Alt-Az Tracking Device - Astronomy Help!

Hi Guys.

I'm a Mechanical Engineering student who has been tasked with developing a design that consists of tracking known constellations.

Part of this design is to create a control system which will read two user inputs, one being altitude (Right Ascension) and the other being azimuth (declination), and then move the telescope to the desired input. The range of motion for the altitude is 60 degrees of the vertical axis in either direction and the azimuthal range is 360 degrees in the horizontal plane. The telescope will have to be moved by two servo motors on each axis to the desired input. Once the telescope is n the desired position, it will have to constantly track those coordinates as the earth rotates at what i assume is a slower speed.

Being a first year Mechanical Engineering student, I have no idea on how to accomplish the feat and have been recommended to this site by a lecturer from the university.

If anyone can help me from understanding the concept, through the various stages what components i will require to achieve this, and ofcourse assist with the programmable code if one is necessary, that would be a huge help.


I'm sure we can help you with the motor control but you need to explain how this is supposed to work in a bit more detail From what I understand, the user enters (somehow) the two parameters you described . Then what happens ?

I suspect what you will need is a system that can drive 2 stepper motors, ie. one for azimuth and one for elevation.

You need to arrange the gearing on the steppers to give you the desired minimum movement step of the telescope (you'd need to ask your lecturer for this information)

Stepper motors don't generally have feedback, so either you need to add some detection in your mechanics for north and vertical, and drive the motors until the switches / opto breaker detect the position is at that defined point. Or you need to have some other method to define the position (I know that domestic telescopes have a multi-star alignment procedure you have to do each time you turn the telescope on).

i.e you have to manually motor the telescope to point exactly on 3 or 4 stars (which the telescope knows the position of), and the internal software then calculates where the motors are and also determines things like any inclination of the base of the telescope

Re: Tracking stars Obviously the system needs to keep time and also calculate the AZ and ELEV for the telescope to maintain pointing at that position

There is bound to be code to do this somewhere, but I suspect its non-trivial.

Do you need to write all the code your self, this seems well beyond a normal first year project in any discipline

If its just a mech eng project, perhaps you just need to design the mechanics to mount and gear the steppers and have the limit / position switches

And not really build this thing. (So I won't go into what you should be buying to build it)

Thanks guys for the useful information.

I've been told by my lecturer to explore the avenue of using servo motors, and controlling the motors by means of a micro controller and variable speed drives, without the use of a gearing system.

Is this possible to achieve?because i have my doubts, and if so is there any sources for further reading that would be able to assist me.

Regarding how the input will be entered is entirely up to me, I'm uncertain as to what interface to use, if any is needed.

My prior knowledge regarding the electronic side is very limited and any help is highly valued

Hi, google telescope tracker arduino barn door

What you ask has been done before with arduino and some very nifty engineering, stepper motors may be okay if the gearing is low enough.
If you are going to be doing astro-photography, you do not need the mount to stop-start during the time exposure.

Tom… :slight_smile:
Hope it helps.

Pyromaniac: I've been told by my lecturer to explore the avenue of using servo motors,

What do you mean by servo motors?

In the Arduino world they usually refer to hobby servos such as are used to operate the control surfaces in flying model aircraft.

In the engineering /CNC world it usually means expensive motors with digital encoders that are used instead of stepper motors.

I think you also need to tell us about the mechanical arrangement of the telescope axes. How does a motor make the telescope move?


Without gearing? That is most definitely not the way I would approach it. It would require some pretty strong and extremely accurate servos to properly track the stars. Earth rotates once per day (duh), that means 0,004167 degrees per second. Even at medium magnification, you need several smaller adjustments each second to avoid noticeable jumps and telescope shaking. No servo of reasonable price can achieve that precision. Gearing is the only reasonable way and to drive those gears you'd need either a DC motor with speed control and feedback or a proper stepper. The conventional way of driving a stepper is with a microcontroller, so you can at least meet that requirement.

EDIT: Is the alt-az mount a requirement? I would think that equatorial would be a significantly better option for star gazing.

you might also need a digital compass to align the scope to North.

I feature I would like is a reset function that would point to (remembers) the Polar Star

If you can afford it you can get ac servo motors with absolute encoders. You would need to initialize the system and count multiples of the maximum resolution of the encoder to determine system position. If it is a 16-bit encoder then "n" revolutions = n*65536. If it takes 20 revolutioins to get to a certain position, then the revcounter variable would contain the value "20" if you update it (revcounter++) every revolution. You would also need to do the reverse for the opposite direction. (revcounter--). The draw back of using servo motors is speed. (finding the optimum speed motor that is not too slow and not too fast). Personally I think a stepper motor using uC controlled microstepping modes with gearboxes is a better solution for this application.

Pyromaniac: two user inputs, one being altitude (Right Ascension) and the other being azimuth (declination), and then move the telescope to the desired input.

I think what you have done here is ably summarise the difference between an astronomer and a mechanical engineer. The first thing you need do is go back to your lecturer and ask him to recapitulate his instructions. While you are at it, you might discreetly drop the buzzwords "equatorial coordinates". If he looks blank, you will know that he doesn't know what he's talking about either.

What you appear to be looking for has been done a gazzillion times and is called a GoTo tracker whereby you can hop from one star to another in short order and thence observe the star as it it is tracked by turning the telescope on a (buzzwords) polar shaft that rotates to (buzzword) sidereal time. I believe these are two two separate operations. The polar shaft rotates through the night, while the telescope gets up to is antics thereon at your discretion. You will be faced with quite a bit of maths to do this but it will be about a million times less than you would face using AltAz coordinates.

I reckon an Arduino would be an excellent choice for all this and you will surely find help here from the stepper motor gurus, but you would be better off first checking Sky and Telescope magazine for a better understanding of what you are up against, and what your lecturer is not telling you.

Shows how little I know !

I had assumed the OP was talking about what @Nick_Pyner has so nicely described.

However unless the telescope is actually mounted on a polar axis I imagine you would need Alt/Az calcs to emulate one.