Making a Dual Axis Tracker

Fallingaway · May 10, 2021

There’s a website if you look but you can calculate the wind loads for you area on solar panels. Those things are like sails when they’re up in the air and tilted. Lots of pressure on that post mount. Good luck though

jasonhc73 · May 10, 2021

Fallingaway said:
There’s a website if you look but you can calculate the wind loads for you area on solar panels. Those things are like sails when they’re up in the air and tilted. Lots of pressure on that post mount. Good luck though

My tracker has an anemometer on it. When the wind is fast, I have it go to park, flat to the horizon.

Hedges · May 10, 2021

jasonhc73 said:
My tracker has an anemometer on it. When the wind is fast, I have it go to park, flat to the horizon.

I see you have faith in your model of which way the wind blows

Yes, should reduce forces. But you might want to brace or strap in that position for very high winds.

Fallingaway said:
There’s a website if you look but you can calculate the wind loads for you area on solar panels. Those things are like sails when they’re up in the air and tilted. Lots of pressure on that post mount. Good luck though

There's a reason I like trusses. I learned about triangles as a child.
Any single pole needs to provide enough torsional load. For a fixed array, the uprights and horizontals can have diagonal braces. Then only the middle of spans with panels have to withstand torsion.

For a tracking mount, the upright can be braced like the poles of a teepee. The spans of the moving parts can be braced with trusses. Then everything rides on a gimbal, and uniform wind pressure doesn't try to rotate it. There can be flutter, and braces preventing the tracking motion could be put up when a storm is brewing.

Fallingaway · May 10, 2021

sounds like you have it under control. What did you go with for the tracker? I built my own with an arduous nano but I have incorporated an anomeeter. Is yours out of he box or diy?

Fallingaway · May 10, 2021

Be glad to share my code for it if anyone wants to use it or better yet, improve on it. Pretty basic though. Takes samples at a given rate, averages them and then moves actuator if one side is out of a set tolerance. Single axis now but duel would be simple to make happen

jasonhc73 · May 10, 2021

Fallingaway said:
sounds like you have it under control. What did you go with for the tracker? I built my own with an arduous nano but I have incorporated an anomeeter. Is yours out of he box or diy?

I used an out-of-the-box one. SZMWKJ

Almost ready to put the pivots on the post. Entirely overbuilt, especially with the pillow bearings.

Fallingaway · May 10, 2021

Pillow bearings are definitely “overbuilt” but I had a hard time finding anything else that was similar but not “overbuilt” either it just wouldn’t work or way over complicated or needed custom machined pieces. The other options I found were pre made solutions from solar companies. Anything with “solar” in the title seems to carry a high premium too, no matter how basic it is so that was put in principle alone. Pillow bearings are simple, budget friendly, and they work. I got this from a guy on eBay. He designed and builds them. It works well but doesn’t have the lcd screen and bells and whistles. It works great though.

jasonhc73 · May 10, 2021

Fallingaway said:
Pillow bearings are definitely “overbuilt” but I had a hard time finding anything else that was similar but not “overbuilt” either it just wouldn’t work or way over complicated or needed custom machined pieces. The other options I found were pre made solutions from solar companies. Anything with “solar” in the title seems to carry a high premium too, no matter how basic it is so that was put in principle alone. Pillow bearings are simple, budget friendly, and they work. I got this from a guy on eBay. He designed and builds them. It works well but doesn’t have the lcd screen and bells and whistles. It works great though.

That tracker is awesome. Simple and functional.

Here it is (mostly) ready to mount on the stick in the ground.

A bunch of bolts and super strut and super strut nuts. All in all about 50 lbs. As HD as this is I am pretty certain this joint is capable of holding eight or nine 72 cell panels, as for now, just one. (See what happens when you have leftover parts and are OCD.)

This is how the E/W actuator attaches to the N/S bar.

This is the main bar attachment where the E/W pillow bearings pivot on the N/S pillow bearings.

This is the little super strut attached to the big super strut and the pillow bearing.

This orange vertical bar is temporary it is for testing usefulness. It works perfectly. A longer version is what the E/W actuator will attach to.

All in all, it is rather tall for the "joint". Drawing in Sketchup and building in real life is not quite the same. ?

Two cross beams for the panel to actually attach to are all the is left to attach.

jasonhc73 · May 11, 2021

It's on the stick.

Tomorrow (maybe) the actuators and the purlins for the panel.

Hedges · May 11, 2021

jasonhc73 said:
It's on the stick.

View attachment 48711

Looks to me like the usual tilt of the panel has the frame covering zerk fitting.
Spinning that pillow block around could make it easier to access.

jasonhc73 · May 17, 2021

Almost there.
I ordered a new electrical box to mount the controller.

The most important parts are there now. I have an old lawn mower battery that is getting charged with an itty bitty 12v maintainer PV, but I think this battery is done for. I'm thinking about just using a 12V AC adapter to power the PV tracker.

The panel is on a grid-tied micro-inverter. I have A/C right there already so I figure I'll just use it that way.

Looking south, this is the E/W actuator. I'll be cutting off the supper strut just below the mount. I'll use that piece to stick out the top for the sensor and the wind sensor. This is a 1200/1600N actuator, is noisy, and kind of on the "cheap" side, but it works fine.

Looking east, the N/S actuator. This is a 6kN actuator and is silent, it is also fairly slow, but it is very precise.

jasonhc73 · May 17, 2021

Hedges said:
Looks to me like the usual tilt of the panel has the frame covering zerk fitting.
Spinning that pillow block around could make it easier to access.

I drew it like that on SketchUp, but in reality, it is not very easy to do, especially using Superstrut.

I was using a ground trencher last week to put in my irrigation lines, it has two pillow bearings like the ones I am using here. However, they are spinning about 1200 RPMs. The E/W set of bearings will do 1/2 a revolution a day, and the N/S bearings will do 1/8 revolution a day. ? So greasing the bearings won't be much of an issue.

jasonhc73 · May 20, 2021

In the test phase before mounting the big panel.

It has one teeny tiny little 12v car battery charger PV, about 1/2 Sq ft. It doesn't have a spec sheet but it did 20V VOC yesterday.

Hedges · May 20, 2021

How much has it cost you so far?
How many dollars worth of PV panels can it hold?

(I know, I know, it's a science experiment. Do you realize how many billions of dollars have been spent on inertial confinement fusion to generate a reported 50 kJ of energy? Which is about what you get from a single box of ammo.)

jasonhc73 · May 20, 2021

I thought nothing was happening all day, until the very end. The time-lapse shows it moving all over the place.

Hedges · May 20, 2021

jasonhc73 said:
I thought nothing was happening all day, until the very end. The time-lapse shows it moving all over the place.

Clouds? To be expected. Surprising it has the optimum straight-up orientation.
If you start it parked, will it move to that?

Oh the sun came out for a moment!

jasonhc73 · May 20, 2021

Hedges said:
Clouds? To be expected. Surprising it has the optimum straight-up orientation.
If you start it parked, will it move to that?

Oh the sun came out for a moment!

If it doesn't detect sun, it goes to the "defused" or "parked" position. But as soon as it detects sun, it aims as close to perpendicular as it can get to it.

In my location, the sun comes up at 110° (2:00 looking down) the at high/sol noon it is about 78° and sets around 290°(10:00 looking down)

I have some adjusting to do on the N/S so it can go more N. In the video I tried adjusting at the very end the E/W actuator a little, but I kind of ran of interest. It's near the mechanical limits of what I can do without changing some struts around. Last winter I saw a double pivot that I have been considering also. I made a Legos one and it works fine, but scaling it up to this is a little extreme for "just goofing around" experimenting.