Active solar tracking

[Bill Roosa]

Hey all
Got an idea a while back to increase my solar collection capability. I did not want to cover the decks with panels so mounting them on a pole off the stern like a radar antenna comes to mind. This led to the idea of using an active 2-axis sun tracker to maximize collection (or minimize the panel size) and a recent discussion on the site on hot water got me to thinking about mounting a thermal collector under the solar electric collector panel for that purpose.
This would be a DIYer install as I'm (as most of you know) handy and cheap.
Thoughts?

 

[finding41]

Is your boat on a mooring? That may be a problem. It may use more power turning to find the sun then it collects.
If you are slipped that's another story. I've seen some guy on YouTube using a mirror to increase the power of his solar panel. Probably just up to max capacity of the panel. Made a big difference....
Great idea though.

 

[jepomer]

I mounted a 100 watt solar panel flat onto the stern pulpit. Using a boosting MPPT charge controller, the 5 KW battery bank used to power the electric motor. Since I sail typically twice a week, the batteries were always back to full charge after a couple of days. The boat was on a mooring this season.

 

[JohnShannon]

Solar panels obey a cosine law. So Pout=Poutmax X Cos(theta)
So what?
Perfect alignment 100% power
90 degrees off (edge on) 0 power
but surprisingly 45 degrees off 70% power

so the tracker might not buy you quite as much as you think.

http://www.powerfromthesun.net/Book/chapter04/chapter04.html

 

[Bill Roosa]

Hey JohnShannon. It is a little more complicated than just the area presented to the sun. There is also a micro structure to silicon that "prefers" more direct "aka normal to the crystal" lighting. Other sites are indicating a 20-50% increase in output. given mpp is 5-8% increase max it seems to be a better solution.

 

[kenn]

The ideal solar power system doesn't supply 'enough' power, it supplies too much power (in ideal conditions), so that in less than ideal conditions (latitude, haze, overcast) it still supplies useful amounts of energy. So the cost (in energy and $$$) to have automated aiming might well exceed the potential gains from aiming, especially when you're near charged up and the additional energy can't be stored.

But it's still true that the optimum angle will maximize the power output. An interesting compromise might be to occasionally adjust the solar panel array's vertical angle to one setting suitable for the latitude, then use automation to keep the panel bearing pointed at the sun's bearing. This could be an efficient low speed drive system; it could even be a sea-powered system (like a self-steering rig) when you're underway.

 

[JohnShannon]

NMEA interface to GPS so you know where you are and what time it is.
So you know where the sun is. 3 axis G sensor to determine tilt of the panel, couple of motors some software and done!

 

[RichH]

The ideal solar power system doesn't supply 'enough' power, it supplies too much power (in ideal conditions), so that in less than ideal conditions (latitude, haze, overcast) it still supplies useful amounts of energy. So the cost (in energy and $$$) to have automated aiming might well exceed the potential gains from aiming, especially when you're near charged up and the additional energy can't be stored.

But it's still true that the optimum angle will maximize the power output. An interesting compromise might be to occasionally adjust the solar panel array's vertical angle to one setting suitable for the latitude, then use automation to keep the panel bearing pointed at the sun's bearing. This could be an efficient low speed drive system; it could even be a sea-powered system (like a self-steering rig) when you're underway.

The sun rises and sets, goes from 0° to 40° to 0° zenith angle; Plus, goes from E to W from ~0°± to ~180°±. For the middle latitudes, a 3 angle (x, y, z) tracker would allow a much smaller panel area .... 1 / Sin 40 X (~Sin 180) = 1 / .64 X ~2 = approx. 1/3 to 1/4 the size of a stationary panel

 

[Bill Roosa]

The power requirements are pretty small, two 5 volt 1 amp motors that pulse on occasionally. the control system is simplicity itself, some LEDs used as photo diodes feed operational amplifiers ($2.00 for chip of 4) the opamp outputs feed a turn east-west h-bridge which powers the motor to turn the panel into the sun. Same same for the other axis. My test stand draws a smoking 0.1 amps at 9 volts (they operate from 5-32 volts so going up in voltage is not a problem) and it would be pulsing for about a second every couple of minutes. Stops where it is at a dusk the slews to the sunrise once the sun comes up. If the motors check out and I can get the friction down it should even be able to operate while under sail and tacking. it will probably be playing catch up on a tack/jibe but you don't spend that much time doing tacks and jibes as a general rule.
As for the ideal solar panel produces more than you need, I'd agree but I (and most others) don't have that much deck space. alternately we have too much demand for the supply. I don't see this changing any time soon so what I'm really hoping for is a massive extension in the time between running the motor. the panel in question outputs 16 amps max so I'm thinking I could leave the reefer on all the time and the freezer on during the day. this would go a long way to extend my time between recharges.
now if I stick a thermo solar panel under the photovoltaic one..... I'd have hot water too.
You don't need a three axis tracker BTW. An altitude-azimuth is sufficient to track the entire sky. On a boat you have to address the entire sky as you would not know which way you might be pointing relative to the sun.

