At 7:07PM

I managed to figure out the modbusDrv configuration settings to get the binary output of the Acromag working. I've updated the Wiki page to reflect this. I've wired the XT1541 DAC, BIO Acromag unit to the T1EN and T2EN channels on the TTFSS box but I still can't get remote control of it yet for some reason. When the PDH loop is closed and I switch the TTFSS box to REMOTE, the loop stays closed regardless of what I do to the binary outputs in EPICS.

https://nodus.ligo.caltech.edu:30889/ATFWiki/doku.php?id=main:resources:computing:acromag

I've worked out how to get the binary IO to work with the TTFSS box so that we can activate switches in that unit. It wasn't working in the setup yesterday because of physics. Actually - there is a 10K pulldown resistor in the Acromag unit that attaches the output to ground. The actual circuit looks like this:

VCC (5V) --- (4.99K) --- T1EN -----|-----DIO0 ----(6.2V if DOUT set to 1)---- (10K) ------ | GND

.........   TTFSS...................................| ..............................ACROMAG ...............................................|

T1EN is measured by the switch-chip (SN74HCT157D, chip U3 in D040423) to determine whether it should be open or closed. We need to bring T1EN below 0.8V to get the TTL logic to work.

If DOUT is set to 1, then DIO0 and T1EN become the excitation voltage, 6.2V, and the switch circuit reads high. If DOUT is set to 0, the excitation voltage is removed and we just end up with a voltage divider and around 3.33V at T1EN - which does not register as low.

We can get around this by adding a smaller resistor, say 810 Ohms, in parallel to the 10K, to lower the effective resistance of the pull-down resistor to 750 Ohms. The maximum current the Acromag unit will have to supply is 6.2V/750 Ohms = 8.4mA.

So that's what I did. Now, when I switch DOUT to 1, I see 6.2V at T1EN and when I switch DOUT to 0, I see 0.669V at T1EN. The TTFSS box registers these as two different states and I can lock and unlock the PDH loop from EPICS.

Volts (requested) to Counts

C = a1*V_req + a2

a1 = 3003.5 counts/Volt

a2 = -20.8285 counts

Entered counts (in +ve and -ve amounts) and measured the resulting voltage across the output. I'm not sure how stable this calibration is.

In order to have a better understanding of the FSS electronic boards we decided to take

open loop transfer functions (OLTF) of the FSS North cavity loop. In the same time we took 

the opportunity to verify that the broadband EOM is working. However this measurement will

be done again for a better quantitative analysis of the loop performances.

OLTF measured in the common path shows the shape due to the PZT and due to the EOM

paths as expected.The UGF has been pushed up to 170kHz. In the picture below we see the

traces corrisponding to the two different gain settings.

Connections:

Network Analyzer                           FSS

           RF          ----------->        EXC (Common)

           R            ----------->        OUT2 (Common)

           A            ----------->        OUT1 (Common)

These measurements have been done with the Switch Exc on located on FSS interface board

and with the following gain settings:

OLTF measured in the FAST path have the connections applied at the Fast and the switch EXC on

the front panel turned on. This measurement has been taken with "high gain settings" of the previous

measurement.In the following pictures can be seen the previous OLTF too (green/red).

Connections:

Network Analyzer                           FSS

           RF          ----------->        EXC (Fast)   (front panel switch Exc ON)

           R            ----------->        OUT2 (Fast)

           A            ----------->        OUT1 (Fast)

PLEASE note these are not the total OLTF (relative to their path) because is still missing the transfer function

between OUT1 and OUT2 in both the paths Common and Fast.

I would like to model this loop, I need to figure out the best way to do it.

We tried locking the PLL today but failed. Looking at the beat note on the network analyzer revealed some 70/140kHz harmonics. We thought this might be what is preventing us from locking.

Rana suggested that these are from the switching power supplies (which switch at 70kHz). If so, this may be a red-herring. It's possible we're not trying with high enough gain settings ...

The large peak that is offset by about 50kHz is from the Marconi ... we can make it move around by changing the Marconi frequency.

Description

As described in the elog entry n. 1577 we were not able to lock the PLL as has been described in elog n. 1570. I have started by playing with the two PDH gains of both loops (North and South) as this could have been causes of non tolerable noise in the PLL loop. I have also monitored the peaks described in entry n.1577 as we were suspicious for their preventing the PLL locking.

Conclusion

The PLL loop has been locked repeatedly after locking the cavities multiple times. This result has been achieved by setting the PLL gain on the SR570 at 20 (PLEASE NOTE I was not able to lock the PLL with any other gains settings). 

