Tracking

=> telescope drive...

Further data is being collected to record the variation in the periodic errors as a function of telescope position. The resulting measured amplitudes and offsets of the errors will be the basis for a look-up table with high resolution corrections for Alt and Az to improve the tracking as a function of position.

TO BE DONE NEXT 6 MONTHS:

Alt/az encoders periodic error table finalizing.

-Changed brushes for AZ motor 2 (entrnace). The base support has been modiffied for future easier service properly.

PI board development ----------

The PI boards, pre-amplifier boards and pre-load boards for the telescope drive are as old as the telescope and with no capability of detecting faults. To replace these with an up dated digital system would provide the facility to detect potential faults such as from the tachometers or symptom of potential motor brush failures. Also extra information about the telescope performance can be acquired which could help diagnosing possible future faults.

To assist in the development two text books on circuit analysis and system design where identified. It was fortunate that one of the books was available as a free download, even though this results in multiple files of each individual chapter and no proper index or content pages.

With the help of the book a full analysis of the complex analogue circuit design of the PI boards (azimuth and altitude) was made producing equations describing their operation in the continuous frequency domain, using Laplace transforms.

To produce a digital design the equations need to be converted to the discrete digital time domain described by Z transforms. Again this could be done on paper (eventually) but instead a free 30 day trial version of MATLAB was downloaded. Using this industrial standard package with its add-on control system toolbox the S transform equations where converted to the respective discrete time Z-transform versions. Plotting the complex frequency response of both the continuous and discrete functions, with the package, gives very similar results.

With the resulting equations it should be possible to programme some digital hardware to have the same functionality as the old analogue circuitry and then to add test points to extract suitable parameters to monitor its performance.

Initially a board can be designed to be a direct replacement to the existing analogue hardware, so requiring analogue to digital converters on the inputs and a digital to analogue converter on the output to make it compatible. The next step would be to add the pre-load board functionality to the software and accept digital data from the TCS rather than the existing analogue velocity reference voltage. The analogue tachometer signals and power amplifier analogue outputs would still need to be accommodated.