PFS Upgrade Series, Day 7: Reassembly Begins! And Two Problems.

This is part of a series of posts about upgrading an instrument at Las Campanas Observatory. If you want to start at the beginning, it's here.

Today started to feel like we were on the downhill part of a hill we had climbed up starting last week. Steve deemed the optical alignment as good as we could get it without more major changes (like, new lenses), and so this morning Jeff and I dismantled the wavefront sensor and replaced the reference mirror with the grating. Next, with Christoph and Steve's help, we replaced the top of the instrument (the roof), and took off the other short side panel so that Jeff could increase the size of the vacuum valve hole that feeds to the dewar from outside of the instrument. Before lunch we mounted the new dewar+CCD in its new mount in the instrument, and then over lunch we used a vacuum pump to decrease the pressure in the CCD dewar down to ~1x10^-6 millibar. After lunch we wanted to try and get images from the CCD, which we were able to do -- just a simple bias, but still exciting! -- but we ran into two problems that turned out to dominate the day.

PFS, slowly looking more like its complete self. Here Jeff is installing our new CCD/dewar. 

We have life! This is Christoph's computer, running the beautiful GUI he designed for PFS. The real data frame is shown on the left. It wasn't totally dark when we took it (just covered the front of the CCD with some cardboard and turned off the lights).

Hello beautiful new 10kx10k, 9 micron pixel CCD! I helped level you a few months ago, remember?

The first problem was a continuation of yesterday -- we still haven't gotten the slit assembly to function properly, so we can't take images through the slits. Big problem, we know! Shec built a connector that would run from inside the instrument (like, sort of behind the dewar), where the pre-slit assembly will plug into the slit assembly inside the instrument, to the slit assembly itself. This will help make it easier to make tweaks to the parts inside the slit assembly (which I showed pictures of yesterday), versus making a tweak, plugging it in and trying it, unplugging it and making a tweak and repeating; we can do everything with this effective extension cord. I helped Shec put it all together (and again, by help I mean mostly watch). Tomorrow we'll keep trying. I'm sure we'll get it  to work eventually...and then we can take real SPECTRA!

Above: One end of completed connector.
Below: Getting ready to solder the other end of the connector. Soldering, yay!

The second problem was that the CCD dewar did not seem to be holding the vacuum after we filled it with liquid nitrogen to cool it down. The vacuum gauge inside the electronics box was reading a steady 1.9x10^-3 mbars, when it was supposed to be going down to 0.1x10^-3 mbars as the charcoal getting inside the dewar cooled and adsorbed the gas remaining in the camera. Christoph was the one who noticed it really wasn't changing, which was worrisome. Jeff had the great idea of circumventing the software limit on the minimum vacuum needed to get the ion pump working, with the idea that maybe the dewar vacuum gauge needed to be reset, since we installed the new dewar/CCD. And it worked! We were able to get the ion pump to engage, which meant that the vacuum inside the dewar really was at or below 0.1x10^-3 mbars, and the ion pump itself gave us an independent reading of the vacuum inside the dewar, which was something like 1x10^-6 mbar or lower. Problem solved, yay!   We reset the dewar vacuum gauge so that it "knows" now what the real readings are. (Don't worry, Jeff reset the dewar vacuum software limit on the ion pump to where it was before, too.)

Ion pump is on and reading a pressure of below 10^-6 mbar, excellent.

We are happy when we solve problems! From l-r, Me, Christoph, and Steve. You can see the Magellan Telescopes in the background! We've been working in that group of red-brown buildings on the hillside below the telescopes.

What's left to do, at minimum, is get that darn slit assembly working, and figure out how to shim the dewar so the focus in the spectrograph is optimized. Hopefully we can get both of those things done tomorrow...? We'll see, stay tuned!  

Today's animal sightings included a viscacha! They like to hang out up in the rafters, away from prying eyes.