The peakes of entry n.1577 are not preventing us to lock the PLL.

Some settings:

North FSS interface: Common gain =700; Fast gain = 450; PID = 4.451V;

South FSS interface: Common gain = 850; Fast gain =  250; PID = 0.5527V;

Beat frequency = 69.6 MHz;

However these are not the only allowable settings for the gain, but the PDH loop gains are crucial for the PLL locking. Later I am going to give a quantitative analysis for our PDH loops in order to have them in a more stable and/or less noisy locking point.

After two straight days of trying to get iPython notebooks to run on my Mac and two different Ubuntu installations, I gave up in frustration and rewrote the simpler parts of Evan's noise budget code into MATLAB. The following Noise Budget (without any BeatNote measurement from the lab) represents an estimate of the current noise in the system.

Clearly, with the ISS OFF, we are hugely dominated by intensity noise. We will investigate if this is the reason that we can't lower the FM deviation on the Marconi below 300kHz/V.

If you can't install python, you can run it all in the Sage Math Cloud for free.

Also, please post the final design for the heat shields which you've sent out for fab.

We've set up the PEM monitor computer to broadcast the temperature channels to EPICS.

• Computer is a Windows machine at 10.0.1.35
• Has OPC server installed to communicate with the Newport zCDR wireless unit
• OPC server still needs a license. Currently terminates after 2 hours
• OpcIocShell runs to create an EPICS IOC that interfaces with the OPC server and broadcasts the channels via EPICS.

We'll find a permanent home for the machine tomorrow.

https://nodus.ligo.caltech.edu:30889/ATFWiki/doku.php?id=main:resources:computing:network:menu

Removed a few old computers from the diagram (35W laser, old OPC server, PSL Workstation, HWS workstation). Added a few new ones (OPC server in TCS lab, New workstation in PSL lab, Acromag control box in PSL lab).

Summary

Today the beat frequency was very difficult to find. Something changed, and I am not sure yet how to drive

the frequency of the beat. However the beat frequency is at 50MHz while so far it was at 64MHz. The PLL lock

more difficult and noisier.

Note:

Sometimes today or Yesterday we have increased the temperature of the lab (~ 2 degrees F).

Data:

lab computer/control fb2/home/data/20151016_PLL_noise_50MHz

Summary

The AEOM has been installed in the South path replacing the EOM 21MHz used for the PMC. There is a high noise that I clearly see at the photodiode in transmission. 

When I have placed the AEOM in the path I have decided to take the alignment of the previous EOM as reference. Not ideal because the reference should be the incoming beam. The beam is not parallel to the table and it was decided to be as less as possible invasive. The mode matching and the alignment gave at that time 20% of visibility (at each polarization). After the installation parameters where unchanged. Later I have improved the alignment bringing the visibility at 30% for both the polarizations. After that, when everything was in place I have easily locked the cavity but the power in transmition was showing a very high noise. I have spent all the day trying to twick the alignment because and servo loop gain, but we need to solve this before going further. My back does not allow me to proceed for today.

NOTE:

I have also noted that the South Laser which is labeed 2W laser has the lambda/4 and the lambda/2 rotated in a way that at the output of FI we had few mm. I am not sure if damping the power at the FI is a good thing.

https://dcc.ligo.org/DocDB/0120/D1500213/001/D1500213-v1.PDF

Here's the drawing for the shield. https://dcc.ligo.org/LIGO-D1500403

Table height

Currently the height of the optical table is 27.5" (see attached PDF). We would like to increase its height.

The legs are provided with a spacer http://www.newport.com/Pneumatic-Vibration-Isolators-with-Automatic-Re-Leveling/844255/1033/info.aspx 

which allow us to increase the height just by replacing the spacer (2.5").

The new spacers bring the height of the legs up to 28" (or 23.5") and the total height of the table up to 40"-41" (or 35"-36").

I ahave asked a quote for spacers which make the legs height equal to 28". They can come with flanges for horizontal bars. I have asked a quote

for this too.  We need to decide if this height is fine and if we want horizontal bars.

We have raised the table by ~ 16" and the enclosure around the table by about 13".

• Feb-16: covered the table and removed the old enclosure. Equipment emptied from cupboards and stored in TCS lab
• Feb-22: Transportation came across and raised the table for us.
• Mar-3/4: Taller enclosure assembled around the table

This morning I sent an email to Evan in order to obtain PMC documentation. 

For now the only documents found is a mechanical drawing E1400332. I am aware of additional 2 documents that he wrote for this design. I also asked for the electronic/servo

instructions as I am not aware if he built/bought parts for the PMC servo.

