Sun Tracking / Heliostat Electronics Update (Work in Progress)

[Gabriel]

I have recently been working towards updating the sun tracking / heliostat electronics documentation. I am planning on adding a few new components along with rearranging some of the old ones. I think I will probably add some switches for manual control of the machines, but we'll see how that goes with the code first.

The pdf below shows what has already been done. So far, everything has been designed in Fritzing, but at some point I want to draw out a proper schematic and hopefully even put together a PCB for an Arduino shield.

http://www.cerebralmeltdown.com/wp-content/uploads//2012/06/SunTrackerCircuit_bb.pdf


Sometime in the near future, I am also planning on doing a design for the Arduino Mega since I have managed to get sun calculations with much better accuracy to work on it. Technically, the program would work on a regular Arduino as well, but the program is so large that it doesn't fit.

One thing to note is that the circuit in the above pdf uses the Arduino Duemilanove. The newer Arduino boards, like the Arduino Uno, have extra SDA and SCL pins which the RTC can use in place of analog pins 4 and 5. As far as I know, you can still use analog pins 4 and 5 on the Uno, but then you wouldn't be able to use them for other things.

On a final note, I still need to add electronics for controlling multiple machines. I think I should be able to use triacs in place of the mechanical relays used in the past, but am still waiting for them to arrive in the mail before I can try it out.

If anybody has any ideas for things that could be added or improved, let me know.

Thanks!

[iamtawon]

I think by using Triac or relay will cause the problem with step driver since it will just excite wrong coil
that will cause you to lose steps , but if your gear ratio is high the effect might be reduced (I guess)

do you have more than 1 heliostat to work with? I  can send you my drawing so you can replicate it easily
the motors I use is from U.S. store so that make it easier for you.

[Gabriel]

I have always turned the power on and off when the stepper motors aren't moving, but I haven't tried it with multiple heliostats (at least not recently). I have found that with my stepper motors and driver boards tend to gain a lot of steps when using 1/8 stepping. But when using 1/4 stepping (and possibly less), the effect hasn't been noticeable. Even after a reset, the reflected light always seemed to end up in the exact same spot.

Of course, using multiple heliostats could cause the driver board to lose its "coil position" (I guess). I'll have to look into it.

I had thought you had compensated for this problem in the code, which is what I was planning to do it it turned out to be too much of a problem. I wouldn't mind seeing your drawing. I'm sure that you understand this stuff a lot better than I do. 

[Paul L]

Hey Gabriel and fellow Cerebral Meltdowners!

   Just wanted to comment quickly to let you know that I check up on the forum from time to time to see how things are going and am super impressed with the progress you and everyone else on this forum is making.  I've taken down my heliostat (Gabriel's original version) for now as I am cycling across Canada starting Monday and then moving.  Wherever I end up though, I plan on setting her up again - maybe even a few more to keep her company, so it'd be nice to have more documentation for controlling multiple machines.  Looking forward to see how much progress you all make when I check back in a few of months!!

Paul     

[Brendan]

Just to throw two cents in...
- I've always thought a next step might be an ethernet shield that would communicate to a PC on a reachable network.  The PC would handle calculations and keep track of each heliostat's info.  The heliostats would check in, give it's number and the server would tell it where to move next.  There are pros (always using the most updated code with lots of features that wouldn't be possible on a microcontroller) and cons (internet connection is required) to the approach but if the design is toward larger arrays, it might make sense.
- Also, redrok suggested a trigger that would allow stats to pull off their target for awhile.  This would be useful if a stat were pointing at an existing solar array that was underused and cool -- like during winter.  If things got too hot, a temp sensor would tell the heliostat to point away for awhile.  Putting additional light on existing solar panel installations would provide some of the best 'return on investment' for a heliostat, especially if the array had microinverters (in a case of a single inverter, increasing sunlight on a few panels wouldn't help anything).

Anyway, hope everyone is well,
Brendan

[Gabriel]

Hey Brendan,

I recently bought a pair of those XBee radios with the idea that I might be able to use them to change targets from my laptop. To a certain extent, I just wanted to learn how to use them. They are actually pretty easy to set up. I don't think that it would be too much trouble to write a simple program for a laptop or desktop computer that could do some basic tasks.
I had never considered using the computer to do the calculations too.

Switching targets based on temperature would also be pretty easy.

I guess what we all really need is a good heliostat design that is easy to put together. All the features in the world won't make much of a difference if nobody has the machines to make use of them all.

Oh, and hey Paul! I must have missed your post earlier.

[Brendan]

Hey Gabriel,

Yes, I think a good design that's cheap and easy to assemble is the key.  With that, people can try out heliostats in different applications.  I think we shoot for kits that people can buy like in the reprap community.  (FYI, on that topic, I'm leaning to a MeldelMax over Ultimaker as of this second -- I keep flopping back and forth).

The most efficient approach for power generation is ultimately very large arrays with a power tower but the problem with that is it costs too much for individuals to set up on their own.  With a kit, lots of different people with different requirements could try things.  I'm hoping for a time lapse today and then have plans for a next version focused on being cheap and buildable.

