3
\$\begingroup\$

I am building a system using an embedded board that appears to use about 1/4 to 1/3 AMP according to my bench power supply. Is this low enough power consumption for this system to be run off of some battery(s) charged by a solar panel no bigger than 3'x 3'?

EDIT: Why the down votes? Is it not a reasonable question? I'm not really an electrical engineer and I've never worked with solar panels before.

More details:

1- The board will be "active" for about 10 hours a day.
2- Board uses 5 volts.
3- I'm probably not understanding what my bench power supply is telling me about power usage. Each supply has an "AMP" reading in the form of X.XX and is currently reading about 0.22. I "assusmed" that meant the board is drawing nearly a 1/4 AMP. However the documentation for the board says typical power usage is about 250 mA.
4- The board is an Olimex SAM9-L9260 ( AT91SAM9260 ) that is sitting running Linux and sending output to a serial console as fast as possible ( counting ).

\$\endgroup\$
15
  • 1
    \$\begingroup\$ Yes. Or maybe no. Depends on the specifics. \$\endgroup\$ – Ignacio Vazquez-Abrams Sep 23 '13 at 5:31
  • 1
    \$\begingroup\$ Then you're asking the wrong questions. \$\endgroup\$ – Ignacio Vazquez-Abrams Sep 23 '13 at 6:17
  • 1
    \$\begingroup\$ Tough crowd with the down votes. The documentation for the board says it should only be using about 250 mA. I must be misreading my power supply. It only has readings for "AMP" in X.XX format. Perhaps when it says 0.22 that's not 1/4 AMP at 5 volts? \$\endgroup\$ – Chimera Sep 23 '13 at 14:29
  • 4
    \$\begingroup\$ @ChrisStratton the downvotes are to state when a question is not researched well enough, missing information to be able to answer is kind of fitting that, no matter intention of asker. A few edits and people will upvote though. \$\endgroup\$ – Kortuk Sep 23 '13 at 15:00
  • 2
    \$\begingroup\$ @ChrisStratton As someone who has written a million comments to new users, without down votes or closing to drive a question to improvement users often sluff it off, with improvement comes upvotes. \$\endgroup\$ – Kortuk Sep 23 '13 at 15:31
4
\$\begingroup\$

As you say, tough crowd, especially as you've done a bit of research into your actual power consumption.

Your device consumes 5V * 1/4A = 1.25W. Lots of solar equipment is available for use with 12V lead-acid batteries, as used in caravans, so let's assume that. Let's assume that you drive the device through a buck converter (switchmode power supply), so the current draw at 12V is about 100mA. Let's round that up to 125mA for headroom. You should allow at least 48h of operation (through dark days in winter), so you need at least 6Ah of usable capacity from the battery. Choose deep cycle gel batteries, although for this level of discharge a car battery will be an OK substitute if you have one lying around.

Now you need to size the charger. Allow twice as much charge into the battery as you expect out, for charging over an 8 hour period. 12 / 8 = 1.5A charge current. Note that this is a lot more than the drain current! That's 18W. So e.g. this is a little undersized, but the right kind of thing to be looking for. You could buy a panel and charger; whole thing should be under 1m^2.

Don't forget to align the panel optimally and weatherproof everything.

\$\endgroup\$
7
  • \$\begingroup\$ Sure - it can be made to work. But 1.25 watt is a lot of power consumption unless the controller has non-trivial computational work to do. It makes me think that the controller is something like a raspberry pi - designed for computational power/ease rather than efficiency for a battery powered system that may not really need to do all that much work most of the time. A simpler control task could likely be accomplish at a tens to hundres of milliwatts; or even less than that if the processor can sleep with its clock stopped a fair fraction of the time. \$\endgroup\$ – Chris Stratton Sep 23 '13 at 15:18
  • 2
    \$\begingroup\$ @ChrisStratton seems like a darn good answer to me, there's a enough ammo there to be extended to other applications/use cases. Which is what SE is about. (I've run out of votes for today - will +1 later) you're totally in your right to only answer the questions that meet your criterion of optimality. But that isn't reason to complain. \$\endgroup\$ – placeholder Sep 23 '13 at 15:31
  • \$\begingroup\$ @ChrisStratton well - we don't know anything about the application, platform, how much work it's doing, whether it has a radio, or is driving something, whether it has to interoperate with an existing system, whether it's legacy code, whether OP even has all the source ... it would be nice to know, but the core question of "can I power this from 1m^2 solar panel" has the answer "yes". \$\endgroup\$ – pjc50 Sep 23 '13 at 17:27
  • \$\begingroup\$ And I agree that they can, which is why I upvoted your answer. I just happen to think there may be additional aspects of the power situation which merit some thought from Chimera as well. \$\endgroup\$ – Chris Stratton Sep 23 '13 at 17:34
  • \$\begingroup\$ @ChrisStratton Agreed. I'm in the prototype design phase and just wanted to know if I'm in the ballpark for being able to use a solar charged battery. I actually have a lot to learn to complete this project as I've never used solar to charge a battery and I'm a software engineer and know only enough about electronics to be dangerous. :-) \$\endgroup\$ – Chimera Sep 23 '13 at 17:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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