below a picture of the first setup using one fiber coupled laser diode (right) mode matched to the FS cavity (left). First modes can be seen in transmission of the cavity. Further optimization required.
can lock it for a second or so before loosing lock, but didn't optimize loop parameters.Still having problems getting sidebands so this is currently No 1 on my priority list.
I actually don't know what i'm locking to, but can lock to individual modes. It looks like there is kind of a zero crossing, but not really.
Using Zach's resonant EOM at 33MHz at the moment. Had to switch PD from Thorlabs PDA10CS to the 2GHz diode.
With a 13dBm mixer and 23dBm into the resonant EOM the "signal" is only about 7mV large and does not like a n error signal at all.
More like some funny oscillation. Probably due to the high frequency noise. Also tried other frequencies, mixers, photodiodes. Needs more investigation.
Modematching into the cavity is only about 25%. Could be either a non impedance matched cavity or our calculation for the mode matching is wrong. Will check this tomorrow.
Nevertheless this can't explain the non-existing error signals.
I want to generate a more coherent (recorded) roadmap for the cryolab so that we can better prioritize our course of action. Before I put together some visual diagram representing this information (which I think will be useful for us), I am going to write down a list of the stuff we know we want to do, as well as the more far fetched things we might one day want to do. Here is the list by category, PLEASE ADD THINGS IF YOU WANT - if you reply to this post with suggestions, I will try to incorporate them into edits of this post. In no particular order, we have:
META (LAB, ETC)
PK borrowed two nice Scopes for evaluation (Agilent MSO7104B and Tek MSO4101, both 1GHz and 4 or 5MSP/s). They look more like small TV's with a lot of knobs (nice large screens).
So we have two ~$20K scopes to have a closer and probably more detailed look into what's going on when beating the laser diodes and measuring the linewidth / frequency jitter.
We started last night and got a little more familiar with the Agilent scope and how to sample the beat signal, do an FFT and some statistics. It's really nice !
We will continue today and the next days and post the results as soon as we know we think to measure
made some slight changes to the spacer drawing. updated files below:
Checked the update - w.r.t. a certain company- they were nonresponsive (to use the nomenclature) when I asked them to quote on the spacer. As far as we are concerned, they are just another Si supplier. Your concern about that supplier sounds to me like we are asking them to give us certain surface quality / precise geometry, neither of which I think we need from the raw materials, but from what I understand, we have very limited requirements for the spacer BEFORE machining.
Or did you mean you were concerned that there weren't things like purity, doping, not being broken specifically in the quote. This all seems sort of moot, since it seems like a not-so-hot idea to use something with the fabrication/shipping process and lead times given in the elog.
i thought the company we are talking about is doing everything (drilling, polishing) except the polishing of the part which requires flatness for contacting. The second company is then doing the re-polishing of the end surfaces, right?
submitted the order, my credit card has been charged already so they might actually ship them this time
bought one of the NIR intensified CCDs which have the phosphor coating on the chip as we need at least one good one for mode identification. My home-made ones need still too much power, so you can't identify higher order modes
Edmund has one with 15% discount at the moment, so it's "only" $1700. Also bought two turning mirrors to see how they are. Specs are not bad, but so far i couldn't find any turning mirror not being frosted on the back.
A full beam analyzer for 1550nm is about $6k, which me might wanna buy in the future if we do more 1550nm stuff. For now i think we don't need one.
UPDATED THE SPREADSHEET
Unfortunately the concave mirror i've ordered and which was in stock according to their online shop turned out to be out of stock
So i've changed the order to a similar one. The only difference is that the new one doesn't have a wedge which shouldn't matter for our prototype as the transmission is only 50ppm (and it's still AR-coated on the back)
I agree - wedge probably doesn't matter with a T=50 ppm
Here are the drawings I submitted to the local machine shops for our 1550 nm triangular metal PMC.
The round trip path length is 2 x 6.3 inches. We want to use either 1m or 0.5m curved mirrors in the back, which give a waist size of:
1m => 425 um waist
0.5m => 340 um waist
Cavity FSR is***:
The PZT we have ordered and plan to use, the PI P-016.10H, has a 15 um range with a 1000V max voltage. Assuming linearity gives us 15 nm / V.
The voltage required to move 1.5 FSR is:
1.5 * (dL = L x FSR / f) * V / 15 nm = (6.3" x 0.0254 m/") x 936 MHz / (1.93x10^14 Hz) * V / (10^-8 m) = 77 V
Frank wanted to get some (specific) 60V power supplies, if we use these we only get a 1.17 FSR full range. Is this "good enough?" It means that we might not be able to sweep out two TEM00 modes in the full range unless we tune its length with a heater.
