Vacuum & Cooling

Since there was no time for a two-stage baking of NOTCam after the May 2011 opening (which was not planned), and since the vacuum obtained was a merely 2$^{-4}$ mbar, we had decided to do a proper baking at the next planned opening. This took place in March to install the two new blocking filters. However, the time between NOTCam runs was not long enough to do a two-stage baking, so we did only a one-stage 8 days baking at 37 degrees. It resulted in no better vacuum than we had from before, an order of magnitude worse than what we had from 2008 onward. We need a longer-term monitoring to say whether we have out-gassing or micro-leaks. A proper two-stage baking is foreseen for a future opening (none planned for the moment).

There was a major concern about the LN$_2$ holding time for NOTCam since early January 2012. In storage (not mounted) NOTCam is refilled with LN$_2$ typically every 5th day. This interval became shorter and shorter, and each time it was only possible to fill less and less LN$_2$ into the tank. The problem with the short holding time and the apparently small capacity of the tank fixed itself every time that NOTCam was mounted on the telescope. Thus, moving the instrument around seemed to help, and it was speculated that water had accumulated in the LN$_2$ tank, forming a frozen layer which would divide the tank in two parts and inhibit filling the whole tank.

When opening NOTCam in March, the relative humidity measured at the fill and vent tubes of the LN$_2$ tank was about 40% while the ambient air in the clean room was at 15%. Based on a few experiments it seemed that: 1) water in the tank may freeze as a layer, and 2) pumping a bottle filled with water will take out the water eventually. Therefore, the tank was pumped until the relative humidity of the air coming out of the tank was $<$15%. Since then the holding time is better than ever (10 days between LN$_2$ fillings!), but we need to monitor this for a longer period.

Due to problems with the block of TMS sensors which include the water temperature of the PTR cooler for NOTCam we do not have a way to detect if the PTR is on or not. An independent system was developed to monitor the NOTCam PTR compressor status. It was decided that the water temperature would provide the most useful information about the compressor since it defines both if the unit is running and also its performance. When this system is implemented the water temperature will be read using a digital temperature sensor and provided over the Ethernet.

A Helium leak was detected in the PTR system. The leak was found to originated in a crack in the flexible part of the supply hose, close to the instrument end, near the previous point of repair, a very vulnerable location where hoses are bent during operation. The tube was cut 10 cm, re-welded, vacuum pumped and refilled with Helium.