firmware for home-made guide camera

[bojan]

Some time ago I designed the hardware for CCD camera (using TC237, still in active production by Texas Instruments).
Camera has RS232 interface, and it is supplied from 12V...
The first firmware was designed as a driver only for CCD chip (in C, for Silicon Labs F330 processor, later replaced with faster Philips /u controller ), the actual application (that controls the camera) was written for PC in Delphi.
The whole thing was inspired by Audine project, long time ago...
After some initial success (first light on the bench) I abandoned the project due to lack of time and obviously better astrocameras that appeared on the market (Canon..)

Now I am thinking about the new firmware for this camera which would convert it into a guider for EQ6 (or any other LX200-compatible mount), directly coupled via RS232 interface - no PC in between.
I was thinking about conversion of Marti Niemi's Guider software (written in BASIC, and freeware, source code is also available, http://users.ameritech.net/mniemi000/auto.html) into C, and compiled for this hardware...

Is there anyone out there (with C and embedded apps skills) who would be willing to participate in this project?
B

[miketheobscure]

I'm interested in seeing how this project turns out, does that count? :D
A cheap or kit standalone autoguider's on a lot of people's wanted list, I think.

On a semi related matter: Do you know (or does anyone else know) if it's possible to get hold of the newer/bigger chips in small numbers? Maybe it's time to design a Cookbook/Audine 2? Something cooled & around 3 meg resolution would be nice. Might not be worth Kodak et als whiles to supply them, but that QHY guy must be getting them from somewhere. SBIG too for that matter.

- Mike

[Tandum]

I don't see how it's possible to have a guide camera without a PC. You need to pick a guide star in the first instance and calibrate which way the camera has been plugged into the scope.

If you are talking about embedding that functionality into the camera, why would you, you'd still need a pc to talk to it to set it up. Easier to make drivers for it that PHD/Maxim etc can use.

[miketheobscure]

Couldn't drift show the embedded software which way the camera was oriented after a few pixels worth of movement? Couldn't the embedded software then chose the brightest x number of points that moved together as its 'guide star'?

Focus could be with a near parafocal EP, or it could be auto too, or may not turn out to be that critical depending on the guiding algorithm ... all problems to solve to be sure, but the solution to them seems not too hard at first glance. (Famous last words in programming ;)). I still think this camera could/should be stand alone.

To answer Bojan's original question: I would help if I had the time spare, but sadly work at the moment means I can't make any commitments. In any case I moved from C to C# and PHP a few years ago and my embedded device programming's resticted to the one AVR chip.

On another semi related topic: Are you thinking about lowering the price on your 40D yet Tandum?

- Mike

[bojan]

Thanks guys for showing interest in this :-).
I also do not think there is a real need for PC at all here.
First, as Mike said, the firmware should automatically pick the star/pattern and start guiding. Actually, one of the existing firmware application for this camera is doing something similar (it is in proto stage still because there was a problem with response time of the system: it was too slow): it provides guiding for laser-guided drill, no PC.. How can anyone expect workers somewhere deep in the hole in the ground, full of mud, to fiddle with PC along with everything else? No way....
This current firmware "recognises" the bright laser spot (it assumes it is circular, brightest at its centre), then calculates its diameter and the x-y coordinates of the spot centre, and outputs them in analogue form (via 2 PWM outputs). This is used as a guiding signal for the drill head hydraulics. I believe it should not be very hard to make it suitable for astro guiding application.
It also automatically adjusts the exposure time so only the brightest spot is taken into account.
Second, I think the need for any kind of PC in the field is a nuisance and necessary evil at the moment. Too many wires, connections, driver issues... I spend half of my available time fiddling with all this (and I am not autoguiding at all yet...)
Yes, it provides a lot of (software) flexibility etc, but is this really necessary? and there is power supply issue as well...
Anyway, we shall see..

[AlexN]

Not true.

If you were at Cambroon this weekend, you'd have seen Paul's TV Guider (Yet another thing I now need!) Its a stand alone guiding setup, where a small camera is in the guide scope, and plugs into the TV Guider control box, then an ST4 lead from the TV Guider control box into the mount. Thats it.. It has a small (3" ish) LCD on it to display the field of view, and a few buttons to control it... but basically, thats the No-PC answer to autoguiding.

[bojan]

The first autoguiding systems were built around two photomultiplier tubes, beam splitter and two knife-edge masks... one for each direction.
Guiding star was held on the knife-edge in the closed loop system which was completely analogue.
Of course, the high voltage required for photomultiplier tubes to operate was abig disadvantage and actually a problem I would think (dew and high voltages.. brrrrrrrrr :-(

Anyway.. PC is not necessary ingredient.

[Tandum]

Call it a pc or a control box as Alex calls it, I believe you still need a way to change settings, focus and generally interact with the device. Even the basic program you list is designed to work on a dos pc. Good luck with your project.

[AlexN]

Fair call. :)

[miketheobscure]

Or call it one or two buttons and LEDs on the camera itself.



I don't see your point here - is this rhetorical or are you saying something about input to the program, or is it something about programming an embedded microcontroller?

And I take it that the answer to my question about the price of the camera you have for sale is no then?

- Mike

[amoeba]

A stand alone autoguider made from parallax components would be cool, all that would be required for others to make the device would be a comprehensive guide and the compiled source. It gets around any custom surface mounted pcb's and the like.

One their more recent proto boards incorporates video output too, so a nifty interface and visual star guide could possibly be used. (ref; http://www.parallax.com/tabid/407/Default.aspx)

Some food for thought anyhow :) I always like pie-in-the-sky ideas.

[bojan]

Well.. what did I say about Guider program was I was considering converting it (after necessary modifications and adding necessary functionality, whatever that may be) into embedded application for the existing processor which controls the CCD chip.
So I do not really see any controversy nor inconsistency here..

General interaction with the guider is of course necessary... After all, it needs pointing, at least :-)
Being the practical person, I only want to simplify the procedure of taking astro photographs, by avoiding current and all future problems with various OS's and compatibilities with drivers and cameras and who knows what.
No, I do not like PC's in the field .. not too much, that is, only as much as it is absolutely necessary :-) I do not want to play with it in the field and in the dark, what I want is to take the photos of stars... in the simplest possible way (according to what I want to capture, of course).
If the PC (and Microsoft etc) is the only answer, so be it...



All the best :-)