This morning I have also poked Rich, in order to have advices on what type of epoxy we need to buy for pzt gluing. Not sure about that. I am going to investigate

on what to buy and especially what should be the best procedure for gluing.

We have been collected the materials bought from Evan:

1. Spacers;

2.  Mirrors;

3. PZTs;

Things to buy:

1. O rings;

2. Epoxy for PZT;

3. Screws;

I am considering to use the table in the TCN lab, once we have a functionally lab. All the information about the PMC will be collected in a section on the PSL wiki page.

O rings and screws have been ordered!!!

I've ordered a 4U rack mount kit with DIN rail for mounting the Acromag stuff. It's open on the top and bottom but is quite deep. I've made up a front panel for this kit which will connect to the slow controls with 8 analog inputs, 8 analog outputs, 4 binary inputs/outputs, 2x 25 pin connectors for connecting to the FSS boards and a 15 pin connector for connecting to the ion pump current supply.

With the (intentional) demise of the VME crate, the slow controls for the laser were brought back online today. So far, only the controls for the laser temperature are running. However, if you run the LASER SLOW MEDM screen, the channels C3:PSL-ACAV_SLOWOUT and C3:PSL-RCAV_SLOWOUT are now active. 

To achieve this: I did the following:

1. Restored FB2 to the local network to allow for GUI interface to EPICS.

2. On the computer running the MODBUS IOC (10.0.0.33), I created some channels in the DB that are "ai" channels with the names C3:PSL-A(R)CAV_SLOWOUT. There are some CALC channels that use these as inputs to convert to counts and then the output is sent to the approrpiate DAC register on the acromag unit via the modbus protocol.

3. So now any computer running EPICS on the local network can interact with the SLOWOUT channels

We connected BNCs up between the Acromag terminals and the SLOW IN BNC inputs on the front panel of the two NPRO controllers and double-checked that a requested voltage actually yields that voltage at the NPRO controller.

With the (intentional) demise of the VME crate, the slow controls for the laser were brought back online today. So far, only the controls for the laser temperature are running. However, if you run the LASER SLOW MEDM screen, the channels C3:PSL-ACAV_SLOWOUT and C3:PSL-RCAV_SLOWOUT are now active. 

To achieve this: I did the following:

1. Restored FB2 to the local network to allow for GUI interface to EPICS.

2. On the computer running the MODBUS IOC (10.0.0.33), I created some channels in the DB that are "ai" channels with the names C3:PSL-A(R)CAV_SLOWOUT. There are some CALC channels that use these as inputs to convert to counts and then the output is sent to the approrpiate DAC register on the acromag unit via the modbus protocol.

3. So now any computer running EPICS on the local network can interact with the SLOWOUT channels

We connected BNCs up between the Acromag terminals and the SLOW IN BNC inputs on the front panel of the two NPRO controllers and double-checked that a requested voltage actually yields that voltage at the NPRO controller.

The old 3123 cards were 16-bit ADCs. Are the acromag units 16 or 12 bit? Also, what about the DAC range? I would not drive the laser directly through the DAC without a low pass.

Summary

Analysis of the frequency modulation for the FSS loop has been done (may be I have repeated it )

as we want to increase the frequency modulation from the current 21.5MHz to something

bigger than something like 30MHz.

Higher order mode analysis

I have considered a cavity of length L=3.45’’=3.68cm and radius of curvatures R1=R2=1m.

1.

Plot 1 shows the frequency distance vs frequency modulations. I care that HOM (Carrier and

SBs) do not overlap with carrier and SBs TEM00. The closest mode is the number 23 which

interfere with the positive SB around 31MHz. I do not consider this to be a problem, but I go head

anyway. At that frequency we also see the mode 46 crossing the the SB (TEM00).

2.

In order to better visualize the distances shown in the previous plot I use the minimum distances from the TEM00 (carrier or SBs).

The minimum distances are referred to the closest mode which may change along the SBs scan.

We see that on 31.7mHz we have a minimum as noticed previously and it is because of the mode 23.

From here if we would like to avoid the mode 23 I would choose a frequency modulation of ~39MHz.

3.

I fix the frequency modulation at 39MHz

I wanted to check if the mode 23 can be “avoided”. I consider the tolerance on L as L+- 0.0005 and Rc of +-1cm

There is a chance that the mode 23 overlap depending on what the variations from the nominal values are.

