few hours here at the 40m.  Kiwamu, Koji and I were all sitting at our desks, and the computers / RFM network decided to crash.  We brought all of the computers back, but now the RefCav and PMC don't want to lock.  I'm a wee bit confused by this.  Both Kiwamu and I have given it a shot, and we can each get the ref cav to sit and flash, but we can't catch it.  Also, when I bring the PMC slider rail to rail, we see no change in the PMC refl camera.  Since c1psl had been finicky coming back the first time, I tried soft rebooting, and then keying the crate again, but the symptoms remained the same.  Also, I tried burt restoring to several different times in the last few days, to see if that helped.  It didn't.  I did notice that MC2 was unhappy, which was a result of the burtrestores setting the MCL filters as if the cavity were locked, so I manually ran mcdown.  Also, the MC autolocker script had died, so Kiwamu brought it back to life.

Since we've spent an hour on trying to relock the PSL cavities (the descriptive word I'm going to suggest for us is persistent, not losers), we're giving up in favor of waiting for expert advice in the morning.  I suppose there's something obvious that we're missing, but we haven't found it yet......

  2564   Wed Feb 3 01:17:19 2010 KojiUpdatePSLIFO isn't playing nice tonight

I checked the situation from my home and the problem was solved.

The main problem was undefined state of the autolocker and the strange undefined switch states, being associated with the bootfest and burtrestore.

- MC UP/DOWN status shows it was up and down. So I ran scripts/MC/mcup and scripts/MC/mcdown. These cleared the MC autolocker status.

- I had a problem handling the FSS. After mcup/mcdown above, I randomly pushed the "enable/disable" buttons and others, and with some reason, it recovered the handling. Actually it acquired the lock autonomously. Kiwamu may have also been working on it at the same time???

- Then, I checked the PSL loop. I disconnected the loop by pushing the "test" button. The DC slider changes the PZT voltage only 0~+24V. This is totally strange and I started pushing the buttons randomly. As soon as I pushed the  "BLANK"/"NORMAL" button, the PZT output got back under the control.

- Then I locked the PMC, MZ, and MC as usual.

Alberto: You must be careful as the modulations were restored.

Quote:

[Jenne, Kiwamu]

It's been an iffy last few hours here at the 40m.  Kiwamu, Koji and I were all sitting at our desks, and the computers / RFM network decided to crash.  We brought all of the computers back, but now the RefCav and PMC don't want to lock.  I'm a wee bit confused by this.  Both Kiwamu and I have given it a shot, and we can each get the ref cav to sit and flash, but we can't catch it.  Also, when I bring the PMC slider rail to rail, we see no change in the PMC refl camera.  Since c1psl had been finicky coming back the first time, I tried soft rebooting, and then keying the crate again, but the symptoms remained the same.  Also, I tried burt restoring to several different times in the last few days, to see if that helped.  It didn't.  I did notice that MC2 was unhappy, which was a result of the burtrestores setting the MCL filters as if the cavity were locked, so I manually ran mcdown.  Also, the MC autolocker script had died, so Kiwamu brought it back to life.

Since we've spent an hour on trying to relock the PSL cavities (the descriptive word I'm going to suggest for us is persistent, not losers), we're giving up in favor of waiting for expert advice in the morning.  I suppose there's something obvious that we're missing, but we haven't found it yet......

 

  2565   Wed Feb 3 07:57:01 2010 steveUpdatePSLPMC transmission is low

The low PMC transmission alarm was on this morning. The PMC alignment needs a touch up.

Attachment 1: pmct40d.jpg
pmct40d.jpg
  2566   Wed Feb 3 09:01:42 2010 robUpdateloreIFO isn't playing nice tonight

Quote:

I checked the situation from my home and the problem was solved.

The main problem was undefined state of the autolocker and the strange undefined switch states, being associated with the bootfest and burtrestore.

- MC UP/DOWN status shows it was up and down. So I ran scripts/MC/mcup and scripts/MC/mcdown. These cleared the MC autolocker status.

- I had a problem handling the FSS. After mcup/mcdown above, I randomly pushed the "enable/disable" buttons and others, and with some reason, it recovered the handling. Actually it acquired the lock autonomously. Kiwamu may have also been working on it at the same time???

