Solar Tracker Project

[Xringer]

Hi Duane,
Thanks for dropping by again! I understand what you are saying,
since I've seen 20-50 watts coming off the array without any direct sunlight at all on the panels.
Just plain old blue sky. Seems weird.


Before this weekend, It almost always got well ahead of the sun..
(Because it was pre-pointed almost dead south when the sun cleared the trees.
It just wouldn't reverse back all the way to dead-on).

But, I did see a some increase in power, when I tweaked to track dead-on.

Maybe it does harvest more power while looking at the brightest area
of the sky, when direct sun isn't available.

In my experience, if the sun is out in the clear, I get a bit more power if it's aiming more accurately.

I'm willing to make the trade-off, less power when the sun is obscured and
more power when it's in the clear.

To heat my hot water, I need the load hot-plate to get a lot hotter than the water.
If I get get 300+ watts in there for 6 hours, I'll forgo the 20 to 50 watts
I can grab from clear sky harvesting.

Today, with a little filtering (with the cloth cover), it locked on when the sun cleared
the trees at 10:30 and it's still locked on tight right now (13:40).



If it keeps on like this, (running from zero to 2-3 degrees behind the sun) I'll be a happy man!
It's running so close to zero, the power out is at max.
Right now.. 442 watts! 1,508 BTU..

The only way I see to get any more out of a 500w PV, is to get that MPPT board built.


I just need to observe tracking during partly cloudy days, before adopting the idea of using a filter.
If it gets lost because of the filter, I'll go back to direct sun mode
Which works way better than the old tracker board I was using..


Thanks,
Rich

[Dave]

Hi Duane,
I just received your tracker (s24Vc3RI) down here in NJ. I've been following this thread (thanks Rich) in preparation of build a two pole mount. I am going to use a 36V linear actuator (I think). Could you recommend a rotary actuator (or design) and how it would apply in this application (single or two pole) and when the belt or chain is bolted to the opposite side of the sheave?

I was always told to ask a question because twenty other people were already thinking it.

So... In Rich's example, in a clear sky, the sun plus the sky is the greatest output. correct? Do I assume the reason his tracker leads the sun by 20 or 30 degrees, is because in the early part of the day, the tracker "locks on" the the brightest part of the sky, then, later on in the day maintains its "locked" position even though sun plus sky is brighter? Does the cloth filter Rich wraps the PCB in, delay the tracker from "locking on"? I guess I'm not clear about how it operates.

Finally, can you tell me how the Reverse inhibit feature might come into play in the above example.

David

[redrok]

Hi Dave, Rich, and All ;

(Sorry, but I don't have link posting privileges yet.)

The DG280 is a nice rotary actuator:
redrok(dot)com/led3xassm.htm#dg280
This is rated at about 23ft^2.
However, I would think that if there was a bearing on the second pole it could support much more weight. I would try to keep the panel array as narrow as possible to limit the "Stop Sign" wind flutter dynamic torque. The DG-280 does move a bit to fast and requires the use of the RI, Reverse Inhibit, option.

Of course, if you use the DG-280 as just a motor much larger arrays could be used.

Rich's tracking problems may be more than just sky radiation averaging. Especially since he describe being able to "Lock On" to the sun. The LED3X is simply not that smart. I suspect there are still reflection or refraction problems where there are multiple stable positions.

Much of the problem I see with thick wrinkly glass domes can be eliminated if one used a dome made from PETE plastic, the stuff used in pop bottle/peanut butter jars. This plastic really has a long lifetime in sunlight. I have some that have been outdoors for over 20 years. The Greenies hate this plastic for the same reason I like it.
Not that there is any thing wrong with being a Greenie.

Reverse Inhibits primary function is to prevent quick reversals from fast motor drives. It has limitations though. The overshoot must be less than about 1.25 degrees.

Duane
redrok(at)redrok(dot)com

[Xringer]

It was overcast this AM and the RedRock started seeking the bight clouds, starting about 8AM.
Steered away from the trees and was looking south when the sun came out around noon.
Then it went eastward to the sun and locked on, within a couple of degrees!

The sun was in and out all day. Every time I checked during bright sun, it was dead on track!

