Dual axis tracking controller?

[willock]

I know, the current opinion seems to be that active tracking isn't worth the expense and maintenance issues. Adding panels is simpler. That seems reasonable since the cost per watt of panels is at all time lows. There are however instances were the real estate to do that isn't available.

I have a couple of portable designs in mind I'd like to explore and have a simple timer relay system sketched out to control them. I might also try a sensor based controller

Does anyone have any experience with those dual axis tracking controllers i see for sale?

Btw, Will, this forum is looking pretty good! Congratulations.

 

[svetz]

We're DIYers! We can do things for fun, it doesn't always have to be about the cost! From what I've heard tracking should get about 20% more power throughout the year (like adding 2 panels for every 10, so you can see why folks say not cost effective).

They don't have to be optical trackers either. Saw a "clockwork" mechanism once, a valve opened in the morning long enough to fill a drum with water whose weight caused the panels to swing east. During the day it dripped out in time with the sun. Not sure how he ever got the thing balanced, and in a hurricane zone like I'm in best to have panels flush.

 

[gehowi]

If you have to buy a commercial system, I agree it’s not worth the expense.
I made 2 portable stations with 12*100W panels from all kind of recup parts (except the panels). (cost around 200$ + 900$ panels)

I track the sun / brightness centrally on the top of my roof and adjust the panel rotation to that data over wifi.

I point to the brightest place in sky, even away from the sun if it's clouded. This may add 50-100W more than pointing to the sun covered with a dark cloud. It depends on the weather type and cloud reflection.

Tracking panels in the long sunny days is very effective. In july my fixed panels, directed to south, start to deliver current around 9 30AM. At that time one tracked panel delivered already >1KW from 6AM to 9AM; same after 5PM.
In very clouded days it does not matter where they point...

The biggest problem is strong wind. I made my panels 5cm off center (right is longer than left). Strong winds (>60km/h) turn the panels automatically out of the wind to some degree. Of course they are not pointed to the sun anymore but that’s better than damage. The tracker adjust them later again automatically.

It’s also necessary to leave 3 cm gap between the panels to reduce lift from wind. If the wind is really stormy, then I move the panels behind the building and make them horizontal.

The total construction weights around 200kg and that makes it stable enough. The tires and the H profiles absorbs most of the shocks.
The wood never comes wet and makes it lighter. In 2 years, no damage or maintenance at all.

As you can see in the graph I get charge from 7AM to 7PM.

 

[Thor]

Gehowi: Very interesting. Can you provide more details on how you monitor sunlight position and adjust the panels?

 

[gehowi]

@Thor ,
There are many ways to do it. There are ready made sensors for tracking but I made an array of simple LDR’s. I placed them in small compartments like honey rate structure , in 360° in 2 rows.
You need only 180° and minimal 4 sensors to do the job but I use the same sensors as detector for car lights when dark. So I know any direction where a car or light comes from, even on the road 200m away.

I also use it as lightning detector. .They are all illuminated +- the same at the same time due to strong reflection.
With a mic I get the sound. This way I can +- calculate the distance of a coming storm.
A wind speedometer is connected as well.
It can help to protect my panels but actually I did not need it very much.. It's just added for fun.

I also measure the total brightness. So if a group of panels (6) does not deliver enough current due to shadowing or the wrong direction, I get a message.

For the LDRs I use an ESP32 with extra I2C analog IO. ESP 32 has only 8 ADCs available if you use wifi.
I read LDR’s continuously to be able to detect lightnings and light flashes but for suntracking 1 read in 10 minutes is enough..

The data is polled by another central ESP that makes decisions, logs and warnings.

An ESP 8266 on the panels drives the DC motors with full bridges depending on the position commands.
Waterproof potentiometers are used to get the position from the panels. Since my panels are mobile, I have to reset horizontal zero pos after a move. If I forget then they point to the wrong direction...

All communication is over wifi with UDP commands + acknowledge data.
Anything can be controlled from anywhere since ESP works with a browser.

 

[azgrogg]

I think having a system that changes the angle on a periodic basis as the sun changes pitch throughout the year (north to south). East to west daily seems like too much. I thought of just having panels mounted on a hinge and adjust them monthly up or down (depending on the month) to maintain an optimal angle. maybe supports with slots in them that you just adjust up or down monthly.

 

[gehowi]

@azgrogg
You are right. Angle adjustments depends on where you live and which season. In my area I don't have to adjust up/down from October to March. Just rotate.
But in summertime I must tilt from 40-10° every day to get max power. Of course you can do less. May be you lose 10% power.

 

[azgrogg]

@azgrogg
You are right. Angle adjustments depends on where you live and which season. In my area I don't have to adjust up/down from October to March. Just rotate.
But in summertime I must tilt from 40-10° every day to get max power. Of course you can do less. May be you lose 10% power.

I was just wanting to balance the amount of effort to get as best performance as possible. I think a system that can full track the sun daily, through put the day, would be really cool. Maximize as much as you can. But, if you have the real estate to put up more panels then I can understand that side as well.

 

[gehowi]

@azgrogg
The mechanics can be made very simple (depends on the weight). The driving motors need very little power during the day.
The wind is the only problem.

 

[svetz]

Ran some numbers on SAM to see what the benefits were mathematically, here are the results for a 7.6 kW array at latitude (25N) when compared to fixed flat (laying on the ground).

Seasonal Tilt: 13,549 kWh, +12.8%
Dual Axis: 17,761 kWh, +47.8%

Tilt angle	Fixed annual kWh, % increase	1 Axis annual kWh, % increase
0	12,016	15,812, +31.6%
15	12,994, +8.1%	16,507, +37.4%
25	13,140, +9.3%	16,646, + 38.5%

 

[gehowi]

Interesting.. I think this is correct for summertime. In wintertime it's far less than this.

 

[svetz]

SAM is probably the most accurate simulator there is available today. The numbers look high as those percentages are based on a fixed system at 0 tilt; most people are fixed at some tilt so you'd need to subtract off ~10% for the average system. Also, the numbers are "annual", so it's percentage based on yearly production.

 

[Wiley]

Gehowi, Love the solarium you have in the background. Did you build it or have someone build it for you?
On your mobile unit:
What did you use for the rotational pivot point? Did you simply butt weld the vertical support to the cross member? Could you show a close up of that portion of your unit? When you say it automatically wind cocks, does it disengage the drive assembly or does the wind force simply drive the motor. When it weather cocks does it disengage the tracking senors so it doesn't fight itself?

I'm planning to build something similar (but much less sophisticated). I stopped by a local wrecking yard yesterday and picked up a small axle and wheel as well as one of the newer style front wheel axles. All the vehicles I've owned had stub axles for the front wheels the newer style is a contained unit that bolts to the A frame/swing arm assembly. I'm in the dark ages regarding vehicles so the auto mechanics in the group will hopefully come back with the proper name. These are short units and attach to the vehicle with four bolts. I saw units with eight studs (off one ton or larger trucks) so weight of an array would be no issue. Opens up all sorts of possibilities for mounting rotating arrays.
I'm going to put together a single panel rotating/tracking array with the small axle as post and pivot, more as prove of concept to myself and work out the bugs before building a larger (for me four 100 watt panels). And always looking to see how others have made theirs.
Thanks, Wiley

 

[azgrogg]

Very interesting stuff.