Azimuth Rotating Mechanism for Solar PV Array

go from east to west

http://arraytechinc.com/wp-content/uploads/DuraTrack-HZLA1.pdfThe DuraTrack HZLA is rated to 90mph wind loading and has been proven to stand up to high winds and harsh environments while providing maximum efficiency.

Good catch, I was thinking East- West but typed North- South, my bad

I'm thinking just throw something out there as a starting point to build on...so

http://www.interpv.net/tech/tech_view.asp?idx=451&part_code=030140054

With a motor driving the worm gear, through the drivetrain of the one ton truck and the axle spindles connected to the support structure holding the PV array...the axle would turn very slowly geared down maybe 100 to 1.....

You might check out if there are any old TVRO satellite dishes in your area. They were designed for tracking and the bigger ones ( 10.5 ) should handle the weight and wind well. Also a lot of those systems were built with decent controllers.

East-west ~180 each day at ~constant rate; north-south ~47 each six months at ~sinusoidal rate (somewhat optional).

What do you mean by a "strong wind"? A low level F1 Fujita storm will likely turn this into a pretzel or missile? I know that I'm being a little absurd but I dislike the your desire to just add an amount of mass to keep a big sail on the ground in a storm.

I live out in the country and the array would be mounted out in a field, I guess a strong wind would be a gust front moving through from a thunder storm with 50 to 60 MPH winds, which I have recorded before, my 16 ft sat dish with 6 panels held up no problem. My neighbor has a fixed 12 panel single pole and has had no problems with that amount of wind. I guess if I used a truck rear end I could hook the brakes to a vaccum boost and pump and an anemometer hooked to a micro and when the wind exceeded a preset limit it would throw on the brakes

Thats not strong!! Ask anyone in the UK what a strong wind is (and its nothing to do with Baked Beans!!!) See here:-

strongest-winds-in-the-uk

But I would be surprised if where you are they are only that strong. Design for maybe double that? Or even check with your local weather buro.

You are right, build everything as strong and heavy as you can.....no harm in that.

Maybe, once its running, you could design in a system whereby it sets itself, side onto the wind above a certain wind speed, to reduce wind drag.....should be easy to design I feel.....

I would personally rather invest in more solar panels than moving them, if I wanted to sue a solar array. Here most people populate their roof and do not move the house at all......

Possibly arranged maybe in a half circle to catch the sun early. Completely unmoving, except maybe for a simple elevation adjustment for summer and winter time, simply done by hand, though a years average middle elevation, would probably do it fine too.....

That way I could achieve a strong, solid installation, with no need for truck axles and the needed controls/drive. I am a strong believer in K.I.S.S., as are many here.

We also have large "fields" of such arrays all over Germany, none are movable. Surely you have to ask yourself why? Years of experiment came to that conclusion.....

Alone the problems of picking up power from a moving array and keeping it weatherproof would put me off.

Also, modern systems cool the arrays with water, to not only improve their efficiency quite dramatically, but also supply warm water for other domestic purposes at a very cheap rate. Overall efficiency becomes quite stunning, I am told!!! Getting the warm water "off" a moving array, will complicate things heavily I feel, hoses will be an expensive problem.....

Furthermore, the amateurs among us building such things, make far too complicated systems for following the Sun.....Photo sensitive cells providing up to date information as to where the sun "is" are a waste of time and money, What many forget is that the Sun's position can be accurately calculated without even looking out the window!!! EVER!!

From any known position on earth and knowing the local time and the day, the Sun's position in the sky, both azimuth and elevation, is easily worked out (even for me!!).

A simply system, without any elevation correction, just needs to know the position of the hour hand on a 24 hour clock, or even the time from a 24 hour digital clock. Convert that to a horizontal angle in say degrees, and that is the azimuth angle where the sun will be relative to that position! Even if its on the other side of the planet (night?). Simple....

There are also sunrise and sunset tables, that can be obtained online, tailored to suite exactly where you live, and knowing that between sunset and sunrise, there is no need to move the array, except (maybe) to move it back to its "start position" for the next day!!!

Two angles only, one for where the sun will arise and one for where it will set for each and every day of the year....a really simple data set!! Even if you are a day or two out, the efficiency loss will be difficult to assess!!

