Jeff and I started looking at HAM ISI signal chains to see if there were enough AA inputs for the planned upgrades during this install cycle. We realized that we still have all of the cabling for the original distribution of STS channels, where each STS is connected to one of the BSC front-ends, the signal goes through a distribution chassis that then sends copies of that STS to each of the other front-ends before sending signal to the source front-end AA, sort a physical ipc. This means each HAM rack has 3 db9 AA ports taken up by a signal distribution chain we no longer use, each BSC has 2.

Until today, ITMY sent its collection of 1 directly connected STS set of channels (the HAM2 (or A) STS) and 2 distributed sets (ITMY (or B) from BSC2 and HAM5 (or C) from BSC3) over IPC to SEIPROC, where it was calibrated, then used for a sensor correction signal that got sent back out to all of the corner seismic. This means we can free up AA ports by removing the rat-king STS distribution set up and just use the IPCs. Doing this required redistributing the IPCs for each STS to the front end model the ground sensor was directly connected to.

Arnaud is working on the ECR (with the helpful cartoon of what is being changed in p. 2 here), but I wanted to test the changes needed to do this today. I disconnected the extra connections for the shared STS signals between each BSC, leaving just the AA cable for each chambers directly connected STS. I then split out the 9 IPCs from ITMY to each chamber. Dave and EJ found a bug where I had mistakenly used an ADC selector instead or a bus, so I also fixed that in a couple places on the top level for each chamber. Somehow this had been before this point, but today meant the wrong channels got picked out for the BS STS IPCs. Dave also had to do some editing of some other files that had to do with the IPCs, because the sender models were changing, but not the names of the IPCs.  

As far as I can tell at the moment this seems to be working. Corner is vented, so testing sensor correction performance will be difficult, but I can at least check that the signals look like ground motion and do a couple other checks.