Platform, tiedown motion 15may08

apr07

Introduction:

project bd120 did a ten second vlbi cal scan using cband at 13:28:02 to 13:28:12 on 15may08 (UTC). After reducing the data they saw a 5 deg phase change in the 10 seconds. Could this phase jump have come from the platform motion? A 6 deg phase jump is 1 mm at 5 GHz.

How the platform height is controlled:

The platform height is controlled using the 3 tiedowns and 6 distomats that measure the height of the platform every 2 minutes. The sequence is:

At the start of an even two minutes start the distomat measurement. It can take up to 15 seconds.
Compute how far the platform has moved from the optimum position.
If the distance is greater than 1 mm then command the tiedowns to move to compensate for this motion.
Tiedown motion is 1.7 times the requested platform motion (cable and boom stretch). The tiedowns will spend 45 seconds doing the move. There is a ramp up to a velocity, a constant velocity portion, and then a ramp down to zero velocity.

The plots show the platform, tiedown motion around the calibration scan (.ps) (.pdf):

Top: The platform distance from the optimum platform height in mm.

* The data is sample once every two minutes.
The dashed green lines are spaced every two minutes when the distomat measurement starts.
The dashed blues lines are the threshold for tiedown motion.
the dashed red line is when the calibration scan occurred.

2nd: The tiedown motion during this period.

The units are in platform mm rather than tiedown mm (1.7 td mm per platform mm). I've also flipped the sign so the tiedown direction matches that of the platform.

3rd: blowup of the platform motion.
4th: blowup of the tiedown motion.

Conclusions:

The cal scan occurred during a 6 minute interval when there was no tiedown motion.
From 13.47 to 13.54 there was no tiedown motion.

The platform dropped 1.6 mm. This was only from thermal effects.
the drop rate was 1.6mm/120secs or .13 mm over a 10 second period.

The 5 degree phase jump was not from the platform motion.
If a measurement occurred during a 45 second tiedown motion then there could be a 2 mm change in 45 seconds (minus the change from the temperature change).
The tiedown motion computation does not try to predict how much the platform would move in the next two minutes. It tries to compensate for the last measurement. Predicting the motion could reduce the platform deviations from the optimum, but the current algorithm is probably more stable.

processing: x101/081515/vlbamotion.pro

 home_~phil