[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [ATM] Tracking Systems for Astrophotography



The original question posed open loop guiding against closed loop guiding 
with no internal tracking model.   The discussion is being turned more 
towards the latter having a straightforward practical implementation.  It's 
an interesting question because if solvable in a simple inexpensive manner 
might be significant.

Open loop guiding consists of some sort of internal tracking model, be it 
mechanical with an equatorial mount, or, computed as in an altazimuth mount. 
Some sort of predictive error correction is added.  One can achieve unguided 
tracking to sub arcsecond accuracy from 30 seconds to 10 minutes this way. 
The latter was reached recently by someone using my freeware software with 
careful attention to the mechanics.

Closed loop with no internal tracking model in the proposed scenario 
consists of a quadrant detector feeding back on/off signals to a motor 
controller.  Others and I have thought about how to make this work easily, 
but I don't have any breakthrough thoughts.  There's the issue of 
sensitivity of the detector, a tremendous problem in not only getting the 
feedback loop to work at a high enough frequency, but also simply in finding 
bright enough guide stars next to the desired object to track on.  Then 
there's the issue of simple on/off feedback loop: no proportional, and no 
derivative is available.  All you know is that the guide star is either on 
one side of the divider or on the other.  You don't know how far.  So the 
control loop has to be stable enough to handle large deviations to either 
side but yet stable enough to keep the guide star basically sitting on the 
divider, half the time on and half the time off.  Remember that perfect 
guiding to a certain threshold is required - not even a single deviation out 
of bounds is allowable.  I think that's hard but maybe someone can point me 
to a successful high precision system using simple on/off feedback and the 
theory behind its control loop.


Mel Bartels 

_______________________________________________
ATM mailing list http://www.atmlist.net/