1
\$\begingroup\$

enter image description here

Is it possible to run a solar tracker with a 10 W panel that connects to a 5 V buck converter and two MG946R servo motors without a battery and under a clear and sunny sky?

If not, can you suggest solar panel specs to run it without battery? I am doing a project to run it stand-alone, without battery.

I am just wondering if the panel is able to sufficiently power the servos for a few minutes; I don't need it to run the whole day. The project demo will showcase just a few minutes (under sunny and clear sun of course). If the weather is bad, I will show a recorded video during the test.

What if I use a 6 V NiMH battery pack between the solar panel and the load, or a 6 V, 4.5 Ah SLA battery?

\$\endgroup\$
17
  • \$\begingroup\$ What will the solar panel be powering? \$\endgroup\$
    – Bruce Abbott
    Commented Jul 22, 2021 at 2:12
  • \$\begingroup\$ With some large caps, and the panel in bright sunlight it can probably run 1 of these servos. But it would be better to use a simple gear motor instead, since that will behave more predictably when the available power/voltage is low. \$\endgroup\$
    – Drew
    Commented Jul 22, 2021 at 2:13
  • \$\begingroup\$ 1 Amp start current. Unlikely to produce useful power or work. 110 mA idle current. So NG. \$\endgroup\$
    – D.A.S.
    Commented Jul 22, 2021 at 2:17
  • 1
    \$\begingroup\$ Why are you do a project with no specs expected to fail badly? \$\endgroup\$
    – D.A.S.
    Commented Jul 22, 2021 at 4:02
  • 1
    \$\begingroup\$ "I am just wondering if the panel is able to sufficiently power the servos for a few minutes. I don't need it to run the whole day." Doesn't matter. Since you have no energy storage (like a battery or very large capacitor), only short-term requirements matter. And 5 volts times 1.2 amps (stall) is 6 watts. 2 times 6 watts is 12. Add another 10% for losses in the buck converter (if you're lucky), and you need a minimum of 13.2 watts. Of course, if you can guarantee that the servo will never stall, that's not an issue. Can you guarantee that? \$\endgroup\$
    – WhatRoughBeast
    Commented Aug 23, 2021 at 2:06

3 Answers 3

Reset to default
1
\$\begingroup\$

If you were to charge a small battery from the panel then you could determine when there’s enough stored energy to make an adjustment; move either the x or y servo but perhaps not both at once. When the panel is reasonably well aligned you should be ok to make small adjustments (a few degrees) quite frequently. If the axis of rotation is close to the middle of the panel then wind should have little effect provided that the panel is in free airflow.

\$\endgroup\$
3
  • \$\begingroup\$ yes im getting a 6v sla battery to power up the servos and other components. Sadly, i was aiming to do one without a battery but it seems to be much of a hassle than having one \$\endgroup\$
    – curiouser
    Commented Jul 23, 2021 at 1:47
  • \$\begingroup\$ Possibly you could make do with a large capacitor, if you make small movements and let the capacitor recharge before moving again, but there’s no getting away from the fact that the panel won’t generate enough to run the set is continuously, especially if the panel is off-axis. A battery is probably the simplest solution. \$\endgroup\$
    – Frog
    Commented Jul 23, 2021 at 5:55
  • \$\begingroup\$ thanks frog. yeah it seems that way. i am going to use a 12V battery which will be connected to a 5V buck to power the components \$\endgroup\$
    – curiouser
    Commented Jul 23, 2021 at 6:08
0
\$\begingroup\$

OK, so you have a panel, or panel array, that pivots to track the sun.

The #1 load acting on that panel will be wind loads. That force will act against the tracking mechanism, and will be hundreds of pounds.

That little motor won't be able to bear wind loads.

So you will need some sort of "one-way action" jackscrew or worm drive arrangement to bear those wind loads. Depending on how you gear and lubricate the jack screw, the motor should be able to turn it. It will take many revolutions to swing the panel around.

Your little panel should be able to make enough power to run the motor, even somewhat mis-aimed. However, you will need external sensors to tell the motor which way to spin.

