|
Hi uspl
I didn't do anything spectacular. I just adapted the basic from Peter Duffet Smith's book to suit my own needs. Those programs only took about 3 months to get running and about 3 years of refinements. The basic idea was the same as your requirements.
I wanted a way to automate a 8" Dobsonian in the days before goto telescopes. I got a dob11 driver and adapted it to the mount. The mount was first modified to have ball races in the altitude supports. Then I worked on the azimuth to make it roll on ball races with a proper spindle. There are full details on the mount mods on my web site.
Initially when controlled by a computer 386's were the best chips running at about 33 Mhz. This required a couple of tricks to get the processing fast enough. Also the DOB11 needed to be driven sequentially alt/azm and this was done once per second from the computer clock. the result was a zigzag movement when tracking at high resolution. I got around this by making a synthesized .1 second clock from the timer xtal. This did not work too well because the computer was too slow. When I got a 486 66Mhz computer it worked well.
At this stage there was no alignment routine. This took some time to get coded. Basically you take a guess at the scopes attitude and do a systematic fit till the error gets to zero. Normally about 5 refinements gets the error down. There is also a document on my web site explaining the procedure.
There was a bit of slippage on the azimuth drive so I designed a belt drive for it but never got around to implementing it.
I never really used the telescope after I got the system to work because by this time I had bought a LX200 10" scope and I gave the old dob to my daughter. She just uses the Dob 11 in its standard non computer driven form.
Barry
|