It’s after 8:00 PM Friday January 13 and we’re all working hard to try and close up Black Cat one more time. The second cool down only lasted a few days as we found some problems we wanted to address before leaving the cryostat cold for a year. So we opened up, tore the camera apart yet again, and started working on revisions and engineered some fixes for the problems we found.
One major aspect of the revisions is swapping out some old 90 GHz pixels for some new ones, which is labor intensive and really tricky. That’s why I’m writing right now, actually – we’re all ready to go to rebuild the focal plane and the folks doing the 90-rework are finishing up. Another major problem was that we had microphonics pickup problems. Essentially, as the receiver was bumped or moved, parts of the camera would start vibrating like tuning forks at particular frequencies. This vibrational energy has to get dumped somewhere, and it gets deposited on our cold plate, which heats it up. We actually ran some experiments where we blasted the receiver with sound waves at different frequencies and watched the temperature of the cold plate in real time. Sure enough, there were particular resonance frequencies that made the cold plate rise almost instantly by 10 mK or so. That doesn’t sound like a lot, but it eats into our pixel sensitivity pretty hard so we needed to remedy the situation.
To fix the vibrating parts, (the readout towers for the 150 modules and the 90 GHz pixels), I designed a set of solid rings that the towers can screw into, essentially making all the readout towers one solid piece. The more rigid the part, the higher the resonance frequency, and all we want to do is push the resonance frequency above vibrations from moving the telescope and the like so parts in the camera won't shake and heat everything up. So these rings should do the trick.
It’ll probably take 4 hours or so to reassemble the camera, so we’re planning on getting that far, maybe bolting the camera into the cryostat , and then calling it a night. We’ll attach readout thermometry, readout cables, and close up the cryostat and start pumping down tomorrow morning. It’s getting pretty late in the season, and we still have to measure the bandpasses of our detectors, how efficient they are, and their polarization angles before we start the process of handing the reins over to the two winter over’s for SPT this year. It’s a lot to do, so this will be our last close up for the season barring a major catastrophe– the cryostat will remain cold until next fall. So, we all have to be super careful over the next 20 hours or so… Here goes!
Edit: It's now nearly 5:30 AM on Saturday. Didn't quite make it as far as we wanted. Got the detectors installed with the new rings that make everything more rigid (they work amazingly well, by the way), and we covered the green circuit board for the 90 GHz detectors with a blanket of kapton-coated aluminized mylar to reflect light away from the board - otherwise the board would absorb it and heat up. But the back side of the camera is ready to go and other folks will go out in a few hours to keep going as the rest of us sleep to meet them after lunch and actually close the cryostat. Exciting!
Before I head to bed, here are some pictures of the new additions to the back side of the receiver. The rings and braces holding all of the readout towers up make them as solid as a rock. Before, you could flick the towers and feel them vibrating with your finger. Now you flick them and they just take it, no questions asked. I'm really pleased with how well they turned out. And the mylar blanket Liz and I made turned out really awesome too, though Liz made it work as well as it did - I just coated the back side with kapton. We affectionately call it the "Big Top" tent. I'll have a new post in a few days after we close up and the 18 hour work days stop. Almost there...
The readout towers shake around a lot and heat up the focal plane, which is bad. Jay had the idea to connect all the towers by a solid ring to make them act like one big piece. From that concept, I designed a set of rings and braces that hold the towers together. Here’s the lower ring installed. Two aluminum bars hold the center tower to the two towers next to it. There are two more towers, not installed, that attach to the rings by those spaceship-looking braces screwed onto the outside of the ring.
The green circuit board in the above picture absorbs stray light really well, which may have also been heating our camera. To keep the circuit board from absorbing the extra light, we covered it in the “Big Top Tent,” a mylar blanket that reflects light away from the circuit board. The back side is coated in two layers of kapton tape so the aluminum coated mylar doesn’t short out all the pins going to the 90 GHz pixels. The Big Top rests on the lower tower support ring in the above picture to keep it floating above all those nasty pins. Once the tent was installed, we placed the second bracing ring on and attached everything to the two outer readout towers. It works great! The readout towers are made of “LC” boards, (an inductor [L] and capacitor [C] in series with a resistor has a particular resonance frequency, and we use these different frequencies to bias up many different detectors with the same set of wires). I like to call this new configuration with the rings and everything “LC Henge.”