Next: Measurements - September
Up: No Title
Previous: Data recording with
In order to correct the pointing for the irregularities of the track, the TEL
task is required to read a file called MODEL1.dat. This is the track
model, a lookup table giving three parameters called F1, F2 and
F3 for each specified azimuth. F3 is directly the elevation correction, while
F1 and F2 are combined with the elevation to provide the azimuth correction
(see MTUN025 or Inclinometry : The Past and Future).
A software package to analyse inclinometry data and create a new track model
therefrom is called by the command
$ @ disk$data:[jcmtuser.inclinometry.progs]INCRED2
A menu is presented that allows processing and displaying of inclinometry data.
The shortest route to a track model is via options 2 and 9 :
- Option 2 uses part of the overlap region of a dataset to remove any
linear zero-point drift, removes any residual mean offset, converts the
mV output to arcseconds using the file SCALES.dat, and reduces the
data to the azimuths 0 - 360. Future versions may make temperature
corrections to each channel of output.
- Option 9 uses the data from the Y-axes of the inclinometers on the
A-frames (Channels 5 & 8) and from both the inclinometers on the TMU (1
& 2) to create a model based on the premises laid out in
Inclinometry : The Past and Future. The validity of this approach has
been questioned, but this option will be updated to reflect any later
changes in the theory of the construction of track models from
inclinometry data.
The current suite of programs available via INCRED2 are tailored to the
perceived performance of the inclinometers. In particular, it appears possible
to take data without going back and forth to
measure reference positions and without suffering large zero-point drifts.
The collection of point-wise data at a resolution of 1degree, with, say, 6
seconds spent waiting and sampling, and covering more than 400degree of azimuth
takes about 2 hours. The zero-point drifts that occur between the start and end
of such a run have not exceeded 2" in the data gathered so far. Assuming
that the zero-point drifts are linear in such a time (and from the available
data that seems a reasonable assumption), we can use option 2 of INCRED2 to
correct for this, and to convert the units from mV to ". Future
processing algorithms may include temperature corrections also.
Other options in the INCRED2 menu allow for
- display of data,
- averaging or differencing of similar datasets,
- derivation of statistics,
- testing of the symmetry of a dataset,
- calculation of the track profile,
- production of a track model,
- display of the 3 components of a model, and
- averaging or differencing of models.
Next: Measurements - September
Up: No Title
Previous: Data recording with
Iain Coulson
Thu Mar 13 13:57:23 HST 1997