The big news of the month is that Dan, Merle and I have found the problem with the Memory boards. It was an improper reset sequence generated by miss-drawing of the logic. Once the proper sequence is performed, the memory boards just work. At least the "good" ones work. Some still need to be fixed. But it now appears to be possible. This is a real weight off my mind. I now know how to do things to make more systems run. It is just a question of working through the details. No more mystery "a good system suddenly stops working" problems. I hope!
We built a nice test fixture so that we could uncover the Memory board problem. I bought a computer cart and we put a computer on it with a I/O board so that we could write test patterns into the memory. Just figuring out what we needed to do solved the problem since we had to take time to really learn how the memory board worked in order to design tests for it. As usual, setting up to really solve a problem solves it.
45 CDs were generated during May by TOM1. The stack is now 524 high. On average, I seem to get a new variable from each pair processed.
ROB has been moved out on the porch and many things seem to work. I was stopped from getting first light by the lack of a good camera. I will have to go to work on the dozen or so potentially working cameras sitting in the basement.
We have started setting things up so that we can run TOM2. Everything is there, I just have to try to make it go.
This has been data analysis month. Michael made the last necessary fix to the pipeline, and I have been running disks through it. I am about 60% through the accumulated data stack. I picked a not so modest sub set and wrote it to CD and sent out 26 copies (so far). Since I picked a dense field, it turned out to contain about 25% of the stars measured for the engineering run. We are finding lots of new variable stars in the data on this CD. It will no doubt point to improvements to be made. A brief analysis of the data quality is in TN-83.