So far, I'm VERY please with this little sun-filter mod.
Right now, it's just wrapped around the outside of the glass 'jar'.
My plan is to install a filter inside, under the glass.. So it will last a while..

~~~

Also did some more insulation work on the boiler, might see some real
improvement with this one.

I figured the bare steel walls inside the large fire-wood compartment were
radiating heat into the compartment, warming the air.
That warm air would go up, and seep into the cracks and right out to the flue. Wasted heat.

Since I don't plan on burning any wood (unless we have a long black out),
I filled the whole solid fuel compartment with fiberglass insulation.

This should cut the inside surface radiation & warm air circulation a lot. Hourly heat loss should improve.


Currently, if the power goes out during cold weather, we have the 5KW
gas generator as backup, to run both heating systems (Oil & ASHP).
It will also handle the freezer and refrigerator.

In case of an extended emergency, It's going to take about 5 minutes
of messy work to restore the solid fuel burning side of the boiler.
We have a couple weeks worth of firewood in the backyard.
After that's gone..?. We go to the mall !!

http://www.woodboilers.com/admin/upl...sssection2.jpg

[Xringer]

I removed the cage around the glass.



Installed a semi-transparent plastic filter around the board, centered over the sensor LEDs.
(The filter is from the side of a plastic milk carton).

This should work like the cloth-wrap, masking out the tree line a bit,
and perhaps getting rid of some of the off-center tracking problems.

Tomorrow is going to be partly overcast in the AM and mostly overcast in the PM,
so we might see enough different conditions to determine the filter's usefulness.

Edit 9/9/2010:
Well, today was a bust. It was overcast about 90% of the day. I only saw a few minutes
of direct sun on the array, and it's tracking wasn't very encouraging.
I installed a cardboard light blocker inside the jar's bottom (inside the nose cone).
Just in case light is bouncing off the lower panels into the jar.
Friday looks to be partly cloudy all day and the weekend is looking great so far.
So, I should be able to get in some worthwhile tweaking.

Edit 9/11/2010:
I've also been tweaking the resistance in series with the motor. It's now at 3.3 ohms.
It still has some reversing once in a while, but it smooth, slow and short.

[Xringer]

9/10/2010:

A pretty nice day, partly cloudy.. In the 70s, which must be good for the PV, since it's doing 465 watts right now!
With 10 watts of cable loss counted, that's 475w my 500w PV is putting out!
That seems like I'm getting about 95% of the rated PV power!

Beside being a cool day, the RedRock light filter mod (with bottom light block) is doing great.
In the early hours, (before 10:30 tree blockage ends) it moved to center.
The area of brightest sky. Not much power there 50-100watts maybe..

But when the sun cleared the tree line at 10:30, it locked dead-on!

I'll be checking on it today, to see if I've done the final tweak..

If it's satisfactory, then I'll move on to building the MPPT board..
I plan to install it indoors, next to the load.
The higher voltage should cut the line-loss down a bit and increase the
power delivered to the hotplate load..
Heck I might be able to squeeze out 500w on a day like this!

[Daox]

What mppt board do you have?

[Xringer]

Originally Posted by Daox

What mppt board do you have?

I'm going to copy the pump controller design I found on the web.

http://ecorenovator.org/forum/solar-...-diy-mppt.html

I've been looking at the design and thinking about how to make
use of the parts that I have been slowly accumulating.

It's a real simple design that maximizes the amount of power
you can get out of your PV under a limited range of conditions.

A real MPPT design is self-adjusting. The pump design is not.
With this simple design, you get to select the voltage (with a pot)
that you want to deliver to your load.
If you select a voltage that gives maximum power output in typical
solar conditions, you will get peak output for 'part' of the day.

Since my array is tracking the sun, and the output voltage/power
doesn't really change much, during the prime 6 hours it's got clear sky
(unless it's cloudy).

I should be able to find a good compromise max power point voltage.
(I'm thinking around 66 to 70 volts).



Check out top plot line. If you set your Pump controller to take power from
the PV at exactly 13.3 volts, you would get max power during best solar input.
However, when the sun's input drops off, you wouldn't do too bad if you
were still using 13.3 volts.. According to these plots, the losses would not
be significant, considering the lower price of Manual Power Point Tracking.

With the array accurately tracking the sun, a simple pump controller
(MPPT board) should give me a significant power gain into my 10 ohm load.

From what I saw earlier today, I believe it's possible my array can deliver
500+ watts, into the load on good solar days..

But, hitting a 500w peak (or more) isn't the whole point. The board will allow
the harvest of more power under slightly less-than optimum conditions.
So, on hazy days, I might end up getting 400w, when I would normally only get 350w.

[aleedogar]

hello, i have taken solar tracker as my final project, would you please help me out with schematics.
Aleedogar@gmail.com

[AC_Hacker]

Originally Posted by Xringer

With 10 watts of cable loss counted, that's 475w my 500w PV is putting out!That seems like I'm getting about 95% of the rated PV power!

Do you have enough info yet to know how many watt-hours your system is producing per day, per week, per month?

-AC_Hacker