A single set of Sunrise sunset tables for a whole "leap" year, letting the control software ignore the 29th Feb for the years that its not valid, will be far more than just "good enough" for accuracy....or forget even that and correct it manually when needed!!! Just look and see occasionally, "is the date wrong or not?"

I would guess that I am not the first to realise that "time" is enough really for azimuth positioning infos, but I have not seen anyone here on CR4 mention it before.....if anyone knows different, please point it out. Thanks in advance.....

I hope this provides some "food for thought" so to say.....

If anyone has further questions, just ask them, I will do my best to help you further......

I agree that a microprocessor and memory chip loaded with the sun's movement data is the easiest way to go if you have a servo motor and encoder to keep track of where the array is pointing. In the low tech system I designed I initially wanted to use a spring wound clock to pulse a solenoid every half hour but that option lacked any feedback mechanism and would not reliably synchronize with the sun. There are about thirty installations of my design mostly in Mali and Chad but a few in Kenya and three in Afghanistan. All have been scratch built by the local villagers with a little help from overseas aid workers. The material for the cylinders varies from steel or plastic pipe and in one case ceramic sewer pipes. The light sensors I used are simple two pin cadmium sulfide LDR photo resistors (light dependent resistor) with a Wheatstone bridge circuit to compensate for the wide tolerance on the outputs of devices that cost £0.06 ($0.1) each

Everyone goes the way he must.

But may I suggest a far simpler mechanism than a servo, with all its extra work and control?

Stepper motors are the way to go for such a job, no need to have feedback at all. They are stepped/driven, from a known position by "dead reckoning". They are used in many objects nowadays, some requiring high accuracy like scanners, photocopiers and printers for example....

There are plenty of cheap chips around to drive them, many under $10.

Depending upon the size/power of the motor, you will need to gear it down. In fact a relatively smaller motor, with a good reduction gearing, could be really good. You don't need speed.

Bike chain and gears are a simple method, that can be found for small money in any, even third world, countries. Gearing down, will also help to smooth out the steps dramatically.

If you want to be "up market", then a screw drive will also lock the unit in place when not being driven, though with a stepper motor, over gears, it will also peg it into a solid position!

Some users simply drive the stepper motor into an end stop position for a specified time, so as to "know" the starting position. This does not damage them.

Others use micro switches to "find" a specific position. Some use both, the micro switch as a "warning" that the end stop will soon be found for example. Hall effect switches and magnets can usually be made weather proof quite easily.

I am sure that there are more modern ways to do it, and someone here will know better than I.

If I was doing it, I would use a "PICAXE" PIC, using BASIC commands to drive the stepper motor and keep track of positioning etc..

They are SOOOO flexible and the programming is really simple.....it can even be programmed to drive a stepper motor via some heavy duty transistors and some simple circuitry directly if required, user's choice.

You can find more here:-

rev-ed.co.uk

All software, books and many helpful examples are completely FREE to download and use for anyone. The chips are all very low priced as they were originally designed for school children, which still use them, but the chips have progressed dramatically in terms of speed, memory size, built in interfaces and other possibilities.....

You will find a lot of helpful people online here:-

picaxeforum.co.uk

But please do your own thing as you see fit, I only wanted to make you aware of how it could be made both simpler, cheaper and more robust..........

Nice sturdy design...

Here's a passive setup...

Here's a guy that built his own tracker....

https://www.youtube.com/watch?v=AHYCLvaeAbk

Cheap 2 axis solar tracker kit....

This looks like it would handle a small array, but not a large one that I want to install.

Here are couple photos of my dual tracker

That looks like a Paraclipse Dish,which is Very Strong,modeled after the large NASA dishes of the time.

It appears to be about a 12 foot dish.

I do not recognize part of the positioning system,however.

It looks like a manual adjustment for elevation.

This will be required due to gradual seasonal changes in the relative sun position.

You could modify the mount by replacing the fixed turn buckle,or manual adjustment,

with an actuator,similar to the one already in use,and using a modified dish controller

to give the elevation movement,and reset the programmable limits to give a full

north/south movement.

