I took an additional look at the noise being picked up on the LVDT’s for both Azimuth and Altitude and found that there was a ground loop through the iso-buffer that was inducing the noise. I removed the buffer and went directly to the ADC of the PLC. This change brought the operational noise down to a few hundred counts or less with all servos on. This appears to be an acceptable level.
The altitude loop gain was set for the proportional part of the PID loop. Additional tuning will be needed for the I and D terms when the actual windscreen is installed and wind loading and other structural resonance’s can be explored. Based on what I observed I believe the servo system will be "stiff" enough to meet the tolerances at the telescope truss.
One operational difficulty that will have to be addressed is the functioning of the telescope servo amplifier interlocks. The interlock for the bi-directional inhibit require someone press the reset on the front of the amplifier. I have spoken to Peter Prieto and he knows of the problem and has a new interlock PAL to install.
On departing the site, the windscreen system was operational in both Azimuth and Altitude and had tracked the telescope motion for over twelve hours while work on the telescope servo system progressed. A small amount of my time was spent on other miscellaneous system issues.
Last modified 05/27/98
boroski@fnal.gov