- Then, I checked the PSL loop. I disconnected the loop by pushing the "test" button. The DC slider changes the PZT voltage only 0~+24V. This is totally strange and I started pushing the buttons randomly. As soon as I pushed the  "BLANK"/"NORMAL" button, the PZT output got back under the control.

- Then I locked the PMC, MZ, and MC as usual.

Alberto: You must be careful as the modulations were restored.

Quote:

[Jenne, Kiwamu]

It's been an iffy last few hours here at the 40m.  Kiwamu, Koji and I were all sitting at our desks, and the computers / RFM network decided to crash.  We brought all of the computers back, but now the RefCav and PMC don't want to lock.  I'm a wee bit confused by this.  Both Kiwamu and I have given it a shot, and we can each get the ref cav to sit and flash, but we can't catch it.  Also, when I bring the PMC slider rail to rail, we see no change in the PMC refl camera.  Since c1psl had been finicky coming back the first time, I tried soft rebooting, and then keying the crate again, but the symptoms remained the same.  Also, I tried burt restoring to several different times in the last few days, to see if that helped.  It didn't.  I did notice that MC2 was unhappy, which was a result of the burtrestores setting the MCL filters as if the cavity were locked, so I manually ran mcdown.  Also, the MC autolocker script had died, so Kiwamu brought it back to life.

Since we've spent an hour on trying to relock the PSL cavities (the descriptive word I'm going to suggest for us is persistent, not losers), we're giving up in favor of waiting for expert advice in the morning.  I suppose there's something obvious that we're missing, but we haven't found it yet......

 

 

This is a (sort of) known problem with the EPICS computers: it's generally called the 'sticky slider' problem, but of course it applies to buttons as well.  It happens after a reboot, when the MEDM control/readback values don't match the actual applied voltages.  The solution (so far) is just to `twiddle' the problematic sliders/button.  There's a script somewhere called slider_twiddle that does this, but I don't remember if it has PSL stuff in it.  A better solution is probably to have an individual slider twiddle script for each target machine, and add the running of that script to the reboot ritual in the wiki.

  2567   Wed Feb 3 10:46:12 2010 AlbertoUpdateelogelog restarted

 Again, this morning Zach told me that the elog had crashed while he was trying to post an entry.

I just restarted it.

  2568   Wed Feb 3 11:13:15 2010 steveConfigurationGeneralplaned power outage for Sat. Feb 20

The electrical shop has to connect the new power transformer at CES. This means we will have no AC power for ~8 hrs on Saturday, February 20

Is this date good for us to power down ALL equipment in the lab?

Rana:  Yes

  2569   Thu Feb 4 00:59:52 2010 ranaUpdateelogelog restarted

 I restarted the ELOG on NODUS just now. Our attempt to set up error logging worked - it turns out ELOG was choking on the .ps file attachment.

So for the near future: NO MORE .PS files! Use PDF - move into the 20th century at least.

matlab can directly make either PNG or PDF files for you, you can also use various other conversion tools on the web.

Of course, it would be nice if nodus could handle .ps, but its a Solaris machine and I don't feel like debugging this. Eventually, we'll give him away and make the new nodus a Linux box, but that day is not today.

  2570   Thu Feb 4 12:29:04 2010 josephbUpdateComputersMartian IP switch over notes

What is the change:

We will be moving the 131.215.113.XXX ip addresses of the martian network over to a 192.168.XXX.YYY scheme.

What will users notice:

Computer names (i.e. linux1, scipe25/c1dcuepics) will not change.  The domain name martian, will not change (i.e. linux1.martian.).  What will change is the underlying IP address associated with the host names.  Linux1 will no longer be 131.215.113.20 but something like 192.168.0.20.  If everything is done properly, that should be it.  There should be no impact or need to change anything in EPICS for example.  However, if there are custom scripts with hard coded IP addresses rather than hostnames, those would need to be updated, if they exist.

What needs to be done:

Each computer and router will need to either be accessed remotely, or directly, and the configuration files controlling the IP address (and/or dns lookup locations) be modified.  Then it needs to be rebooted so the configuration changes take effect. I'll be making an updated list of computers this week (tracked down via their physical ethernet cables), and next week, probably on Thursday, and then we simply go down the list one by one.

