JCMT Inclinometry

The track is composed of 14 segments of 2 sizes, all of which bow somewhat in the middle. The segmental shape and the joints, generating pointing excursions of size ~10", especially when the wheels pass over the joints. The antenna wheels are truncated tapered cones, made of thick, but machined, aluminum; with surface accuracies of (xxxx). The locations of the wheel/joint interactions, in terms of antenna azimuth are given here . (Note also that the track segments were welded together during 1997 ).

The track profile must be determined - by inclinometry - and appropriate corrections made to the pointing. The need for, use of, and reduction of data from the inclinometry system are described in several technical reports from the Pointing Project.

The inclinometers - their location, orientation and performance
Inclinometry is achieved with (wait for it !) inclinometers ! Our meters are 701-series (-2A) products of Applied Geomechanics Inc (though see also similar products by Wyler of Switzerland, as used, eg, at GBT). The inclinometers measure tilts in 2 orthogonal directions, and we usually have them at 3 locations :

• #2036 is on the TMU platform - which is as close as we can get to "on the elevation axis".
• (#4486 was once on the TMU itself, and was therefore 'tiltable' (see detail)).
• #1969 is on the Left A-frame, and
• #1970 is on the Right A-frame

• #0354 is our current Spare - recently (Jan 2004) repaired by the manufacturer.
• #5652 was purchased in Jan 2004. We will hold it as a spare, but its arrival prompts thoughts that perhaps there are new measurements we could make - not necessarily to define the track profile, but perhaps as diagnostics for other characterization of the antenna motion. Placing inclinometrers on the 'radial arms', between the central plinth and each wheel, was previously deemed useful in diagnosing problems with the wheels, but would obviously also provide the track profile relevant to that wheel (convolved with any vertical plinth motion, of course . . .).

In their normal configurations, the T, L and R inclinometers have their Y-axes pointing towards the front of the antenna/building and the X-axes pointing to the right. The radial-arm configuration should have the Y-axis pointing radially outwards, the X-axis pointing in the CW direction.

• TX gives roll (F2),
• TY gives pitch (F3) , and
• via the geometry of the telescope, LY, & RY give yaw (F1)

• TX, LX, RX - showing the strong correlation between all these tilts, indicating that the entire antenna structure rocks (for want of a better word) almost as a single unit
• TY, LY, RY - note the strong correlation between RY and TY (as may be expected since the tilting of the right-hand A-frame feeds into the elevation encoder which, via the SERVO, drives the elevation axis to compensate) and that the LY data is merely of the reverse sign and different in phase by 180 degrees. This is more easily seen if the LY data is plotted after the sign reversal and phase change . Such a good correspondence implies good symmetry in the response of the antenna to the track.
• F1, F2, F3 - the track model derived from TX, TY, LY, RY
• The long term stability of the track model.
• the differences in (F1,F2,F3) between a recent track model and the previous one. Although mainly noise, these spectra reveal joint-related spikes that have been disturbingly large - see the many such reports during 1997 and 1998 - although an explanation has recently emerged.

Each of the two orthogonal tilt meters is accurate to 1 microradians (0.2") - which we can confirm empirically. They operate effectively within about a half degree of horizontal. With the gain settings at low, and filter settings off, their output voltages scale as 20mV = ~1". Other documents discuss scales, and see stability in more depth.

For a discussion of the use of inclinometers in other scientific fields, see, for instance, the GRAVILUX Project, by Nicolas d'Oreye, and Nicolas's estimations for Earth-tidal tilts upon Hawaii

Follow these links for summaries of

• Data acquisition
• Data reduction
• Trouble-shooting.