My Portable Solar Tracker (single-axis)

Discussion in 'Power - Site Power/Batteries/Generators/Solar' started by Dubbya, Apr 27, 2015.

  1. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    The unit ships from China so as you might imagine, documentation and specs are few and far between. In practical application, I ran it for an afternoon and the battery monitor didn't budge, so it appears to be pretty efficient. I'm hoping to get a solid weekend of service out of one 9.6v rechargeable battery. Going forward, I may add a couple of small solar panels just to maintain the charge in the drill battery but we'll see how this works first.

    This controller doesn't have "parking" and it's an "always on" scenario, this is another reason why I felt it necessary to add an on/off switch. Though it constantly monitors the sun's position, it doesn't use much power at all to do that, only when turning on the motor to reposition the deck. Even then, the duty cycle is extremely short, as it only moves the deck two or three degrees at a time.

    If power consumption is an issue, the TinyTracker HD has parking built in and is available in either single or dual axis. It runs a few post checks, positions the tracker then checks the sun's position every 10 minutes. More efficient but double the price.
     
  2. rabird

    rabird Howdy!

    6,321
    159
    Mar 3, 2006
    TX
    Very nice.

    looks easier than rotating about the tilt, new to me 'Vertical single axis tracker'

    Solar comparison

    [​IMG]
    embedded for educational purpose!
     
  3. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    Neat idea! I guess the idea there is to tilt the panel diagonally to follow the sun. I'm not sure I understand the concept correctly but it seems that the apparatus would be mounted to a universal joint of sorts so that the bottom corner goes up and toward the West while the top corner goes down and toward the East. That's pretty clever!
     
  4. rabird

    rabird Howdy!

    6,321
    159
    Mar 3, 2006
    TX
    Your's would be C, A is what I've understood as single axis tracker, B is the same as A with no tilt (good locally for summer).
    With my portable, I rotate like you but increase the angle in the morning and eve, (if I'm gonna be around) or just set it for best noon time, all manually.

    [​IMG]
    embedded - educational
     
  5. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    Right, "Single Axis tracking on a Vertical axis". Lookie here, I learned sump'm! Thanks, Bupkis! [:D]
     
  6. MickeySp

    MickeySp Member

    107
    0
    Apr 2, 2015
    Dubbya, any chance you could update your parts list with prices? I'm curious what the tally is...
     
  7. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    Are you curious as to how much I've spent on this so far or are you hoping to find out how much it would actually cost someone else to build something similar?

    Guesstimating, I'd say $150-$200 should be enough to build something similar to this but again, you can certainly do this cheaper.

    Everyone is going to have at least a few of the items on the list and they're going to find their own solutions to tackle the same issues. Hopefully they build to suit their own purposes using whatever tools and materials they have on hand. Some solutions will be less expensive, some will cost more. Some will be more complex and some more creative and simple. I think it best just to share what I've done and let others price out the items they might require for their project locally.
     
  8. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    I dissassembled the contraption and ended up spraying the base and top deck with three coats of Duplicolor Truck Bed liner and a finishing coat of Bed Armor from a leftover can I had on hand. I left things to cure in the sun for a couple of days before I reassembled the top deck.

    Made a little more progress this past weekend and managed to get the 12v 2 pole rocker switch installed. This cost me about $5 to do and it's not marked "on/off" but it does the trick. Pressing the rocker switch down turns off power from the battery and up turns the unit on.

    [​IMG]

    I also got some 2" non-marking "feet" installed. $4-$5 for a pack of four at HomeDepot. I used 1/2" screws and a dollop of Marine Goop to effectively glue them on as it seemed the screws are a little too small to hold them on reliably.

    [​IMG]

    This is the underside of the top deck, fully assembled with the wheels, chain tensioner, tensioner spring, and drive sprocket. I've also attached the lanyard ends on the pull pins for the drive post underneath the deck (on either side of the drive sprocket) so they're out of the way.

    [​IMG]

    And here's the top side of the assembled top deck showing the three mounting points for the solar panel.

    [​IMG]

    Just waiting on the 54T chain ring to arrive so I can join the two halves. Meanwhile, the bedliner paint gets to cure some more.
     
  9. MickeySp

    MickeySp Member

    107
    0
    Apr 2, 2015
    I'm curious what it would take to replicate. Which you answered, thanks!
     
  10. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    I've made some progress in the past two weeks and the project is nearing completion. I've attached the 54T chain ring, new 3/32" x 1/2" chain and finished reassembling the unit.

    Here the solar panel is attached and laid flat atop the rotating deck and base, ready for storage and transport.