 

[Sumner]

Do you have pictures yet of your test rig? Are you working with one large panel or a number? If a number would the frame be one frame or multiple? Making the frame/frames moveable and also stable on high wind days will be interesting. Also lots of items to make work in a corrosive environment.

Hope you post pictures and results. I'm interested,

Sum

[FONT=Arial, sans-serif]============================[/FONT]

Our Endeavour 37

Our MacGregor 26-S Pages

Our Trips to Utah, Idaho, Canada, Florida

Mac-Venture Links

 

[RichH]

The power requirements are pretty small, two 5 volt 1 amp motors that pulse on occasionally. the control system is simplicity itself, some LEDs used as photo diodes feed operational amplifiers ($2.00 for chip of 4) the opamp outputs feed a turn east-west h-bridge which powers the motor to turn the panel into the sun. Same same for the other axis. My test stand draws a smoking 0.1 amps at 9 volts (they operate from 5-32 volts so going up in voltage is not a problem) and it would be pulsing for about a second every couple of minutes. Stops where it is at a dusk the slews to the sunrise once the sun comes up. If the motors check out and I can get the friction down it should even be able to operate while under sail and tacking. it will probably be playing catch up on a tack/jibe but you don't spend that much time doing tacks and jibes as a general rule.
As for the ideal solar panel produces more than you need, I'd agree but I (and most others) don't have that much deck space. alternately we have too much demand for the supply. I don't see this changing any time soon so what I'm really hoping for is a massive extension in the time between running the motor. the panel in question outputs 16 amps max so I'm thinking I could leave the reefer on all the time and the freezer on during the day. this would go a long way to extend my time between recharges.
now if I stick a thermo solar panel under the photovoltaic one..... I'd have hot water too.
You don't need a three axis tracker BTW. An altitude-azimuth is sufficient to track the entire sky. On a boat you have to address the entire sky as you would not know which way you might be pointing relative to the sun.

Bill, what sources or examples of 2D trackers are you talking about? pics., URLs, schematics, etc. ? 12 vdc stepper motors?
Very interesting concept .... especially for a sailboat where 'windage' is always a problem, so optimizing solar panel orientation to keep down 'drag' is a big 'plus'. ;-)

 

[walt]

I’ve been watching an 80 watt solar panel in my side yard that is connected to a MPPT grid tie inverter recently. I’m watching watts produced using Kill O watt meter (October in Denver)

Some interesting observations. This is about one week apart in October and on a warm day I was getting 60 watts out of the 80 watt panel with the sun shining directly on the panel. The panel got significantly warm even though it’s October in Denver. However, today is somewhat chilly and I’m getting 69 watts in almost the same conditions. Ie, Im getting 15% more power and as far as I can tell, its just a colder day. The extra power is likely because the panel is cooler so the maximum power voltage is higher. On the day I was getting about 60 watts, if I had the panel pointed directly straight up, I would have been getting only 30 watts. Ie, I would be only getting about 50% of what Im getting if the panel was not angled toward the lower sun in the sky.

If you use the solar where the sun is directly overhead and it’s cold, MPPT with panels pointed straight up would be your best bang for the buck - but of course that condition doesn’t really exist. If it’s warm, it appears MPPT is significantly degraded. In the winter months where the sun is at a lower angle, its cooler but the sun is at a lower angle and if the panels are pointing directly up, you might only get 50% out compared to an optimum azimuth angled panel.

So especially in the winter and in a warmer place, I think some sort of tracking device might give the best bang for the "parasitic power buck". I.e., MPPT consumes about 100 ma just to run and if your panel pointing system also used about 100 ma, the panel pointing system would produce higher power. In the summer, a panel pointed straight up with MPPT would probably be best – and a lot simpler.

For this idea, why not just fix the azimuth since that really depends on the time of year and only control the rotation (with whatever method)?

 

[Maine Sail]

Bill,

No doubt that aiming panels can increase the daily output. The Solar Stick already does this, though without the "electric tracking". IIRC they do tracking in their military Stik systems. I have spent considerable time talking with the designer, Brian, and they limit the pole to 100W. These poles are ROBUST, well built and bomb proof but still they don't want any more than 100W of panels due to windage etc... To support a panel capable if 16A would require some serious engineering for it to rotate and articulate well..

I have quoted a few "Stiks" but can't get anyone to bite. I just did 125 watts "invisible" on a standard bimini using a Solbian panel. The owner did not want "ugly" as he has a gorgeous Chuck Pain designed Able 42..

I have done over 400W using bimini & davits and it was quite unobtrusive. You'd be amazed at how well flat mounted panels do. This time of year I am getting full rated output for about 4:15 per day or averaging almost 34Ah / day from a Kyocera 140. In June I can get well over 40Ah's per day.

