adding a "skew" setting?

[Paul-Bostwck]

Hey...

I'm hoping to find a tracking program for my tip and roll array collector. The extra tricky bit is that my collector is made of a group of tubes that contain optics that need two axis tracking (so far so good) But the tubes are anchored at the foot and turn on the central axis for east west and so tip up and down at the top for north south. The hard bit is installation of the array can be along non east west lines (in the graphic that is the foot line) so there needs to be compensating math for the difference... any ideas about how to implement those "corrections" in the code?

[Paul-Bostwck]

Those tubes are the collector surfaces... the sun is at the top of the image shining down on the tubes.

[graham22]

I suppose you could modify the run function in the linearActuator code, the current implementation assumes a linear movement/positioning, ie one inch of travel equals a given angle of movement, you could add a function to calculate a non-linear position to angle. I'm not sure if this would be accurate enough for what your doing, you might need actuators with position feedback...

On Tue, Nov 15, 2016 at 12:03 PM, Paul-Bostwck notifications@github.com wrote:

Hey...

I'm hoping to find a tracking program for my tip and roll array collector. The extra tricky bit is that my collector is made of a group of tubes that contain optics that need two axis tracking (so far so good) But the tubes are anchored at the foot and turn on the central axis for east west and so tip up and down at the top for north south. The hard bit is installation of the array can be along non east west lines (in the graphic that is the foot line) so there needs to be compensating math for the difference... any ideas about how to implement those "corrections" in the code? [image: screen shot 2016-11-14 at 6 39 06 pm] https://cloud.githubusercontent.com/assets/16853997/20291257/cd6ab14e-aa99-11e6-80b2-ed9284d1151f.png

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/graham22/SkyeTracker/issues/1, or mute the thread https://github.com/notifications/unsubscribe-auth/AFGATQPqMajXeJb0coU1wQIR6d2avktdks5q-eXGgaJpZM4Kyxgw .

Graham A. Ross (613) 769-1331

[Paul-Bostwck]

Hey Graham22 thanks for responding! I think you described it well (I have a hacking session with a clever friend this weekend and this way of describing the problem will likely help) I suspect you are right: the position feedback is very likely to be needed. I do have the advantage of a large diameter az wheel (.1m) and the length of the tubes (1.8m) as the radius of the elevation "gear" , so many stepper turns will add up to very small movements of the system but even still, "where are we" will need some kind of verification system. The non-linear aspect of the function is going to squeeze my brain hard! I guess both axis will have their own adaptation and together they should arrive at the aim... OK up too late. Thanks for the feedback. More as I know it!

Another image of my collector design, for your entertainment:

[Paul-Bostwck]

That is a section view of a section of one of the tubes.

[graham22]

The code assumes linear actuators