    [​IMG]

    From the front at ground level, you see the base, drive sprocket (chain ring) with the chain in place, the top rotating deck, the drive post in the stowage brackets and the frame of the solar panel.

    [​IMG]

    Again at ground level, you see the base, drive chain, chain ring, tensioner, drive sprocket, rotating top deck, and the underside of the solar panel. At left between the top deck and underside of the solar panel is the stowed drive post. Atop the drive base is a 1/2" foam ring which has been glued to the underside of the solar panel with Marine Goop.

    [​IMG]

    Here, the solar panel is partially lifted while remaining hinged at the bottom. This exposes the stowed drive post and retaining brackets.

    [​IMG]

    Here, the solar panel is tilted all the way forward so that the bottom of the frame rests on the front edge of the rotating top deck and the drive post has been removed from the stowage brackets, ready for insertion into the drive base.

    [​IMG]

    With the solar panel tilted forward, the drive post is inserted into the drive base and secured on either side using quick pins.

    [​IMG]

    Another shot of the inserted drive post and the solar panel tilted fully forward.

    [​IMG]

    With the drive post secured in place, the solar panel is tilted back and the top eye bolt is secured using a quick pin.

    [​IMG]

    With the solar panel locked in place, all that remains is to place it in the sun, plug in the power cables and turn it on.

    [​IMG]

    When powered up, the controller automatically positions the solar panel so that it is centered on the sun. It moves a few degrees every 6-10 minutes even when overcast. Surprising, but the LED's are extremely sensitive.

    [​IMG]

    I've tested the unit under clear skies in full sun and in hazy/overcast conditions and it performs exceptionally well for approximately 2 hours before it burns the PCB directly under the heatsinks. After some reading, I've discovered that a 4 amp rating (by the controller's manufacturer) is either inaccurate or woefully inadequate. A second unit was purchased, this one said to be a 10 amp unit but it was exactly the same as the 4 amp, identical in every respect. Despite my inquiries and the seller's assurances that it wouldn't be a problem, guess what? Same result. 2 hours and a dome filled with white smoke. GRRRRRR! I tested the drill motor and it was using between 3.2A without a load and 3.9 amps under a moderate load. Of course, if the unit is binding or the motor is partially restricted, the amperage soars and can jump up to over 13A at stall. That's a problem!

    No problem, I'll get a refund (again) and I've since purchased the TinyTrackerHD single axis controller ($38.95 US) which is rated for 8 amps and 200 Watts. This one is made for larger motors and shouldn't have any problems meeting the amperage requirements of the 9.6v drill. Note that I purchased mine on eBay since international shipping fees were $11 US for USPS vs. $33 US for FedEx to Canada.

    Additionally, the TinyTracker HD turns off for 3 minutes between cycles in order to save power and so shouldn't overheat like the cheaper unit did. Bah, what can I say? I tried to save a few bucks. Meh, lesson learned! Next time spend the dough and get the one you know will work. [;)]

    http://www.homecsp.com/

    TinyTracker HD Features:

    At less than 2" square the TinyTracker HD lives up to its name. Integrated sensor, controller, and motor driver eliminates the need for any other electronic components (besides the power supply). The Atmel ATTiny microprocessor applies advanced logic and data filtering techniques to the LED light sensors for controlling any sort of East-West tracking system.

    A built-in power regulator provides 5v logic supply voltage from the main motor power suppy, and accepts a wide range of input voltage, up to 30vdc max! Making it perfect for connecting to most 12v or 24v solar power systems. Ultra low power use makes the TinyTracker efficient too. Nominal tracking frequency is 3 minutes. Automatic nighttime detection will reset the unit to an Eastern position.

    We recommend mounting the unit inside clear glass or plastic jar/dome to provide full 180 degree visibility for the LED sensors. Mount the jar sideways to get a clear view through the side. This is easy if you attach a water-tight PVC conduit fitting to the lid, and attach the lid your array. Then the jar just screws down over it. Take care to shade unnecessarily exposed areas with silver paint or foil to avoid creating a little solar oven that will overheat the electronics.