However the same is going to happen if we fix the Frequency modulation at 35MHz. So between  35MHz and 39MHz

I do not see substantial differences. (Please note that the color are normalized at 1MHz detuning from carrier).

4. While the scan of L and Roc has been done I have checked that no other HOMs come into play. The mode 23 and 46 are

the modes which determins the red lines at the point 3.

5. Mode distances at 35MHz and 39MHz.

From what I see here 35MHz is fine as mode 23 is the closest one to TEM00. As a reminder the pole of the cavity is ~180kHz. In principle 35MHz would be 4MHz

apart considering the nominal values L=3.68cm and RoC=1m.

I showed Andrew Wade the CTN lab and gave him a safety walkthrough. We also went over the safe operating procedure for turning on the laser.

I have a set of two Wenzel OCXO added to the electronics chassis, courtesy of the gentry from the Cryo Lab. The chassis has outputs for 36MHz and 37MHz.

We can use these temporarily and exchange the oscillators in future when we select a new sideband wavelengh.

As we are doing mode matching I compiled a list of presently available lenses to save time looking or ordering. See below.

A dynamically updated version is here: https://docs.google.com/spreadsheets/d/1x4-VQ85Wl7kGUH-V7HffCr6B7wIcD1y3VxXO4j6H5T0/edit?usp=sharing (this may not be a stable link far into the future)

At the time of posting the list is as follows:

I showed Andrew how to set up the slow controls. We got the previous 3 ADC channels back up.

• C3:PSL-TRANS_ACAV_ISS_DC
• C3:PSL-TRANS_RCAV_ISS_DC
• C3:PSL-TRANS_RF_DC

Also:

• rewired a few BNC feedthroughs up through the front panel of the Acromag chassis
• bolted the front panel to the chassis independently of the screws that mount the chassis to the rack

Summary (Entry 1633 revised)

Analysis of the frequency modulation for the FSS loop has been done 

as we want to increase the frequency modulation from the current 14MHz to a

higher frequency in the range of 25 to 45 MHz in order to increase the FSS  unity

gain frequency. Below 25MHz we do not have too much benefit as currently we

are modulating at 14MHz. Above 45MHz we have photodiodes noise and we want to 

deal with that.

Higher order mode analysis

I have considered a cavity of length L=3.45’’=3.68cm and radius of curvatures R1=R2=1m.

1.

Plot 1 shows the detuning frequency vs frequency modulations. We care that HOMs (Carrier and 

SBs) do not overlap with the carrier TEM00. The closest mode is the number 23 which

interfere with the TEM00 around 31MHz. I do not consider this to be a problem. The morde order

is ment to be as the sum of n and m. The legend shows color of the modes in a dashed style. However

each mode order is showed with carrier (solid line) a positive SB (dash-dot) and negative SB (dashed).

From this plot we can see that in the range 25MHz to 45MHz we are pretty safe from HOMs overlaps.

If we care of the mode 23 than we should consider modulation frequencies above ~35MHz at least.

I consider a cavity pole of 135kHz (see elog 1480 https://nodus.ligo.caltech.edu:8081/CTN/1480)

2.

Plot 2 shows a zoom-in version of plot 1.

3.

In order to better visualize the difference between the detuning frequency from TEM00 and the closest HOMs (either carrier or sidebands) 

I provide the following plot. Please keep in mind that in general this way to plot the differences does not take in account of

which mode order is the closest. However in this case it is only the mode order 23.

We see that at 31.3MHz we have a minimum as noticed previously and it is because of the mode 23.

Choice of the Modulation Frequency

Currently in the TCN lab we have two oscillators; one is at 36MHz and the other one is at 37MHz; I was tempted to use them

but Rana suggested to have 2MHz separation between the two modulation frequencies in order to avoid to see noise in the beat note 

after the demodulation process.

So if we want to use one of them we can pick 37MHz and buy a 40MHz one.

If we have to buy two of them anyway we can go for the same 37MHz and 40MHz or shift a bit the frequencies to 39MHz and 41MHz.

I would pick 37MHz and 40MHz leaving the option to buy or not buy the second 37MHz.

Before buying all the parts and do all the work which is consequent at this choice, it is good to have some comments/opinion!

Post documenting the set out of the PLL electronics and cable routing.

We were having trouble with the PLL electronics in obtaining a beat note error signal and lock. We can see a beat note of 17 dB clearance above the DN level on the spectrum analyzer.   This post is to document the present configuration and the powers present at the mixer before I start to troubleshoot (aka pull it apart).