A simple relay,switching from AZ/EL would work for incremental changes,or a photo

diode array sensor, with collimator for continuous tracking to keep everything

centered.

A 24 hour timer will need sidereal time clock to keep proper position.

You realize of course,that you could cover the inside of the dish with mirrors and put

a smaller higher rated solar collector at the appropriate focal point?(Not the calculated prime focus,this would be too hot.)

This would make tracking less critical because light from any angle would be focused to the center.

My original system as shown in the photos is a knock-off 16 foot Paraclipse Dish with horizon to horizon mount. For the counter balance I had to add a couple of hundred pounds of lead and beef up the support for the extra weight of panels. It was modified with a center pipe acting as a shaft to rotate the entire dish. A large sprocket is located at the bottom connected to a gear box, and operated by a standard 36 VDC satellite drive motor. I was part of the design team that made tracking satellite system to track The Russian Molniya satellite that would pass over the US. This was our proto-type and latter we made a 30 foot system for the US Army PSYOP's , back in the 80's. I ended up with the 16 footer and used it for C band reception, in fact when Desert Storm was happening, we could watch the live feeds from CNN from an Intelstat Bird. The dish was almost touching the ground to pick up siginals. Then C band programming died so here came the PV array

Try searching for used gear motors.

Here is one to get you started:

There are many offerings on the web.

This may be a cheaper and easier route than trying to reinvent the wheel from scratch.

http://www.surplusindustrialsupply.com/products/Gearmotors-Speed-Reducers-GEA/Used-Gearmotors-Speed-Reduce-Flexaline-100-1-Ratio-Speed-Reducer-GEA1362.html

Excellent link!

But it looks like a hand held remote operated unit.

I do not see position feedback for automatic control,however it could be coupled

with the power wires.

Looks like someone else found a use for the old C Band sat dish actuators.

Actually, the whole system could be much simpler if you mounted the axis vertical and allowed the array to turn 360 degrees in azimuth over a 24 hour period. That saves the complication of a reversing mehanism. During nighttime the array won't be converting light to electricity anyway, so you might as well let it continue turning through North and let it start catching light at dawn during the Midwest summers. If it only turns 180 you'll miss an hour or two of morning and evening light.

If the array is well balanced, a small motor is all you'd need. Then you could use a gear, chain, or belt drive to covert the RPMs of the motor into a 24 hour clock-drive for the platform that the array is mounted on.

Yeah but you keep it turning 36 you'll need slip rings for the electrical connections, and that spells problems.

as for wind, a building permit requires design match your areas wind maximum, right? and the additional cost, including maintenance will not IMHO be offset by the additional electricity. Running it at night is a waste of energy, and need a battery system if off-grid.

don't mind my asking this, but is it worth to spend the extra energy to turn the axis due north, considering the entire unit is probably weighing close to 200 pounds or more. will two hours in the morning generate enough energy to compensate the energy spent in axis rotation?

The idea is good, but picking up the power may be a lot more complicated..... and also be unreliable....

Then you have to provide commutator rings on it also.

You are fully correct of course. A problem that few others consider before posting it would seem.....

Also, if its an efficient modern system, with built in panel cooling, you need to get the fluid on and off the array too.....that is if you want to actually use the heat for something useful....

That would all be WAY too complicated for me personally.

Seems easier to simply build a Newman engine.

https://www.youtube.com/watch?v=RMYo1QlvK5g

I don't know why the U.S. Patent Office failed to issue him his patent for so long, this has been going on since the early 1970's.

Don't believe the self promotion. Newman Engine analyzed.

More's the pity. I was thinking of hooking up a plethora of "crystal" batteries to one of these, and powering God knows what. Shucky dern.

The video is blocked here, probably there is music that the artist does not want to be played for free.

What are the pluses of such a motor?

Thanks

Newman has umpteen 9 V batteries connected in series to his motor, and even though the witnesses could hardly turn the motor with a pull cord on the shaft, he was spinning about 220 rpm (depends on voltage of the batteries connected in series, and # turns wire) using a weak current source operating on a commutator between and upper and lower coil, and large magnets spinning on the shaft with an air gap, I believe.

