Re: [ATM] Tracking Systems for Astrophotography

["Mel Bartels" <mbartels@bbastrodesigns.com>]

> Mel, in a recent response to my earlier question, you intimated
> that the problems associated with tracking a moving object are
> difficult to resolve and costly.

No, that's not what I said.  Let me try again.  Tracking closed loop with a 
simple autoguider that has no knowledge of a tracking rate (tracking rate as 
expressed in terms of an internal motor control loop such as a servo control 
loop with PID filter, or more coarsely, a voltage regulated drive) is more 
difficult than tracking open loop with error correction.  Though a 
subjective conclusion, I daresay that all of those who build these kinds of 
systems will agree.

> While they are not astronomers, others, EE's for the most part,
> are suggesting that the circuitry needed for accurate optical
> tracking in an x and y axis does not require a novel solution.

Yes, for your EE friends, point them to this Sky and Telescope article: 
March '74, pg 191- (though the article describes a motor control loop set to 
purposefully drag the motor across the quad detector, so this is not a pure 
implementation of autoguiding against a guide star with no knowledge of a 
tracking rate)

However, if you wish to pursue this, one way to go about it would be for you 
to describe in functional terms the mechanism of the feedback loop between 
autoguider and motors.  Along the way don't forget about the exposure time 
required for solid state photodiodes to detect reasonably situated guide 
stars and the consequent impact on the motor control loop.  Tell your EE 
friends that they must maintain motor velocity to 2^20 resolution spatially 
and to 60 Hertz frequency in the time domain.  This gives 1/4 arcsecond 
tracking resolution - the defacto default in serious amateur drives.  They 
should also separate out the guide star's scintillation from its true 
motion, since again, there is no internal motor control loop to predict the 
star's position.

Of course, there's a wee bit more to this than simply a quadrant diode 
detector...

> They based their responses on what I told them. They understand
> the motions of the earth. However, it may well be that I left out

Oh?  EE's understand celestial mechanics?  The motions of Earth from a 
highest precision view are not understood even by astronomers.  To wit, the 
chaotic variance in UT - see Meeus Astronomical Algorithms ch. 10, and for 
that matter, the rest of his book.

Mel Bartels

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