Schematic of the PLL electronics stage attached below.  Settings for the Marconi and SR560 are also below. I would noted that there is a low pass filter on the output of the phase detector (which I assume is a mixer) that extracts the (omega_LO - omega_sig) but has nowhere for the dumped (omega_LO + omega_sig) signals to go.  I'm guessing the non-DC signals are getting reflected directly back into the ZRPD-1+. I can't imagine that is great if back reflections start to form parasitic competing signals with the beat note signal.

Marconi 2023A
Carrier Frequency = 75.526 MHz (close to PD beatnote)
RF Level = +13 dBm
FM Devn 10.0 kHz (set by Ext DC)

SR560
Filter cutoff = DC
Coupling = DC
Source  = A (B port terminated with 50 ohm)
Inv = On
Gain Mode = Low noise
Gain  = 50
 

The spec sheet for the ZRPD-1+ indicates that it is a Level 7 mixer with a damage threshold at 50 mW, so the +13 dBm probably isn't a risk but may be saturating the mixer.  I don't have a calibrated power level for the RF PD input, but it is low (definitely not +7 dBm).  At this stage we either need to improve the beat note or implement some kind of amplification so that both the PD-RF input and LO inputs into the phase detector are matched at the ideal +7dBm level. 

At this stage both PDH control loops are ringing with DC signals that are very angular (and about 10% of the total DC signal). The angluarness would indicate that they are probably both railing somewhere or converting an control-error signal to amplitude modulations.  Previously we have diagnosed this back to imperfect polarization input into the fast control BB-EOMs.  I checked both North and South path and the inputs to the all EOMs (the 14 MHz and fast control) are very well aligned to 's': modulator polarization is not the issue. Antonio had this very well optimized form yesterday.

Tuning laser temperature around on the north path, it is apparent that there are some new/stronger HOMs around the region of our TEM00 modes. This is only a qualitative observation. It is possible that over time our mounts have drifted and the alignment is off.  This is now on my list of things check and to tune up.  

At this stage the PDH loop issues need to be resolved as they are impeding efforts to reliably form an observable beat note.

The process of unlocking and re-configuring for alignment and other types of diagnostic measurements is cumbersome.  There are BNCs popping out of the experiment everywhere that are also mostly unlabeled.  It would be much nicer to have all the useful control signals and PD signals coming out at a central point, preferably next to the computer interface and electronics rack where we lock and tune gains/temperatures.  I will gradually start migrating the cabling over to that side of the table and then install labeled 'patch lines' that are routed to mounted BNC panels around the table. This way there is a central location close to where we perform locking etc but it is relatively easy to patch through signals and oscilloscope trigger lines for temporary oscilloscopes situated around the table. This also means that consistent labeling can be applied.

It is not a great idea to label the cables with label-writer tape or color code them with electrical tape.  From experience the glues in these breakdown in 2-5 year timeframes either flaking off or leaving a sticky residue that is almost impossible to remove with solvents that don't also dissolve the BNC cable plastic. Heatshink can be good as it can be cut off and doesn't have glue. Otherwise a cable tied labels work well. We have some of these.

It also distresses me a little that BNCs cables are routing HV around the table without a clear system for distinguishing these cables. By Murphy's law these will one day be plugged into something sensitive. Some of them are labeled by are not marked clearly as HV. If I can find some red heat-shrink that fits over the ends of the BNC cables I might start labeling HV lines with a consistent markings to lower the risk of mix ups. 

These are the RTDs we use in aLIGO TCS

http://www.omega.com/Temperature/pdf/1PT100KN_2PT100KN.pdf

The heater driver in the old design (iLIGO PSL) is this horrible programmable power supply with high output noise (see measurement of 40m PSL heater).

What we want instead is something that takes in +/- 10 V and drives a DC current (not PWM) into our heaters (which are ~50-100 Ohms). A BUF634 is almost good enough; it can do 200 mA at 10 V. Is there a BUF634 equivalent which can do more like 500 mA? Otherwise we can just use a opamp + transistor.

What is being used for the ring heaters at the LIGO sites?

I assume the key parameters are the noise characteristics here.  They seem to commonly quote the input noise voltage which I'm guessing gives a measure of the equivalent noise injected by the buffer itself. 

Data sheets list the input noise voltage of the BUF634 as 4  nV/√Hz @ 10 kHz . There is another buffer (300 mA), the LMH6321, its datasheet claims  input voltage noise 2.8 for >=10 kHz. I don't know what this means for noise at mHz.

Is there a reason we can't just buffer with more general purpose high current OP Amp? Is the the noise just a killer for that whole catalog of devices? Looking at a random selection, all the ones I've looked at have higher input noise voltage values but the digi-key product selector doesn't let me filter for noise characteristics so maybe there are low noise ones. 