This was driving a fairly large diaphragm pump known as a mud pump, I think the suction and discharge hoses were 3" sized. Not a small task at all. The mass of the motor is around 3.5Mg. (megagrams), or about 3500 Kg, if I correctly recall.

The U.S. Patent Office would never grant this fellow his patent, as he may have claimed over unity, and that he was converting mass to energy using magnetics only.

It apparently would run indefinitely on even old used up 9V transistor batteries (or any other grouping of batteries delivering the same potential at the commutator). Apparently, the current draw was positive at times, and negative (back into the source at others), and could apparently even charge up old batteries slowly, that they did not explode.

Maybe it is all snake oil, I don't know. Just sayin' that if it does what he claims, and runs indefinitely off the same batteries, should we not look into it?

First, decide if you need a single or dual axis tracker. This will help:

http://www.strathconasolar.com/single-axis-vs-dual-axis-solar-tracking/

My personal preference is a Tilted single axis solar tracker. This paper will get you going:

http://www.enggjournals.com/ijet/docs/IJET13-05-02-213.pdf

Next, decide on your programming platform. I am partial to Arduino.

Then, decide if you want to program array based on lat / lon, time of day, and day of year or use a sensor to keep the array pointed at the sun. Either will work.

My gut feel is your rig is too heavy for 12v.

My present system is using 36 VDC motors with a Heliotrack dual axis controller which uses a sensor. Since it has been on line since 10-1-13 I have generated 380 KW with Six 190 watt Enphase grid tied micro inverters and six 220 watt panels. I have turned of the elevation motor and had come to the same conculusion that it wasn't worth the extra mechanics for the new install, but since I already have it on my present one it has worked flawless since it was installed. that why I was looking for a cheap way to rotate the array and a truck rear end and a little concrete and a little welding looked like the way to go.

I wish I had looked up the Heliotrack system before I replied.

It has essentially the same thing I suggested in my reply.

It looks like with the appropriate mechanical advantage,the existing actuator could be

used on your larger system.

Some of the C Band actuators had a capacity of 1000 lbs.

If you have a worm gear in the drive somewhere, that will keep it in one place.....therefore no need for brakes.

There is a gear ratio for worm drives, that once above this amount, they cannot be "back driven".....I have simply forgotten the minimum value, sorry, but you should be able to find it out with sa Google if the idea interests you further.

Furthermore, quite a small motor will then do the job, maybe even a wiper motor with gearbox from say a car or truck, cheap and easy to source....cheap to find a spare as well and usually reasonably weather proof.....

I wish you well, show us some pics one day.....

Here is one I designed earlier for distribution via NGOs to rural communities in the third world. I grant an open licence to use this design free provided it is used on a non profit basis.

Low cost, low tech, solar panel tracker for local manufacture.

Objective

Fitting a tracker to a solar panel can boost it's performance by 50%. Conventional tracker technology is expensive to buy and maintain. This offers a simple, reliable alternative.

Mechanics

Fit each solar panel onto a three point mounting L, M & N.

Mount L is fitted at the center of one short edge and comprises a ball and socket joint on top of a short pole stuck into the ground. This will be the lower edge (to the south in northern hemisphere and the north in southern hemisphere)

Mounts M & N are fitted at the corners of the other short edge and are attached by rods to floats (plastic ball cocks) that each sit within a vertical cylinder (plastic pipe sealed at the base and open at the top). Displace liquid (old engine oil) with a pump (oil car engine oil pump fitted with a windscreen wiper motor) from one cylinder to the other and the panel tilts sideways.

Electrics

A single printed circuit board (PCB), can be purchased ready made, assembly from a kit or scratch built locally. Mount onto the front of the solar panel with the tube perpendicular to the panel surface.