    Specifications:
    Voltage: 30vdc max, 12-27 recommended
    Tracking Freq: 3 minutes
    Max Motor Run: 5 minutes
    Sampe Period: 0.5 seconds
    Maximum Current: 8A @ 12V, 5A @ 24V
    Eastern Return: 10 minute delay

    Still Remaining:

    [list type=decimal]
    [*]Fabricate a "parking brake" to lock the top deck and prevent it from rotating when the solar panel is folded down.
    [*]Fabricate a method of securing/supporting the top eye screw (attached to the solar panel frame) so that the solar panel can't accidentally flop down when in the stowed position
    [*]Attach safety cable(s) or folding bracket(s) from the top deck to the top edge of the solar panel frame to reduce stress on the frame when the panel is tilted all the way forward during assembly/removal and stowage of the drive post.
    [*]Install TinyTracker HD
    [*]Figure out a convenient, efficient, durable and cost effective wiring/plug system for the solar panel cabling.
    [/list]
     
  11. ~erik~

    ~erik~ Active Member

    287
    44
    Jan 19, 2015
    North Georgia
    Thanks for the update. Great job. Frustrating about the first two trackers drawing too much current and filling with smoke. Now I know what to buy and not repeat your experience. [:D]

    I guess the only thing that bugs me about your design is the use of the original drill's rechargeable battery. I think it would be more elegant to use the power of the camper battery that the solar system charges, rather than to have to unplug the drill battery and charge it when it stops turning the solar tracker.

    Thanks again for the update!
     
  12. HobieNick

    HobieNick Member

    181
    2
    Mar 16, 2015
    +1 on using the 12 VDC from the camper's battery or the solar panel.
     
  13. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    Yeah, I thought about using the pup's battery to run the motor but at 12v, the motor turns even faster and the solar panels I tried we're not big enough to run the motor by themselves. It needs a battery of some kind and most units I've seen seem to support that method. There are tons of solar tracker that use 12v low amperage motors and smaller solar panels to run them but I wanted to use what I had on hand.
     
  14. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    I managed to fabricate and install a locking mechanism this weekend. I made a 90 degree bend in a 1/2" steel rod, drilled a hole and inserted a spring pin, then drilled another hole for a cotter pin.

    I modified a 3/8" fender washer by drilling out the center hole to a 1/2" then slotted it.

    I drilled a 1/2" hole through the top deck and base then bolted another fender washer to the base to act as a receiver for the end of the locking pin.

    I slotted the 1/2" hole in the top deck so that the spring pin in the locking lever could slide through to lock/unlock the top deck.

    [​IMG]

    Here, the locking lever is in place. On the underside of the top deck, is another fender washer on top of a compression spring and retaining washer. These are secured in place with a cotter pin.

    [​IMG]

    This is the locking lever in the unlocked position so that the top deck is free to rotate. Pull up and lock/unlock with a 1/4 turn of the lever.

    [​IMG]
     
  15. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    I've been working on this off and on over the past month, so here's where things are at:

    Still Remaining:

    [list type=decimal]
    [*]Fabricate a "parking brake" to lock the top deck and prevent it from rotating when the solar panel is folded down. - The rotational brake is installed!
    [*]Fabricate a method of securing/supporting the top eye screw (attached to the solar panel frame) so that the solar panel can't accidentally flop down when in the stowed position - Scrapped this idea. A trip to the thrift store yeilded a used camera back and a backpack with 1-1/2" webbing and clips that were attached to the panel frame and tracker base to secure the panel in the stowed position.
    [*]Attach safety cable(s) or folding bracket(s) from the top deck to the top edge of the solar panel frame to reduce stress on the frame when the panel is tilted all the way forward during assembly/removal and stowage of the drive post.
    [*]Install TinyTracker HD
    [*]Figure out a convenient, efficient, durable and cost effective wiring/plug system for the solar panel cabling.
    [/list]

    A problem that I've had with the drill motor is that it spins too fast and that it requires more power (amperage) than the controllers can handle. While they didn't seem to struggle with 8-10 amps while the motor was running, the start/stop and stall amperage was well above that, sometimes as high as 30 amps at stall.

    I took a two-pronged approach to solving these problems.

    Firstly, I ordered a relay booster module that gets placed between the motor and controller so that the controller doesn't have to deal with 30 amps.
    $29 US from SBE Solar Technologies.
    [​IMG]
    http://www.sbesolartech.com/relaymodules.html

    I separated the relays from the board, stuffed everything inside the drive post, connected the controller and motor and tried it out. Just as promised, the relay allowed the controller to do its thing while remaining nice and cool. Success! I ran the tracker for a full day in direct sun without any heat-related issues.

    The problem I now had is that the relay module is an on/off, full-speed solution that doesn't come with an easy way to reduce the motor speed. I didn't want to mess around with electronics, so the next option was a mechanical one. Since the drill's gearbox contains two sets of planetary gears that reduce the motor RPMs to 300 RPM at the chuck. Another visit to the thrift store yielded a $5 Black & Decker 9.6V cordless drill from which I harvested the gearbox. After a few tests, I decided to stack another set of planetary gears on top of the old gearbox.