Here are some numbers on the heat radiated away from the vacuum can by leakage through the foam insulation and from the exposed metallic parts. The heat loss is dominated by the foam as it has an larger surface area. However, these numbers are maybe a little rough as they don't account for the Al foil cladding on the outside and ignore the details of the cylindrical geometry.  

I've been working on documenting the thermal aspects of the vacuum can.  Info is spread across the elog in various places but not in one place.  This stuff is gradually getting added to the wiki which will be the central collecting point for information to avoid this iterative amnesia.  I am also almost finished on a graphic that summarizes the setout of the vacuum can and its sensors. This is just the heat loading calculations.

The tank is 22" long with 8.3" tube diameter with two 10" flanges on the ends.  This apparatus is clad all around by 2" (average) thick CertiFoam 25 see post PSL:178 for characteristics (note that Frank's values are for 1" thickness only, this must be scaled to the thickness used). Total dimensions of the foam box are 12x12x36" giving a surface area of ~1.3 m^2.  For a tank held at 35 C above room temperature 20 C this is 11.3W of heat loss.

At the top of the vacuum can there are three half nipples welded along the top, these hang out above the insulation as access is needed for the turbo/roughing pump connection, the ion pump and the sub-D 9 feedthrough. The exposed surface is a mix of shiny stainless steel and matt/sandblasted bits.  Shiny and matt exposed areas are respectively 0.0296 m^2 and 0.0168 m^2 (not including the ion pump) which is not big.  With emmisivities of 0.09 and 0.18 for these two surface types we get a total of 0.5265 W radiative heat dissipation for a tank held at 35 C above 20 C room temperature.  

Thus total estimated heat for a 35 C tank is 11.78 W. We don't need to run it at this temperature but I use it as a reference value.  See attached graph for heat loss as a function of vacuum can temperature.

A summary of these numbers and details is in the attached matlab file.

---

For reference the tank has 4 resistive heating mats wrapped around it. The small ones near the ends are 30 Ohm and the two larger ones near the center are 70 ohms each. These have been connected in parallel+ series network that gives a total resistance of 50 Ohms and can be driven with up to 115 V.  To just maintain the tank at 35 C we would need 24.5 V with ~0.5 A. This seems like a lot but is almost doable with available OPAMP buffers.

---

The next step is to work out heat capacity. I can't find design drawings for the tank itself on the elog/wiki, maybe its too far back in ancient history. There is a solidworks drawing on the SVN but was made in the student version of solidworks so is very buggy.  I will try an extract numbers to get an idea of the mass of metal in the tank. 

Also in progress is a step function measurement of tank cooling.  I spent some of this week working out how to integrate a new RTD acromag card into our existing EPICS setup so we can log the temperature drop after heating is turned off.  This took a while as I was unfamiliar with this kind of setup and also the power supply turned out to be not doing what I thought it was. The voltage current source is now hooked up and a thermister fitted for logging temperature. This measurement should give us some more grounded numbers on the real characteristics of the tank thermal decay rates.

To verify estimates of the heat load and thermal inertia of the system I am conducting a simple step test of the vacuum can heating.

The resistive heaters on the vacuum can were given a steady DC 35 V over the day, the system settled on an equilibrium temperature of 46.14±0.05 C. The location of the thermistor is shown on the picture attached. It took a long time because I initially used a fairly low current power supply (0.5A).  I switched this out for a 3.0 A, 0-60 V supply which was sufficient with the 50 Ohm heaters. 

I have left the tank to cool down (starting at time stamp 18:50:00, Aug 29) with the Acromag cards logging the decay back to room temperature.

Here's the RH driver PCB for aLIGO:

https://dcc.ligo.org/DocDB/0021/D1002529/006/Driver_RingHeater.pdf

It uses this power amp: https://www.digchip.com/datasheets/download_datasheet.php?id=508953&part-number=LM12CLK

The full cool down curve of the tank and a fitted curve ()are attached below.

Least squares fitted values are 

a=23.4145±-0.0005 C, tau=7596±170 s, c=21.00947±0.00004

The variance on the time constant is very large, I think this might be due to a poor fit owing to the fact that the temperature was still rising (rather than steady) when I turned the heating off. We want the whole system to be at equalibrium at the start of the test and we also want a decent tail on the decay. I have therefore repeated the measurment this time heating overnight with 30 V (18 W) which settled on 44.36 C.  I have now left the vac can to cool for the rest of the day (starting about 9.30 am).