\$\endgroup\$
6
  • \$\begingroup\$ i will just be testing it under the sun for a short period. Regarding the wind load situation, i will stop testing if it gets too windy or bad weather. Otherwise, i will try to make it run self-sufficiently without the use of a battery for now. \$\endgroup\$
    – curiouser
    Commented Jul 22, 2021 at 7:06
  • \$\begingroup\$ yes, i will be using LDRs to control its position. \$\endgroup\$
    – curiouser
    Commented Jul 22, 2021 at 7:37
  • \$\begingroup\$ A 10W panel will not generate hundreds of pounds ... maybe in hurricane conditions. If the OP is seeing such high wind loadings, consider a wind turbine instead. However a continuous rotation motor and a jackscrew is a sound suggestion anyway. \$\endgroup\$
    – user16324
    Commented Jul 22, 2021 at 12:34
  • \$\begingroup\$ @user_1818839 Nobody puts tracking on a 10W panel, easier to use a 20W panel for $8 more. They put tracking on a 2000W array, and they use a 10W panel to power the tracking motor. That's what I presume we're dealing with here. \$\endgroup\$
    – Harper - Reinstate Monica
    Commented Jul 22, 2021 at 16:27
  • \$\begingroup\$ @curiouser Yeah the problem with steppers is they only want to "seek to" 90 degree positions. Since I gather you want finer control than merely "East", "South" and "West", you'll need some sort of gear reduction. I am merely suggesting to use a "one-way-torque" kind, like a worm or jackscrew, so you aren't spending energy to hold the array against the wind. Once you do that, your design easily scales - it could track a MUCH larger array with no electrical changes (just gear ratio changes). The lower gear ratio would make it rotate slower, but it has time :) \$\endgroup\$
    – Harper - Reinstate Monica
    Commented Jul 22, 2021 at 16:35
0
\$\begingroup\$

im just wondering if the panel is able to sufficiently power the servos for a few minutes.

The MG946R draws ~2 A peak, so 2 servos could draw up to 4 A. Theoretically with a switching regulator the 0.58 A at 17.5 V that your panel puts out in full sunlight could translate to 2 A at 5 V, but practice it will be a bit less due to converter losses.

Clearly this is not enough to power both servos at once. If you only operate one servo at a time you might get away with it, provided you keep the panel fairly well aligned to the sun. To handle peak currents you should add large filter capacitors (eg. 1,000 μF) to the input and output of the buck converter. Symptoms of insufficient filtering include servos jittering and the Arduino glitching, resetting or not responding to commands when the supply voltage drops out during peak current demand.

Next problem is how to reduce servo loading. RC hobby servos are designed for fast movement of aircraft control surfaces or car/boat steering, not a 0.8 kg solar panel. If you connect the servos directly to the panel with enough 'throw' to track the sun over the sky they may stall out or jitter trying to fight the high inertia of the panel. Linear actuators may work better because they move slower, have more torque and lower current draw, and hold their position better when inactive.

Another option you might want to consider is simply using two fixed solar panels facing in different directions. This may be a bit more expensive, but mechanically much simpler and more reliable.

\$\endgroup\$
6
  • \$\begingroup\$ As of now, i have little time to make much changes. Are there any suggestions from you to add on to what I have currently? Is a 20W able to at least run the two servos? thanks \$\endgroup\$
    – curiouser
    Commented Jul 22, 2021 at 9:58
  • \$\begingroup\$ Theoretically 2 panels would be enough, but only in strong sunlight with aligned panels, and only if you add smoothing capacitors (otherwise peak panel current is only ~1A, which isn't even enough for 1 servo). In a typical practical application a small battery or supercapacitors would be used to handle peak operating demand. Then you have to decide what you want - maximum power during peak sunlight, or longest operating hours per day. \$\endgroup\$
    – Bruce Abbott
    Commented Jul 22, 2021 at 10:56
  • \$\begingroup\$ thanks bruce. guess i have to get a battery. btw, i got a mg90s servo lying around. Do you think it could support my 10w solar panel which weighs about 1kg? if it does, i guess i could use it to reduce the power consumption \$\endgroup\$
    – curiouser
    Commented Jul 22, 2021 at 11:49
  • \$\begingroup\$ MG90 servo is far too small for this application. Get a battery. 12V 7Ah SLAs are cheap and have plenty of capacity. Charge it with a PWM controller like this one:- jaycar.co.nz/miniature-12v-3a-pwm-solar-charge-controller/p/… \$\endgroup\$
    – Bruce Abbott
    Commented Jul 22, 2021 at 18:17
  • \$\begingroup\$ Thanks, i got myself a 12V/24V 10A pwm controller instead with a 12V 7.2Ah sla battery. I did not know that those motors have an initial surged in current at starting. Also, how much current is supplied by the 12V 7.2Ah battery? Assuming at peak charged. I understand that current does not equal to ampere-hours, so i assume it can provide 7A per hour or 3.5 A in 2 hours etc \$\endgroup\$
    – curiouser
    Commented Jul 27, 2021 at 11:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.