ACIS Large Vent Valve Ops Summary

Bob Goeke
Latest update: Sun Aug 23 21:03:48 1998 EST

Instrument Status

ACIS was reintegrated onto the ISIM and immediately transported to the TRW M4 Thermal/Vacuum chamber for a test following the Mission Deployment scenario. After 4 days in vacuum, but always at ambient temperatures, the chamber was opened to fix a design flaw in the water injection system. The same day (Friday, August 14) the chamber was pumped again, the small vent valve opened, and the cold soak phase of the test started at 1530 PDT on Saturday, August 15. Since Monday the entire ISIM has been effectively in the cold survival state: Detector Housing at -72C, Focal Plane at -125C, and the Vent Valve Assembly at -40C.

On Saturday, August 22, at 1300 PDT we turned on ACIS for the first time in a week, commanding the Detector Housing to -60C and the Focal Plane to -50C. At 1315 the SIM operator turned on the trim heaters used to raise the ISIM to its normal (cold) operating condition. We then started to rehearse the door opening procedure by using a 10% duty cycle pulse. When this was ineffective at raising the valve actuator temperature above 30C, we increase the duty cycle to 20%. This, too, seemed inadequate at the time. So that we could finally open the valve and do the (detector housing) pump out we came for, we increased the duty cycle to 40% which did achieve out goal of warming the actuator to functional temperatures. A plot of the three cycles is shown below, followed by an expansion of the final cycle. I manually halted the cycles after the actuator hit 125C without any indication that the ValveTransition switch, much less the ValveOpen switch, had toggled.

The Full Three Cycles of the Open Vent Valve Actuator

3 Cycle Data Plot

The Third Attempt to Open the Vent Valve

1 Cycle Data Plot

The lower 4 traces are bits from the PSMC serial telemetry stream, and have been corrected for the 1 major frame phase shift to be time consistent with the temperature data. One sees the actuator power being applied about every 140s or a bit over 4 major frames in the lowest trace, and the ValveClosed switch tripping as the temperature reaches 75C. It is by design that the ValveOpen switch is invalid until the ValveClosed switch has transitioned (the door mechanism works exactly the same way); this is why one sees the . . . . WAIT A BLINKIN' MINUTE. I'VE BEEN BIT BY THE SIDE A BUG !!! Side A actuator controls, as the experts will recall, do funny things with their logic; unlike Side B which just passes them switches through, and doen't even bother with the concept of "transition". Cycling the power to the actuator, as I was doing to practice the "slow door opening" scenario, mucks up the Side A logic with the result that one never sees the transition, and never sees the open. You ask to heat the actuator and, in spite of it really being open already, it just keeps heating! When you switch over to Side B, truth prevails and the valve is shown to be fully open, so that's an end to it. BTW, one can see fairly clearly the wax phase transition both heating and cooling in this data (perhaps around 75C going up and 80C coming down), which at least gives on confidence that the thermistor is fairly well coupled to the chemistry of interest.

There is a lesson in all this, and Ed and I know it well. No amount of thinking -- about failure modes, or procedures, or much of anything else associated with flight hardware -- is worth squat compared to running the hardware, day and night. And this homework assignment was meant to demonstrate that no new procedure, no matter how well thought out, should be applied to said flight hardware late in the game without a lot of trepidation -- and probably not even then.

Doing It Right

If one uses the B side of the PSMC to drive the Vent Valve actuators, things work more like one might expect. The controller shuts down power to the wax when the far switch trips, no matter how many times you cycle through the ON/OFF commands. This data was taken using a 30% duty cycle with the valve housing temperature at -26C and the bus voltage at 30.5VDC.

A Demonstration of B Side Pulsing

1 Cycle Data Plot

There have been 76 accesses to this page since 9/14/2022.