Either way it looks like the characteitic 1/e decay time is on the order of 2.1 hours.

The Newport temperature, pressure, humidity remote sensor unit (zED-BTH) in the PSL lab was sitting on a recessed shelf of the work bench. This isn't a great location for accurately sensing and logging the lab temperature. I have now mounted it on the center of the western wall of the lab (pictured) 1.524 m from the ground. This location has free air flow.

Indicator LED also wasn't blinking. I checked the batteries and they were down to about 0.200 V for each of the AAs. I've replaced them as of today.  

It seems like these units have an email alert function: http://www.newportus.com/ppt/ZSERIES.html . Maybe we should set one up for battery replacement alarms.

I'm working with Jamie to get a frame builder running in the ATF & CTN for these temp/pres/humidity channels to be stored on. There's no 
long term storage right now.

Post summarizing some changes and current configuration of the acromag crate.

Some useful previous posts:

• https://nodus.ligo.caltech.edu:8081/CTN/1564 (a previous summary, there are also some useful links there)
• https://nodus.ligo.caltech.edu:8081/CTN/1539 (some starter links for initial setup)
• https://nodus.ligo.caltech.edu:8081/CTN/1545 (reference to more LIGO documentation using EPICS Modbus with CDS)
• https://nodus.ligo.caltech.edu:30889/ATFWiki/doku.php?id=main:resources:computing:acromag (Aidan's wiki tutorial on setup of acromag cards themselves)
• https://nodus.ligo.caltech.edu:8081/CTN/1573 (some notes on interfacing the TTFSS box with acromag digital outputs and how to get that working)
• https://nodus.ligo.caltech.edu:8081/CTN/1574 (Aidan's notes on calibrating acromag outputs, can't find input calibration, might be in paperlogbook only)

I've made a few modifications to the Acromag crate.  There are three cards a XT1221-000 (8 differential inputs) a XT1541-000 (8 DC outputs + 4 digital I/Os) and a  966EN-4006 (Six Channel RTD/resistance).  Up to this time we have been adding channels one by one as we need them. I wanted a few more slow monitor channels so I just went ahead and connected all the rest of the remaining analog input and output channels to finish the job.  I also replaced some of the shorter cabling for the BNC front panel connectors with longer cables so there is a bit more slack inside the crate for these wires.

In time it might be good to put a few LEDs in the front pannels from the binary outputs of the XT1541 card so that we can get a visual blink indicator that some of the perl scripts are running. Alternatively, we can go back to interfacing these with the TTFSS boards as was done in PSL:1573.

A DIN rail mountable power supply (PS5R-SC24) was also fitted inside the rack to replace the variable voltage lab power supply.  This had been a temporary setup. This new power supply provides up to 30 W at 24 V from 110 V AC. The AC in line is strain relieved with a cable tie put through a hole I popped in the side of the crate.

Some rearranging of channels:

Previously the PZT/FFS fast controls monitors for the north and south path were inputted through the acromag channels 4 and 5. This was because at the time we only had short BNC cable connector feed throughs.  The order of channels now corresponds the the front panel labeling. The mapping of inputs/outputs+Resistive RTD unit is as follows. The present database file is attached below for reference.

I've put together a draft schematic of the basic layout of the shields and tank as well as the present state of connections of the dual ref cavity setup (attached in both pdf and omnigraffle formate).  There are presently no active controls on either the individual cavity shields nor the whole vacuum can. We are relying on passive stability. 

The tank heaters were hooked up to a large current Hewlett Packard 6267B DC power supply.  The unit is no longer producing voltage or current, either a fuse is blown or its kaput; I have removed the unit from the rack. There are a few more 6267Bs around the lab and all have similar problems. I have stacked them all in the corner for assessment and possible disposal in the next big lab cleanup.

Resistive temperature sensors on the two shields are both broken while resistive heating wires on both shields are working. However, according to Tara the length of wire is different because they didn't know at the time if they would be voltage or current limited in their actuation: to mitigate for both cases they chose for the south path to have 85.6Ω resistance and the north 156.8Ω. When we install the new shields we will bring these to be identical. 

I'm not really sure about the existence of a voltage-to-current drivers used in the past.  Its evident that there was one for the previous implementation of PID control of the vacuum can, but I haven't seen any boxes/circuitry in the lab. We need a low noise high current driver for our new implementation. Aidan has linked to the TCS drivers in PSL:1711. Our design requirements will come down to the characteristics of the plant and how it filters the LF noise.

