SUMMARY
Inclinometry was taken during the early morning of 16 May 1999 and a new track model was created and installed on the 17th. Concerns over a CCW 'loss of symmetry' are abated somewhat.

   Datasets     HST start      mean leg temperatures   Humidity
                               start   middle   end       %
 ( inc990512cw    00:22         5.5     4.4     3.8      40  )
   inc990516ccw   02:43         1.4     1.8     2.1      45
   inc990516cw    05:53         2.1     2.3     2.6      45  

The difference between the clockwise models 990516cw and 990512cw is shown below :

• The rms scatters in the (F1,F2,F3) differences are (1.03", 0.27", 0.47").
• The saw-tooth nature of the F1 difference looks like the inverse of that seen in the 990508-990502 difference.
• The new track model was installed at 13:10 HST on 19990517.

Symmetry
The symmetry remains good (see the plot ), if not at its perfect best.

We add the CCW data to the database started on 990512 ,

      dataset       scatter about the best-fit str line to
                    Ch2 -vs- Ch8   Ch5(phase-flipped) -vs- Ch8
               i.e.  TY -vs- RY     LY(phase-flipped) -vs- RY
    ------------     ----------     -------------------------
    inc971031cw         0.59"                 1.56"
    inc971031ccw        0.79                  2.37

    inc980820cw         0.76                  4.30
    inc980820ccw        0.74                  4.98

    inc990511cw         0.73"                 1.41"
    inc990511ccw        1.05                  1.87
    inc990512cw         0.72                  1.56

    inc990516cw         0.70                  1.65
    inc990516ccw        0.84                  1.68

CW-CCW
The cw-ccw data are shown below :

As seen on 990512 , the differences in (F2,F3) are small, with rms scatters of (0.19", 0.37"), and the difference in F1 is similar to that seen on all occasions when CW and CCW data have been taken :

As stated on 990512 , the amplitudes on the scale of the track segments seems relatively unchanged since 980820, but the overall amplitude now (6") is more like that at 971031 rather than at 980820 (9").

It's going to be difficult to show more CW-CCW datasets clearly in this format. I have experimented with displaying the differences between these CW-CCW differences between one dataset pair and another, but the terminology gets confusing and the displays are not significantly more helpful.