LINUX

For a linux machine, this means checking the /etc/hosts file and making sure it doesn't have old information.  It should look like:

127.0.0.1               localhost.localdomain localhost
::1             localhost6.localdomain6 localhost6

Then change the /etc/sysconfig/network-scripts/ifcfg-eth0 file (or ethX file depending on the ethernet card in question).  The IPADDR, NETWORK, and GATEWAY lines will need to be changed.  You can change the hostname (although I don't plan on it) by modifying the /etc/sysconfig/network file.

The /etc/resolv.conf file will need to be updated with the new DNS server location (i.e. 131.215.113.20 to 192.168.0.20 for example).

SOLARIS

Simlarly to linux, the /etc/hosts file will need to be updated and/or simplified.  The /etc/defaultrouter file will need to be updated to the new router ip.  /etc/defaultdomain will need to be updated.  The /etc/resolv.conf will need to be updated with the new dns server.

vxWorks

Looking at the vxWorks machines, the command bootChange can be used to view or edit the IP configuration.

The following is an example from c1iscey.

-> bootChange

'.' = clear field;  '-' = go to previous field;  ^D = quit

boot device          : eeE0
processor number     : 0
host name            : linux1
file name            : /cvs/cds/vw/pIII_7751/vxWorks
inet on ethernet (e) : 131.215.113.79:ffffff00
inet on backplane (b):
host inet (h)        : 131.215.113.20
gateway inet (g)     :
user (u)             : controls
ftp password (pw) (blank = use rsh):
flags (f)            : 0x0
target name (tn)     : c1iscey
startup script (s)   :
other (o)            :

value = 0 = 0x0

By updating the the host (name of machine where its mounting /cvs/cds from - i.e. linux1), inet on ethernet (the IP of c1iscey) and host inet (linux1's ip address), we should be able to change all the vxWorks machines.

LINUX1

The DNS server running on linux1 will need to be updated with the new IPs and domain information.  The host file on linux1 will also need to be updated for all the new IP addresses as well.

This will need to be handled carefully as the last time I tried getting away without the host file on linux1, it broke NFS mounting from other machines.  However, as long as the host on linux1 is kept in sync with the dns server files everything should work.

  2571   Fri Feb 5 00:52:55 2010 SanjitUpdateAdaptive FilteringOAF at 0.1-1.0 Hz

 

At 0.1-1.0Hz, there is some coherence between MC_L and RANGER_Y & GUR_Y, see the first figure. Also GUR_Z has low noise there. So I used all five of them, increased the gains of GUR_Z from 0.1 to 0.5. Some improvement near 0.5Hz. We possibly can not do any better with these PEM measurement, as the coherence of the adapted error signal and the PEM channels is almost zero, see the second figure. May be we need to think about placing the seismometers at different places/orientations.

However, there is lot more scope at higher frequencies, lot of coherence at 5-100Hz.

 

Attachment 1: OAF_04FEB2010_noOAF.png
OAF_04FEB2010_noOAF.png
Attachment 2: OAF_04FEB2010.png
OAF_04FEB2010.png
  2572   Fri Feb 5 01:04:58 2010 SanjitUpdateAdaptive FilteringOAF at > 5Hz

 

There is lot of coherence between the error signal and PEM channels at 5-100Hz. We had been applying a 1Hz low pass filter to all the GUR and RANGER channels for stability. I turned those off and OAF still works with mu=0.0025, this will give us some more freedom. Kind of annoying for testing though, it takes about 45min to adapt!

In any case, there is no significant improvement at high frequencies as compared to our usual OAF performance. Also, the low frequency improvement (see previous e-log) is lost in this set up. I think, we have to adjust the number of taps and channels to do better at high frequencies. Also, delay can be important at these frequencies, needs some testing.

 

Attachment 1: OAF_04FEB2010_highFreq.png
OAF_04FEB2010_highFreq.png
  2573   Fri Feb 5 11:01:49 2010 steveConfigurationVACslow pumpdown valve

I have installed a slow start throttle valve in front of V3  This spring loaded valve will cut down on the flow at high pressures. There will be no more sand storme

and static built-up during pump down.

Attachment 1: slow.JPG
slow.JPG
ELOG V3.1.3-