Some information we don't have yet:

• The mass of the tank;
• the type of temperature sensors used on the current shields;
• a model of radiative transfer between shields and the tank.  This is significant complicated by the fact that the Al frame and mechanical isolation stages are actually acting as a shield as well (they go all the way around);
• the schematic of the AD590 -> voltage adaptor circuitry.  The LIGO DCC number appears to be for a VCO servo even though it is labeled "LIGO Temperature Sensor Interface board" a search of the DCC or elog doesn't seem to turn up anything;
• What current drivers were used in the past to drive the vacuum can heaters
• Modeling of full impact of end caps? 
• modeling of improvement from moving to 1" windows on vac can?

Note: the omnigraffle file has multiple canvases with different info. 

Settings for the 14.75 MHz LO and modulator source.  The Marconi doesn't keep memory when turned off for a period of time. This post is general lab documention. 

Marconi 2023A
Carrier Frequency = 14.75 MHz
RF Level = +13 dBm
FM Devn = 0 Hz Off [Off unused]
ModF = 1.0000 kHz [Off unused]

Antonio took attached picture a while ago.  I have also attached a schematic of the present routing of RF to TTFSS boxes and modulators.

See Tara's post PSL:1090 for info on the EOM driver

The previous cool down test was repeated over a longer time period.  The tank was heated overnight with a 30 V (18 W) (as stated in previous post) and its settling temperature was 44.36±0.05 C. This gives us some idea of the expected DC heat load that the tank will require to maintain an elevated temperature of ~45 C.  

The supply was turned off and a cool down curve was recorded over the next day. Plot is below.  Fitted values were

a = 19.4291±0.0005 C, b=13980±730 s, c=20.75216 ± 0.00003 C

for the fitted equation .

With this test (with a longer cool down tail), the time constant for the insulated tank to come to room temperature was 3.88 h

Quote:

The full cool down curve of the tank and a fitted curve ()are attached below.   Least squares fitted values are  a=23.4145±-0.0005 C, tau=7596±170 s, c=21.00947±0.00004 The variance on the time constant is very large, I think this might be due to a poor fit owing to the fact that the temperature was still rising (rather than steady) when I turned the heating off. We want the whole system to be at equalibrium at the start of the test and we also want a decent tail on the decay. I have therefore repeated the measurment this time heating overnight with 30 V (18 W) which settled on 44.36 C.  I have now left the vac can to cool for the rest of the day (starting about 9.30 am). Either way it looks like the characteitic 1/e decay time is on the order of 2.1 hours.

Summary post pulling together numbers on heat shields.  For future reference.

Drawings and Soildworks files can be found at the following links:

• Cylindrical heat shields: D1500403
• End caps D1500213

Purchase orders for copper parts: INSERTLINKtowiki
Goldplating specs, vendor and PO: INSERTLINKtowiki

These values were pulled from the SolidWorks files:

*Emissivities drawn from Rathore and Kapuno, Engineering heat transfer, Jones & Bartlett Learning, 2011

edit: Mon Oct 10 17:46:40 2016 added DC heat load and radiative cooling slew rate to TOTALS table

Some tests and characteristic noise of the RH drivers: https://dcc.ligo.org/LIGO-E1000785-v1

The documentation  shows noise measurements of ring heater drivers over 0-1.6 kHz span, it would be nice to have have more resolution at LF.  

Key points about the TCS Ring Heater Driver:

• Conversion 0.08A/V (as is, although this can be adjusted)
• ~0.07 µA/√Hz @ 100 Hz for 100 mA 
• ~0.16 µA/√Hz @ 100 Hz for 600 mA

What it does show is increasing noise as the current is increased across the load. It is low compared to expected driving currents for the tank (0.6 A, for 50Ω total load) but maybe note for the shields (~24 mA for refcav shields, for 50Ω heater load).

It doesn't seem to include a characterization of the thermal sensing circuit's noise.

PDF and omnigraffle file of the PSL lab layout, as of Aug-Oct 2016.

I made this but can't remember posting. Now attached below.

I have updated the graphic outside the lab. This can also be used when the SOP is next updated.

Summary of channels active in the current PSL lab configuration. The .db file is also attached below.

We are almost full for input channels. Also if we wanted to implement an EPICS interface for the FSS control boxes we would need a total of 8 output channels.

List of all the current channels:

Overview schematics of thermal controls' elements and of control loops. These are the basic elements and some numbers.

Power budget for essentials.

APC SMX3000LVNC unit: battery capaciy 738 Volt-Amp-Hour

At the upper bound of consumption (at start of unit life) this will ride out 45 min blackout.