PCB incorporates; a voltage regulator (fed from the DC storage capacity), three light sensors (A,B&C), two differential amplifiers (E&F), a latching relay (L) and a reversing relay (R). To complete the circuit you need a couple of decoupling capacitors and some resistors but the total component cost should be under £2 and add £2 more for the PCB and assembly. Light sensor A is always shielded from direct sunlight by a plastic plate fixed 20mm in front of it. Its output is connected to one input of amplifier E. Light sensor B sits at the base of a plastic tube so that full light only falls onto it when the tube is directly aligned with the sun ± 3^o of arc. Its output is connected to the other input of amplifier E and one input of amplifier F. Light sensor C always remains uncovered in full sun. Its output is connected to the other input of amplifier F. The circuit is connected so that if the same amount of light is falling on both sensors connected to the input of an amplifier, the amplifier output of off. If the inputs are unbalanced the amplifier output is on. Note that it does not matter if there is full sun or partial cloud as the background light level is same for both sensors feeding a common amplifier. There are also two limit switches set to the detect limit of the range in the East and West directions.

Operation

Start in the morning with the panel tilted East and the east limit switch tripped. Nothing happens until the sun rises high enough to shine directly down the tube. At that point the outputs of the always shaded light sensor A and the light sensor at the base of the tube B will be unbalanced. The differential amplifier E will trigger and set a latching 'run' relay L. Once the latch is set amplifier E has no further function but as sensor B intermittently is in full sun it tries to re-latch the already latched relay. At the same time the tube sensor B and the unshaded sensor C will both have the same amount of incident light and the output of balancing amplifier F will be off. As the sun moves round sensor B again falls into shade. Amplifier F triggers, starts the pump and realigns the panel and sensor B tube directly with the sun, then switches off again until further movement of the sun again throws sensor B into shade. This incremental movement is repeated every few minutes all day as the panel tilts from east to west. When the panel trips the west limit switch the run relay L is de-latched and the reversing relay R is latched. R opens a 'reverse direction' valve and runs the pump continuously until the panel trips the east limit switch which unlatches R. The pump stops and the 'reverse direction' valve closes. The system is now set up for the next day's operation.

Every two weeks some oil is added to or removed from both cylinders. This adjusts the angle of the panel to compensate for the different track of the sun as the year progresses.

Provided the panels are mounted on level ground, several panels can be slaved from the master panel by connecting the bases of all the east cylinders and separately the bases of all the west cylinders. The connected cylinders act as a multi leg U tube, so if the level in the master leg changes, the level in all the slave legs change with it. If more than three or four panels are to be connected in this way the size of the pump must be increased to maintain an adequate response time.

These all seem very technical to me, but then I know little about following the sun.

Having 16 fixed panels on the roof (SE) I do notice the loss in power as the sun travels over and often wish I had installed them pivoted in the middle (fulcrum like) at each end of the array to capture the sunlight as the sun (and day) passed by.

That is, I think the gain from a single pivot (fulcrum) at each end used to tip the array say 6 hours each way (6 to the SE in the morning and 6 hours to the NW in the evening) would be relatively simple to install and while maybe not be catching every watt, would more than justify the expense?

What puts me off doing this is the fear of having some 240 square foot of "sail" on the roof ! The tiles already shake in a severe wind, to a sail I think I would be definitely saying goodbye! (likely to the whole roof)

Because I think this would be a very economical way to substantially improve the output does any member have a good suggestion on how to anchor all this down?

Many thanks for your help.

jt.

John got off the elevator on the 20th floor and nervously knocked on his blind date's door.

She opened it and was as beautiful and charming as everyone had said. "I'll be ready in a few minutes," she said. "Why don't you play with Spot, my dog, while you're waiting?" He does wonderful tricks. He rolls over, shakes hands, sits up and if you make a hoop with your arms, he'll jump through."

The dog followed John onto the balcony and started rolling over. John made a hoop with his arms and Spot jumped through - right over the balcony railing!

Just then John's date walked out. "Isn't Spot the cutest, happiest dog you've ever seen?" "To tell the the truth, " he replied, "Spot seemed a little depressed to me!"

Why not just use a weight on a clock-work, and re-wind it every day for some cardio-exercise? If you have to take fluid on/off the platform to provide area heating/cooling, then this simple harmonic motion where man is required is better. Circular motion will be a problem due to wiping seals to transfer liquid that will eventually leak.

Would it not be even better to keep the panels motionless, but change the optical array to direct the sunlight to the panels? I may be speaking naively here, but it would seem so, especially where cooling of the panels is required.

If the panels were located at the focus of a Rowland circle of beam deflectors, would that help (or just get way expensive?)