*** I am guessing a tiny bit. Since the FSR of a Fabry Perot of length L is c/2L, and the triangular cavity is like a flat/curved Fabry Perot with 2L = roundtrip length of triangular cavity, I think the FSR of a triangular cavity is c/(roundtrip length)
I have updated the Wiki (here) with the information about quotes from the two machine shops.
they arrived today
I scanned all the measured spec sheets and put it on the wiki
i had a closer look into cryo temp sensors and what we might wanna use for the first tests.
The comparison posted some time ago (link) showed that platinum RTDs seem to be a good choice in terms of stability / repeatability of the cryo controllers.
I found the following publications which go into more details, but they agree with the above statement that platinum RTDs are a good choice for 120K:
Resolution and Accuracy of Cryogenic Temperature Measurements
A review of cryogenic thermometry and common temperature sensors
I extracted some plots from the above publications. The red line indicates ~120K, our first setpoint. The curves are for different types of sensors. Detailed description below the graphs.
They show that PT RTDs are a good choice in terms of sensitivity and temp resolution for ~120K. Other sensors are much better at lower temperatures.
The platinum RTD plot is for a 100Ohm sensor. So by using a 1k or 10k PT RTD we can get even better. The cryo controller we bought can handle higher RTD values.
Later we might wanna try some high resistive NTCs/PTCs which have much higher gain and so much less noise. The drawback is that the range is very limited and there are issues with the repeatability.
Below a list of items i was working on in the past which i think we can finalize but want to double check a last time.
RA: Go, go, go
Warren checked the vacuum system this morning, so i placed all orders for the remaining parts.
The vacuum pump is the longest lead item with 3-5 weeks (still before the cryostat arrives). Everything else is in stock and should be here end of the week.
the shipping of the cryo PT sensors from Heraeus will be delayed a little bit.
They still didn't ship the sensors as the still didn't receive a correct PO.
Our Techmart database is out of date and contains a wrong company name.
So i'm waiting for the change request paperwork being done before we can resend the order.
They also don't collect sales tax and need proof/conformation that we do not resale the parts or pay sales tax somehow.
I hope we can figure that out by end of the week. Parts are still in stock.
Unfortunately, REO is not making low value or small quantity coating runs anymore. After their internal meeting this morning the sales guy contacted me to let us know that they won't do small coating runs anymore.
This is what i got from them:
We do have the capability but our business model has shifted to focus on recurring optical opportunities or large high value programs, as we focus our limited engineering resources.
i got confirmation that the pump including all options and adapter will be shipped August 5. They don't have any in stock at the moment.
I hope it won't delay anything as the cryostat will be delivered first week of August as well.
Will update the project file on the svn.
We might have a problem with the current vendor for our Silicon mirror substrates.
According to the Caltech export control guys this company is currently blacklisted, so we are not able to buy from them.
However, i've called that company today and they say that they are not blacklisted and can do unlimited business with anyone and our people might have an old list and are not up-to-date.
The have been on that list a few years back when the were accused of illegally exporting defense technology. However this case is settled and they paid a small fine (link).
I've forwarded some contact information to Rod who will forward it to our people on campus so that they can get in touch with their attorney to clarify this. That's what they offered me today.
He was sure that they will figure that out on a very short timescale. I will wait another day or two as we don't have a real alternative at the moment (with short lead time) and would have to start the whole bidding process again.
In parallel i will make a list of companies i didn't ask already and then we start with those. We will also talk to all the others to see if we can get something with less quality earlier.
we talked to a couple of vendors today and relaxed the specs a little bit. Four of them will quote us within the next 24h.
Specs are now:
Silicon mirror, 1m cc, 30min wedge
L/10 @ 1.5um
SD20-10 both sides
(annulus only if possible on short time scale)
checked all open orders. Everything is on time or only delayed by a few days except for the dual power supply which is still missing (we got the high voltage one and the rack mount kit).
The dual 30V-3A supply is currently backordered with a delivery date of Aug 24th. They quoted it to be in stock but sold all of them before our order was placed.
I don't know if it makes sense to look for an alternative place which has it in stock as we don't need this super urgent and we can always borrow something from the PSL for a few weeks.
There are also some special nuts missing to mount them to the rack kit. I talked to the company and they figured out that they are missing in all brand new kits they ship . So will have to wait for them as well.
CAD drawings are now up to date on the svn including all mates and fasteners (outer shield to cavity, no cold plate support yet)
the following details are still missing:
other not important features:
I tried fitting the same curve (radiation and conduction in a simplified cavity setup) in MATLAB to a product of exponentials.
Two equations I tried:
where u(x) is the Heaviside step function. The second one simply allows for a time delay of t_0 in the second exponential.