    Using a side grinder, I cut the top 7/16" section off of the donor drill's gear ring, then used crazy glue to bond the two gear ring sections together, clamped the assembly in a bench vise and left it overnight.

    [​IMG]
    [​IMG]
    [​IMG]

    I removed the clutch's ball bearings (to bypass the clutch) so that the drive would be solid, then coated the inside of the original gearbox housing with Marine Goop. I inserted the gear rings and planetary gears and let this section cure for the weekend.

    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]

    I used an oscillating multi-tool to cut the top 1/2" off of the donor drill's plastic gearbox then sanded the bottom edge of the ring to make it smooth and flat and drilled out the screw holes so that the two sections could be screwed together.

    [​IMG]

    As I did with the lower section, I coated the new gear ring section with Marine Goop then slipped on the top plastic ring and turned in some screws so that the two would be perfectly aligned when the motor section was mounted.

    [​IMG]

    I let the assembly cure for a few days while we were camping then came back to mount the motor using some 1-1/4" screws. Here's the modified drive assembly ready for installation.

    [​IMG]

    After reassembly and testing it appears that I'm going to need to tweak the TinyTracker software, not only to conserve power and smooth out the POST (Power On Self Test) but to further reduce the motor speed. As it stands, it seems that the controller tests the sun's position every 3 minutes, and every 500 milliseconds (1/2 second) while the motor is repositioning the deck. I've tried repositioning the LED sensors but I'm thinking that testing every 100 or 200 milliseconds (while in motion) should allow the controller to stop the motor so that the deck is positioned a little more accurately in relation to the sun.

    Since the motor no longer uses MOSFET transistors to immediately stop the motor when the power is cut off, the POST needs a few milliseconds of wait time between the initial left/right rotation test. Furthermore, the TinyTracker allows up to a 25% reduction in motor speed using built-in PWM (Pulse Width Modulation). This should give it the time it needs to "think".
     
  16. lifespeed

    lifespeed New Member

    693
    4
    Aug 26, 2014
    You can't use PWM with relays.
     
  17. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    Hmmm... after a little light reading and some thought, I see why not.

    Well, I guess that simplifies things and I can still tweak the software to address the other issues and forget about using PWM. Worst case scenario, I just add another set of planetary gears, go from three sets to four and reduce the chuck RPMs from 80 RPM to around 16 RPM (guessing). That'd probably provide enough additional torque to be able to raise/lower the roof on the pup or loosen/tighten lug nuts! [LOL]
     
  18. lifespeed

    lifespeed New Member

    693
    4
    Aug 26, 2014
    I doubt this is very helpful at this late stage of your project, but . . .

    The most direct way to retain the design features of the tracker controller (like PWM) is to use a lower-powered gear reduction motor that won't overload the circuit. Obviously high power is not required for your application.

    Or, you may well find with the significantly reduced electrical load via added gear reduction to the drill you can go back to using the MOSFETs directly as well as PWM (assuming the drill is a simple brushed motor without its own controller).

    Cool project, sounds like you're close.
     
  19. Dubbya

    Dubbya Wherever you go, there you are...

    6,125
    42
    Aug 2, 2011
    Steinbach, MB
    Since the deck was still rotating too quickly, I added an additional set of planetary gears so that there are now four sets, each with a 5:1 ratio, two from the original gearbox and two additional sets from a recycled drill.

    I cut another plastic housing, cut another gear ring, crazy glued it to the original gear ring (same as I did with the third set above), filled the cavity between the housing and gear ring(s) with Marine Goop, screwed the housings together and let the adhesive cure.

    I put the motor back on, reassembled the apparatus then plugged in the 9.6v battery and gave it a go. With four sets of gears, the RPM was reduced from 10,000 at the motor to 16 RPM at the drive sprocket. Perfect.

    The motor works longer now but because there's so little torque on it, it uses a lot less power. While testing, I've had no trouble getting two full days of service out of one charge. That's acceptable and it wasn't completely dead after two days but that's all the time I had to test it and I'm hoping to get at least three days out of one charge in the field. We'll see how it goes.

    I finally got around to taking a pic of the straps that hold the panel in the storage position.

    [​IMG]

    And some pics showing how the 10 Gauge cables all connect.

    [​IMG]

    [​IMG]

    [​IMG]

    And here's another short video showing how the apparatus works now:
    https://youtu.be/dc39H6hxmU4

    And finally, I'm calling it done.
    [​IMG]
     
  20. swordfish

    swordfish Well-Known Member

    2,203
    30
    Sep 27, 2010
    That is too cool! Thanks for showing.
     

Share This Page