We are continuing the acceptance and development tests on the telescope begun in July, 2000. The work for our trip to Arizona, 28 January -- 3 February, 2002, (by Joel Eaton and Mike Williamson) had to do with further tests and modifications of the control system to make the telescope operate reliably during various kinds of failures, furether assessment of the oil consumption and requirements for topping off the oil, and further assessment of any mechanical problems that may have arisen from the first few months of automatic operation.

The work planned fell into three categories as follows:

1. MECHANICAL ADJUSTMENTS and augmentation of the telescope structure.
1. Run a heat-flow test for spectrograph enclosure by measuring ambient temperature with and without a known amount of heating in the enclosure. (We set up loggers to record the temperature and humidity in the enclosure on Jan 28 and let them record conditions without the extra heating for two days. We started the heating test [7.7 amps @ 120V] on 30 Jan at 5:45 PM MST and let it continue to 4:30PM on 2 Feb. We also left one of the loggers in the enclosure with four 40W lights on for a longer test.)
2. Verify placement of wells for legs to the spectrograph table for designing equipment for moving the spectrograph into its enclosure. Run a sway test on the building to see how much clearance there really is between the optical table and the walls. Put the concrete anchors in wells for attaching the extensions for the table legs. (We measured the positions of the wells and set up the sway test on 28 Jan. We laid out the positions of the anchor bolts on 2 Feb but did not have the right anchor bolts to put them in the floor. We also screwed the protective channel for the oil lines to the telescope to the floor in the main enclosure.)
3. Assess the oil consumption by the azimuth bearing further with data Lou Boyd has been collecting on level vs. ambient temperature. Assess the efficacy of the secondary oil-return pump and determine whether the controller hangs. (We used data collected by Boyd to determine that the thermal expansion is probably not the culprit; rather, as Boyd suggested, oil is not being pumped back into the tank as readily as we had hoped. Thus, we must allow for a greater variation of the level in the reservoir by placing the float switch deeper in the oil and we did so on 1 Feb.)
2. ELECTRICAL WIRING.
1. Add a device to the power input for the .JPEG server to let us cycle the power remotely. (Eaton-We acquired the device in early January, tested it in the lab, and shiped it to Fairborn. We tested it and put into operation on 30 Jan.)
2. Assess the feasibility of cycling the power on the control computers remotely. (The problem here is that we have two computers with ATX supplies that require someone physically to punch a button on the front of the box to turn on power to the motherboard so the computer will reboot after a power failure. [Why anybody would design shuch a ridiculous system is a mystery to us.] There may be solutions to this problem, but they're hardware-dependent. We have decided to live with this problem until we rationalize the control computers in a couple of months. This will involve replacing the executive computer [purcell, to be t13c] with a slower computer with a smaller disc and replacing the gateway computer with a faster CPU and a bigger disc.)
3. Assess the controls for the secondary oil-return system and make any repairs required. (This control seems to fail for some reason that isn't immediately apparent. The switch seems to have been jammed by the4 dust cover on the overflow tank, which we have removed.)
3. Work on the DRIVES and CONTROL SYSTEM.
1. Discuss further the operations and control system with Lou Boyd and decide what changes are necessary for him to keep track of the telescope's operation. (We determined that Williamson will write up a tutorial explaining the control system before our next visit to the observatory and discussed routine checks to make on the equipment in the meanwhile.)
2. Run tests of the UPS to test whether the program to check its status actually does what it's supposed to do. This will involve turning the power off and watching what comes out of the program. It will also mean deciding on how long to wait to shut the observatory when the power goes off. (We ran this test on 30 Jan by switching off the power line to the UPS for roughly two minutes and watching the output of the computer program that monitors the UPS. The program did what was expected of it, detecting the fault and keeping track of how long the UPS had been running on its batteries. An example of the output from the program to the power-status file was as follows: [fail 01010 122 355], where the 122 is the time in seconds that the UPS has been on battery power and the 355 is the time in seconds that the UPS thinks it has left on battery power. Whether the latter is realistic remains to be seen, however, since the number seemed to stabalize at 355. On 1 Feb we repeated the test with the telescope running to see if the control program would shude down the observatory gracefully in case of a power failure, and the test worked.)
3. Make changes to the control programs to stow the telescope if communication is lost with the executive computer and to restart the control programs gracefully if the computers reboot. Test these changes realistically. (Williamson fixed the programs and checked them on 1 Feb.)
4. Replace the executive computer and bring purcell back to Nashville to use in the student system. (We have deferred this task because the replacement was not ready in time. We have decided to switch the present gateway computer with purcell, take a similar motherboard out of a computer at the lab for a backup, and bring a faster computer from Nashville to serve as the gateway computer. Most of the hardware for these changes is already available but must still be scrounged up.)
5. Make more extensive backups of the various control programs and configuration files on the control computers and update the documentation. (Eaton began working on this job while at the observatory, completing the backups for the roof-control computer and most of the backups for the gateway computer. We have floppy discs for the other computers with many of the special files required on them, but finishing this system will require a great deal more work.)