I'll look forward to see what happens with the XBee's but an unsolicited comment is that for larger projects wireless is always harder to troubleshoot than wired.  Oh, and the other thing I wonder about is how best to do power distribution out to a set of 5-10 stats a ways remote from the house.  I'm thinking 12V like people use for path lighting but just throwing it out there as a consideration.  Anyhow, XBee-wise, one of my favorite things about this whole process is learning a bunch of components to try to make the larger project work.

Sorry for the jumbly post, gotta make pancakes for the kids now...

[Gabriel]

This link goes to an updated pdf of the electronics with the shift registers added. Everything is wired as it is in the latest version of the sun tracking heliostat program, V0.9.7.b. I still might change some things in the circuit, but I think it's getting close to being finished. I still need to figure out a good way of cycling power on and off to each set of stepper motors.

http://www.cerebralmeltdown.com/wp-content/uploads//2012/07/SunTrackerCircuit2.pdf

@Brendan
I have had a chance to play around some more with those XBees. So far, it looks like they will do everything that I would want them to do. It's pretty neat being able to turn an LED on and off from my laptop. Especially when that LED is on the opposite side of my apartment. I've started writing a program in Visual C++ that could be used to switch targets from a PC. I also think that it would be fairly easy to use the computer to more accurately calculate the sun's position, and then send that position to the Arduino. The Arduino could also switch to the less accurate calculations if the computer was off.

From what I've heard, the XBees are supposed to be very reliable, so I don't think that they would typically cause trouble. I would agree that wired would be better, but even if the heliostats are just fifty feet away it can be costly to run wires out to it if you have to bury the cable.

Heliostat's really don't use much power, so I'm sure that it would be possible to use just a small solar panel and a battery to avoid running a cable for power too. Of course, I guess that has its downside in that the battery has a limited life span.

[Gabriel]

I just found this MCP23017 - i2c 16 input/output port expander  (http://adafruit.com/products/732).
It looks like it would be better (or more accurately easier) than using the shift registers.

I'll have to order some when I get the chance. If they work out, which they should, I'll change the circuit.

[Gabriel]

The update to the Sun Tracking / Heliostat Electronics Circuit is one step closer to being completed. I have swapped out the shift registers for an i2c 16 input/output port expander because it is a little easier to wire together and doesn't use up any of the Arduino's pins since it shares the SDA and SCL pins with the RTC.
http://adafruit.com/products/732

I have also added a joystick into the mix. This makes it a lot easier to program in heliostat targets. Basically, you can now manually move the heliostat's reflection to where you want it just by using the joystick. Two push buttons were also added so that you can select which machine you want to control when you have a multi machine setup.
Note that you both activate and deactivate manual control via joystick by holding down both push buttons simultaneously. The target(s) are automatically saved to EEPROM when manual control via joystick is turned deactivated.
http://www.adafruit.com/products/245

I also added a 5K linear potentiometer which is used to switch between different sets of saved targets.

My intention with these new components is to make it so that you don't have to rely on a computer quite so much when you need to change something.

You can view the full pdf of the schematic at this link. I think it's getting pretty close to being finished, but I won't guarantee that it won't change at some point. Also, watch out for mistakes on my part.
http://www.cerebralmeltdown.com/wp-content/uploads//2012/07/SunTrackerCircuit3_bb.pdf

Note that at the time of this writing (7/16/12) I haven't uploaded the update to the Sun Tracking / Heliostat Program which utilizes the new components described in this circuit. I should have it available sometime within the next few days. I'll also try and give a better explanation for how all of this stuff works when I do.

[iamtawon]

WOW you really get things done very fast!
I guess the heat wave in US really encouraging right

[iamtawon]

So, now you're working with Arduino Mega or the old smaller one?
because in the latest software update post it seem like you're using Mega.

but in the circuit update it's still arduino uno

[Gabriel]

So far, I've managed to design everything so that it is compatible with both. The board in the picture is actually of an older Arduino, the one thing you have to watch out for with the Mega is that the SDA and SCL pins are not on Analog pins 4 and 5, or at least that aren't on mine which is the Mega Rev 3. They SCL and SDA pins are labeled though so they are easy to find.

Right now I'm in the process of soldering the circuit together so that I can move things back out into the sun to try some of the new features out.

[Gabriel]

Sun Tracking / Heliostat Electronics and Program Update

This is a "quick" video of some of the updates to the electronics and Arduino program for the Sun Tracking / Heliostat Project. There's not much to say here that I haven't already said in the video, so um, I guess that's all for now.

Let me know what you think.

[Brendan]

Latest work looks great, nice job.  Always impressive how fast you crank out results.

Love the joystick movement.  That plus the calibration by pointing at the sun and checking shadow look like a much easier way to set things up.

Also, from other thread, the approach of having one motor controller and switching output to different 'stats is great.  Ultimately the usefulness of heliostats is all about cost.

Brendan