I used fminsearch again with these new functions and got out very similar fits.
Essentially no difference. Also, there's very little difference between these fits and the single pole fit I did before. I tried to do weighted fitting like DMass suggested (multiplying my error vector by 1/time) but I couldn't get it to produce a good fit.
I did some simulations of my simplified cavity and surrounding enclosure while changing parameters. The parameters I changed were
I recorded the time constant (assuming a single pole and looking at where the curve reaches 1-1/e of the temperature step).
Temperature Step Sweep
I used 10 second time steps in the simulation and got the following table (with cavity emissivity e_cav = 0.7 and surrounding enclosure emissivity e_s = 0.1)
temp step [K]
time const [s]
As the temperature step gets smaller, variation between any step responses diminishes. If I normalize the responses so that the all start at 0 and end at 1, as the
following graph shows, the curves that correspond to smaller temperature steps lie on top of each other, whereas those corresponding to larger temperature steps differ more.
Top Graph: Temperature response to a boundary step function for different temperature step sizes, as a fraction of the temperature step.
Bottom Graph: Difference between fractional temperature response for different step sizes relative to temp response for Tstep=0.01K (absolute value)
So for smaller temp steps it appears that I am limited by the number of decimal places recorded in the solver. Currently the solver is outputting 10 decimal places,
which may be fewer than the number it's actually using in its calculations, so I may be able to go to even smaller temp steps if necessary. The
responses are essentially identical for 0.1 K temp steps and below. For larger temperature steps, it is likely that the linear approximation no longer holds,
and the time constant begins to depend on the temperature step.
I also tried a range of temperature steps for different input emissivities and saw the same trend.
I chose a temp step of 0.01K. Here are my data and fits:
For e_s=0.01: tau(e_cav)=(3100±30)*e_cav^(0.994±0.004)+(27870±50)
For e_s=0.05: tau(e_cav)=(3080± 20)*e_cav^(-0.996 ± 0.003)+(5980± 40)
For e_s=0.2: tau(e_cav)=(3070± 30)*e_cav^(-0.997 ± 0.003)+(1800± 40)
For a constant surroundings emissivity, the time constant goes as 1/e_cav. For a constant cavity emissivity, the time constant goes as 1/e_s.
This is consistent with David's calculation which gave , assuming small e1 and e2 such that the e1e2 term can be neglected.
Here's a 3-D Plot:
And here''s the actual data in a table:
And here''s the actual data in a table:
Do we want a cavity dummy which is equivalent in thermal mass and size for testing the thermal stuff?
I think it's a good idea to not play with the silicon rod at the beginning but still being able to measure the thermal characteristics right from the beginning.
(and test the mechanical stuff as well)
Is 1k too steep to pay for the dummy? If not we can just get a 4inch length of 2inch Si crystal from Virginia Semi
i thought more about a $20 aluminum tube with same length and diameter and matched thermal mass.
You should play with your new toys. You can start with the simplest cryogenic system, a dewar filled with LN2 (liquid nitrogen), attach the temperature sensors to a wood stick
(careful they can't slip off), wire them up to the electronics, and lower them into the the dewar, pausing for a minute or two when they are in the cold vapor, but have not touched
the liquid yet. Then start to test stuff, like how much temp fluctuation do you see in boiling liquid (probably not small, and how much temp gradient inside the dewar, etc.
Then borrow a simple test cryostat from somebody, mount two thermometers on the same piece of cold plate, cool down the cold plate with a modest stream of liquid, so
it take 30-60 minutes to reach 77K, and then monitor the temp difference between the two thermometers. etc. etc. Have fun.
Frank and I had been unintentionally fighting over the gannt chart because I failed to RTFM for the SVN. I will conspicuously omit most details for now, but all my updates since the gannt chart's inception have not gone into the svn because they were made to the wrong version.
I found a charger for my computer (am overseas) and thus was able to make my delinquent gannt chart update (not full syntax on the calls)):
One conspicuous absence on the chart was the Silicon spacer, though I thought this was already on it, not sure if I was mistaken or it if got deleted some time ago
today the remaining parts including the pump arrived. Assembled everything, programmed the turbo controller and tested the vacuum gauges.
Everything is working.
did some leak test of individual components like valves, gauges and also the first 24-pin vac feedthrough.
Seems to be OK, however i will seal it using the GE varnish to reduce the remaining leakage.
The only problem so far is that i can't read the good vacuum gauge with the software the provide. It reads the model number but not more.
However the same cable, software and computer can read the other one. I can also talk to it using a terminal program, so i guess it's a software problem.
checked with Rick on Friday. I've seen the 4 polished shields already and the PEEK parts for the cavity support.