Last week, in late October here in Maine, after installing the 125W Solbian, the owner was getting 2.1A at 3:30 in the afternoon with his 125W Solbian panel and Genasun GV-10 controller..

Sun tracking can be cool just have not yet seen it built to handle very large panels or executed to hold up in the marine environment. Build some that hold up, work and can handle a couple hundred watts and I bet I will have some buyers...

 

[kenn]

The sun rises and sets, goes from 0° to 40° to 0° zenith angle; Plus, goes from E to W from ~0°± to ~180°±. For the middle latitudes, a 3 angle (x, y, z) tracker would allow a much smaller panel area .... 1 / Sin 40 X (~Sin 180) = 1 / .64 X ~2 = approx. 1/3 to 1/4 the size of a stationary panel

That's assuming ideal sun and perfect tracking, every day! (Ok a few of you are lucky enough to have the near perfect sun part). And the smaller panel would never match the peak output of a larger panel, if/when you ever need extra charging from your solar array.

Some considerations:
- how much of the morning or evening sun produces enough energy to justify 2-axis tracking?
- are there measurable benefits to 2-axis tracking while underway, factoring in pitch and roll ?
- sacrificing simplicity and redundancy (more fixed panels) for complexity (smaller tracking panel) ?
- how much do you depend on your solar array for serious charging, as opposed to 'topping', where most of the solar energy can't be accepted anyway?

I'm not trying to be a wet blanket, just looking to assess whether the effort to put in a competent solar tracking system would realistically justify the effort. (besides the cool/fun part. of course)

 

[Bill Roosa]

MS
Thanks for the insights. I had a talk with Brian when he showed up at the Pentagon to show off his wares on industry day. a very interesting guy.

Kenn
I suggest you go to the wiki site, http://en.wikipedia.org/wiki/Solar_tracker
and yes cool fun does play a part.

Rich
see above, and, in keeping with MS really good idea of KISS, the only real novel concept is the using LEDs as light sensors. http://solartracker.greenwatts.info/solar_tracker_cds_new_sensor.htm Devilishly clever of them. The folks at the tracker link are all enamored with programing a PIC microcomputer though. Me, not so much. think analog control circuit...... so an "east side" and "west side" (altitude is similar) sensor that just produces a voltage signal. Apply some basic electrical engineering to it like it needs to have a null when pointed at the sun and east and west voltages are the same and you come up with a 6 LED, and one opamp circuit (plus some power filtering and resistors to keep things "in range") the big ticket items where the motors with gearboxes. at $4 each

 

[kenn]

Hi Bill,

I've been following the solar engineering exploits of Circuit Cellar founder/publisher Steve Ciarcia as he adds solar power to his house, so I appreciate the potential gains from tracking. At least for on-land arrays.

That idea of using LEDs as sun sensors is cool. It's certainly easier to find LEDs than it is to get photo-diodes or CdS photo-resistors. And the LEDs have lenses built-in.

Finally, if you have $60 to spare, this dev kit and this accompanying book are a great kickstart into PICs for small applications. Don't let the book's juvenile title put you off; I found it to be really useful. Also, the dev kit is now officially deprecated, but it's still very useful, and capable of working with most current low-end PICs. Arduino is also a good platform to start with.

 

[Bill Roosa]

hey Kenn
I have programed a PIC before. Very handy little devices for most applications where "stuff needs to be tweaked" a lot. This solar tracker is so single purpose that a simple analog circuit is more than sufficient. and if it is to going to work in a marine environment it should be operator reparable and of simple construction. I can't imagine a simpler circuit of 12 LED (alt + azi), 1 opamp, 2 h-bridges and 2 motors. and no fancy program to trouble shoot/de bug

 

[sea_casper]

NMEA interface to GPS so you know where you are and what time it is.
So you know where the sun is. 3 axis G sensor to determine tilt of the panel, couple of motors some software and done!

GPS gives you location but not heading, so it doesn't know where the sun is relative to the boat heading, you also need a magnetic heading sensor.

 

[Bob J.]

Wouldn't a tracking system on a boat get dizzy?

 

[Bill Roosa]

You guys are attacking the problem the hard way round. I really don't care where I am on the globe, I don't care which way the boat is headed either. All I really have to know is the following:
Is the panel aimed east or west of the sun
is the panel aimed above or below the sun.

two sets of directional light sensors give that information. So if the sun is west turn west till it ain't and if the panel is not angled high enough aim it higher.
if the boat heals over or I tack her or she sails on the anchor I don't care as all that does is require the panel to recenter itself.

I suppose you could use GPS lat and long and an ephemeris and a clock to understand exactly where the sun is in the sky and then integrate that with a magnetic bearing (calibrated to the panel) and keep track of where I'm pointed in relation to the boat and disengage the system when the sun is below the horizon........and do the math but that sounds a) hard and b) expensive.

Children of the digital age!!!!!!

Go analog and you have more sail time. And is that not the goal????