Building a Multi-function Scope Controller

[Stub Mandrel]

Flushed by the success of my simple RA drive unit, I'm making a comprehensive (i.e. does everything) scope control box.

My aims is for it to work as a stand alone unit, but with the potential to add a USB link to a computer to allow control from Stellarium or guiding. It will also have a port for use with a dedicated guide cam.

So far I have made:

• Two simple gearboxes for RA and DEC steppers.
• Fitted a project box out with an embarrassing number of XLR jacks.
• A controller handset box with L,R,U,D buttons and a function/menu select button.
• Adapted a 64 x128 green screen display from an earlier project (that moved on to use a QVGA display)
• A main board around an Atmel AVR mega644P - like an Arduino on steroids, it will also have three polulu stepper drivers, two FET heater switches, a real time clock module, a USB link and plenty of 'expansion' ;-)

Possible expansions include a pair of relays for DSLR camera control. Anything 'lit up' will be brightness adjustable, mostly software but the main screen will be controlled by a big knob. Green is good as it can be much fainter than a red one and still visible.

The challenging bit is writing the software! Here's a taster:

[Ajohn]

Not sure I have looked at the right parts Neil - 8bit 20mHz and a lot of variants?

John

-

[ringz]

Don't forget, you will most likely need to heat the LCD screen if you are going to be using the controller under cold winter conditions.

They have a nasty tendency to become sluggish and unreadable else

Martyn.

[Stub Mandrel]

1 hour ago, ringz said:

Don't forget, you will most likely need to heat the LCD screen if you are going to be using the controller under cold winter conditions.

They have a nasty tendency to become sluggish and unreadable else

Martyn.

Interesting point, the 2x16 display on the one I used over this winter hasn't had any issues, but its only been down to about freezing. I suspect the rest of the electronics might keep the box a few degrees above the surroundings, but if I do get a problem at least I will know why now!

[Stub Mandrel]

Some progress - here are a set of screen grabs, not brilliant as the LCD doesn't photograph too well. I might change the larger fonts to a sans serif style.

 

[Stub Mandrel]

Some progress, I now seem to have a functional goto controller (I was distracted for a while because my DEC stepper is 96 not 48 steps so the best part of Sunday was looking for non-existent software errors... Now I need to test it, although I can add focusing, heaters and a 'wander' mode on cloudy nights. In time I will add USB control, hopefully!

Control panel middle knob reduces brightness to almost completely dark, right hand knob changes slew speed when under manual control.

Back panel. Sockets are RA, DEC, focus, guide (why use a dodgy standard socket when you can use an XLR - easy to fit and remove in the dark?) and then 12V heaters. Black switch will enable guiding to be switched in and out so it doesn't start leafing through menus on its own and develop a hideous alien intelligence:

The inside! All connections to PCBs use connectors except the 12V supply. Need to add a Shottky diode to avoid accidents. L/H board is the 'brains' R/H board operates the display and was from another project that 'elevated' to QVGA and bristles with unused I/O that might find functions in the long term, including logging data to serial or camera timer control. Or I might change the display to QVGA... You can see a lot of the I/O isn't wired up yet and vacant holes on the main PCB.

[ChrisLX200]

That's a very clean, professional looking job!

ChrisH

[Stub Mandrel]

1 hour ago, ChrisLX200 said:

That's a very clean, professional looking job!

ChrisH

Thanks! It's certainly not down to my usual standard ;-)

[cjdawson]

How about instead of using relays for the camera shutter release, use a circuit based on a couple of transisters and resisters.  Much lower power requirement.

 

[Stub Mandrel]

10 hours ago, cjdawson said:

How about instead of using relays for the camera shutter release, use a circuit based on a couple of transisters and resisters.  Much lower power requirement.

 

I think opt-isolated analogue switches would be better as I don't want to risk nuking a camera, but power isn't a big issue when hooked up to a big 12V battery and compared to running a pair of steppers.

[calli]

Respect!

I could imagine building the hardware, but the software? A Goto system? With all the calculus for the stars? Never.

Will follow this.

Cheers,
Carsten

 

[Stub Mandrel]

1 hour ago, calli said:

Respect!

I could imagine building the hardware, but the software? A Goto system? With all the calculus for the stars? Never.

Will follow this.

Cheers,
Carsten

 

Not as bad as it sounds, honestly! Most of the maths is pretty basic, it's just the slog of programming it all in, made harder by my aversion to C and preference for assembly language. In C it would be a doddle, but I suspect it would have used up a lot more of the chip's program memory.

The mount continuously steps around in RA at a rate you choose, naturally this is usually sidereal speed except for Lunar or Solar observing which don't need GOTO.

For GOTO you just need a catalogue of the RA and DEC of some interesting objects, including bright stars for alignment. You polar align then sync to a bright star. The number gets converted into a number representing the DEC and RA in terms of stepper steps from 0:0:0 0:0:0, which is just multiplication and addition, a number of steps for each hour, minute, second or degree, arc-minute etc..

Now select your target... and calculate its stepper numbers for RA and DEC.

Subtract the synced position from the current position, do some simple maths to make sure you don't go the 'long way round' (i.e. turn 255 degrees one way instead of 5 degrees the other) and send the required number of pulses to RA and DEC.

Obviously there's a bit more to it than that, but nothing really hard. In time I need to add things like avoiding the mount, backlash elimination (a little wiggle at the end) and meridian flipping.

It isn't worth the candle to use anything other than a simple HH:MM RA and DEG:SS Dec or work out slight movements of the stars as the field of view is so big working to arc-seconds is meaningless for anything other than hunting tiny objects with a super-powered scope.

 

The hard maths was in calculating sidereal time from UTC (which needs some custom long division and multiplication to keep accuracy down to a few seconds error) drawing clock faces (which needed me to write a sin/cos approximation)! But none of that actually does anything useful, it's just for show ;-)

What will be useful to add are things like a spiral search to locate a planet just outside the narrow FOV of a webcam.

A fun challenge will be to get it to display a little sky map.

 

[cjdawson]

7 hours ago, Stub Mandrel said:

I think opt-isolated analogue switches would be better as I don't want to risk nuking a camera, but power isn't a big issue when hooked up to a big 12V battery and compared to running a pair of steppers.

Nothing stopping you opto isolating the transistor inputs that will have the same effect in a smaller package that you could get with relays.

[Stub Mandrel]

2 hours ago, cjdawson said:

Nothing stopping you opto isolating the transistor inputs that will have the same effect in a smaller package that you could get with relays.

True, but you have to be sure which way round the switches are wired, analogue switches and relays will work whichever way they are polarised..

These only take 10mA so can run direct from a microprocessor pin, are pretty tiny, and only cost 92p.

But , to be honest I am nowhere near designing  a camera interface yet - I will probably prefer to use my hand controller.