Could have taken those parts already but will wait for everything beeing done to see if everything fits together, especially the cavity support itself which has to slide into the inner shield.
Monday they will start with the PTFE insulator and the cavity SS304 frame.
I made a simplified cryogenic model.
This is a 2D axisymmetric model. The inner cavity is silicon and the outder shields are both aluminum.
The dimensions are the same as those in the drawings for the physical cryogenic cavities being built.
I wanted to see how long the inner radiation shield took to cool to the temperature of the outer shield, first first with no conductive links between the cavity and either of the shields.
I set the entire system to 120K initially except for the outer edge of the outer radiation shield, which is set to 125K at time t=0. The cavity and inner shield cool to 120K during the duration of the simulation.
The inner radiation shield has changed by 1-1/e of the temperature step after 3.0 x 10^5 s or 83 hrs.
The cavity interior hadschanged by 1-1/e of the temp step after 4.25 x 10^5 s or 118 hrs.
This further supports Frank's finding that if time constants of the thermal coupling in this system are to be on the order of an hour, the dominating form of heat transfer will be conduction through the links between the shields and the cavity.
It makes sense that this system I have modeled would have two radiative poles. Fitting this on matlab may allow for identifying them separately from one another.
My next step in modeling is to add a conductive link and run the same simulation. Since I'm working in 2D axisymmetric mode, the link will be a thin disk. To account for the added cross-sectional area through which heat can flow, I'll decrease the thermal conductivity of the material accordingly.
finished the breakout box and testes (measured) all connections and for shorts between channels next to each other.
Missing parts for a final cryo test with the cryostat are:
picked them up from the machine shop and cleaned them today. The only parts missing are now the frame and small adapters for the cold plate connectors. Will start putting one shield assembly together next week to figure out where to drill additional holes for sensors and cables. Will drill those myself. We could also think about putting the assemblies in one of the refcav chambers with an RGA attached to see how dirty they are with all the sensors, heaters and cables.
below the scanned documents for the measured specs for our cavity coatings. Coating run number was V6-593 for the HR coating and V6-594 for the AR coating.
Measured specs @1550nm are R=0.05% and T=165ppm which corresponds to a cavity finesse of ~18000.
DYM: Is it common for the AR coating (%T) to be so wiggly on a ~nm scale? Not saying I think this is good or bad, I'm just used to seeing smoother traces and was curious, though maybe I just haven't looked at many %T graphs
don't know how other people can work in an environment like this. Currently it looks like the situation is out of control and the lab transforms into a dump as foreseen in post 218, some people work on that intensively. After one of the tables and desks didn't have a single square inch to work on anymore people started taking over the workbench and the other desk! Looks like we have to start daily cleaning from now on...
Oh, if you miss equipment in the EE shop check the crackle area. Chances are excellent to find what you are looking for, but you might have to dig a little bit.
As we have to give a plan to the PMA people by EOB this Tuesday for the cabinet and desk installation, we will toss everything which will be left on the desks by Tuesday afternoon. If you want to save something then move it ...
measured the Si cavity dimensions. All the mechanics was/is designed for a 2"x4" cavity, but the cavity is smaller and we might have to modify things...
hole at optical axis: 0.53"
venting hole: 0.25"
As we still don't have
i will stop working on the cryostat (incoming inspection, leak checks, cleaning, etc) and transfer the responsibility of all those things to Dmass
after monitoring the first script for a few days for crashes or other unexpected things i decided to adapt the code for the other gauge as well. Both scripts are running now and frames are written.
Additional things i would like to add (which are already in the code but not tested) and test/modify :
tried to close the exp chamber from the cryostat for baking. Gave up after 4h of trying different positions. The holes don't line up whatever i tried and i can't add more than ~15 screws at a time. There is probably only a single position where it works.
What's actually going on:
We need to figure out what to do about this...
Let's decide this tomorrow at the meeting
Use it as is and just use more elbow grease? - Not an option. We already have one ripped Helicoil and we even didn't use the thing. The risk of breaking the Helicoils is too high if we open it frequently (which we will do). If you break the thread for the Helicoil then goodbye Crystat and we need a new coldplate/LN2 chamber. We would also get all the metal parts from cutting the thread into the clearance hole on the sealing surface.
They are actually not 1/4-20, they are all #10-32.
Rana and Warren don't seem to think this is that big of a deal - suggested solution is to give it to Mike by the 40m to machine out the holes.
Rana suggests doing this through Steve, however Steve is sick. Shall I wait for him to get back, or try to hunt down Mike myself and get this in the Q. If we do this we don't draw on Steve's experience, and I don't know if there are any special instructions I would need to give him in doing the job. Holding off for now.