Inclinometry was done as follows :

    Datasets  El Dirn     HST     mean leg temperatures   Humidity
                          start     start   middle   end       %
  ( 20060129  89  cw      19:40      5.1      5.1    5.0      22 )
    20060821  89  cw      00:29      8.0      7.4    7.0      --

The data stream recorded is from the 3 active inclinometers only: no strain gauge data, for instance.

The inclinometry acquisition software has been rewritten during the shutdown although the code and interaction are similar. The standard CW run used the standard CW azfile, but there is no datum for az=10.09 - a problem to be fixed later; the datum for az=370.09 was copied in its place.

The raw data from channels 1,4,7 (inclinometer X-axes), 2,5,8 (Y-axes) and 3,6,9 (temperature) are shown below,
first, for the previous data of 20060129:

then for the new data of 20060821:

Channel-by-channel comparisons follow:
- TX, LX, RX

Changes can be seen in the X-axis data, but the data quality look good. The TY data (Ch2) are of significantly lower amplitude than before; recall that 200mv on the raw data scales corresponds to 10". The new RY data is affected by a couple of spikes, but otherwise is little changed from before. The big concern is the large scale, wide-spread noise through the second half of the LY (Ch 5) data. The largest spikes (to negative values) are not shown in full, but reach -2000mV and -3000mV. Such a noise problem is reminiscent of the effects of the underloading of the wheels. A central bearing load adjustment planned for last week (Aug 16) was not performed, but seems necessary to ensure full wheel loading. Also, (as TCC points out) the occurrence of the noise/spikes in the new data in quadrant 5 (360 < az < 450 ) and not in quadrant 1 suggest at least a contribution from the cable-wrapper in the plinth.

Because of the uncertain validity of the data for azimuths greater than about 300 degrees, I have not generated a track model of any sort. As a result the old model (of 20060129) is still in place, with a potential for causing pointing errors whose magnitude may be estimated from the plots above, but which could easily reach 10" in some places. Such a situation will undoubtedly frustrate other parts of the Functional Check-Out.

An updated track model should be of high priority and further inclinometry runs are scheduled to achieve that.

20060822
An attempt to acquire CW and CCW data in the morning hours of the 22nd failed, with the data stream ending during the CW run at az=306. The data showed similar behaviour to those above :

with the level of noise increaing as the antenna enters the 4th quadrant. There is also one substantial movement of the antenna at az=194.89 caught by the LY and RY tilt meters. There is a cluster of track joints in the vicinity, although not precisely co-incident with this event.

In the azimuth range they have in common, the differences between the two recent CW runs are shown below:

There has been an evolution in the TY (Ch2) behaviour, however, with the TY profile now essentially reproducing that of 20060129:

The other channels that feed the track model seem unchanged from two days ago at azimuths less than 200. The 'noise' beyond that in LY and RY seems just that: inconsistent from one run to the next. (The event in the 20060822 data at az=194.89 runs off-scale).

A central bearing adjustment is scheduled for this morning, possibly with inclinometry to follow.