correct design of small solar power system with low voltage and good voltage indication?

line nicely from 12.5v to 0v. I don't know when to consider it time to shut down the processor. Lead acid batteries are apparently damaged if you drai n them too low, but apparently the voltage isn't a good indicator of that p oint?

you measure the battery, when the voltage drops under 11V you shut everythi ng down until the voltage raise over 11.5V. you don't want a lead-acid go under 11V.

instead of arduino use this (don't know if it works with rpi zero though):

Bye Jack

ritto:

ecline nicely from 12.5v to 0v. I don't know when to consider it time to sh ut down the processor. Lead acid batteries are apparently damaged if you dr ain them too low, but apparently the voltage isn't a good indicator of that point?

hing down until the voltage raise over 11.5V.

:

yep, it works with the zero:

Bye Jack

"Fight to keep providing 5V no matter what" is a pretty good description of a properly-designed regulator, so, yes.

They don't decline nicely, but the knee in the discharge curve is soft enough that if the battery will keep the thing up for more than an hour or two from fully charged, you'll be able to detect a voltage level that will give you a minute or two to shut down. Check out batteryuniversity.com -- there are articles, and discharge curves.

If the AC power is easily come by, then you'll probably spend less on really high-grade commercial surge protection than you would on full solar. For that matter, a near-enough lightning strike would fry a solar- power system, too, although "near-enough" would be a lot closer for a well-designed solar system than for a well-designed AC system.

Of course, if you have to run the AC any distance to the station, you may find that solar is cheaper and less fuss.

Get a commercial solar power system that provides a 12V rail. Use an efficient 12 to 5V converter, and watch the 12V rail to know when to shut things down.

You'll probably want to experiment a bit to know where to set your turn- off and turn-on voltages, and keep in mind that as the battery ages its capacity will diminish.

--
www.wescottdesign.com

te a solar power system for a weather station. I want it up and running as much as possible, so there's a lead-acid battery involved.

n't resist that) needs to be properly shut down when the power fails, or th e disk is corrupted.

Sure would be easier if you could avoid that

lable voltage is below 4.8v - I think. It can need up to a minute to shut d own. I also need the power removed and then reapplied when the battery is g ood again, so there's an arduino involved to handle power management.

A regulator with shutdown & added hysteresis could do that

but more capable pi).

I might have gone for 6v

days. The solar charger will sometimes dump 16v or maybe even 20v into it. So everything hooked up needs to tolerate at least 20v.

If you're seeing 20v on the battery something is very very wrong. 14.4v is fast charging.

w do I know when the power level is sagging?

voltage

ding 5v no matter what, until the voltage suddenly falls off a cliff and th ere's no time to complete shutdown. Low battery and a cloud, bam.

line nicely from 12.5v to 0v. I don't know when to consider it time to shut down the processor. Lead acid batteries are apparently damaged if you drai n them too low, but apparently the voltage isn't a good indicator of that p oint?

voltage is precisely the indicator. Set your shutdown to 11v and you've got lots of time.

AC. A lightning strike nearby fried it, presumably because enough current g ot induced and found a path to ground via the power line. My thinking is if the weather station is 1) battery powered, 2) uses wifi to communicate and 3) has most of the electronics indoors and somewhat shielded with foil, I can mitigate that problem. But I don't know if that's true. Is the solar pa rt a fool's errand?

lose the foil, that won't do any good. Replace it with a copper hoop around /over the electronics if a dirct hit is possible.

ing iron and may need it explained with part numbers and small words. TIA.. .

A 12/24v mains derived feed is so much easier.

NT

What you need is not just a DC-DC converter, but an MPPT (maximum power point tracking) solar panel battery charge controller. Yes, they exist. Or you can use two different units, one to convert the solar cell output to 12 volts (a DC-DC step down with special features) and a 12 volt input battery charger (a DC-DC step up with special features). Then your computer equipment can be powered by a 12 volt in (12 ish) to

5 volt out which is a relatively simple piece of gear.

Yes, unless it has an input under voltage detector that will give you the shutdown signal. But that is the regulator saying it won't be able to continue to regulate, rather than saying the battery voltage is too low for the battery.

Actually I think a battery *will* decline pretty nicely from 13 volts to zero, but you don't want that. You want to keep some minimum voltage on the battery which although may depend on temperature, is well defined.

A someone else said, the lightning damage to the last one wasn't from lightning getting into the equipment and finding a way to ground... it's from the lightning current inducing current into the power lines and giving your equipment a nice fat over-voltage.

I don't know how much it will cost to prevent that sort of surge damage because it all depends on the size of the surge. I worked on an alarm circuit that got zapped everywhere it ran close to a power wire. It actually melted the relatively light gauge alarm wire at each juncture. How can you prevent a surge like that from getting into gear? A solar powered unit will resist all but a nearly direct strike. Just keep your wires short as reasonable and twist them to minimize pickup.

--

Rick C

No. The battery (unless it is toast) will always clamp the output of the charger to 12V (probably 10-14, worst case). If the charger can pull the battery voltage (AT the battery's terminals) above its nominal unloaded voltage, then the battery needs to be replaced.

Depending on the converter/regulator, the battery can fall to a volt or two above 5V and it (converter) will still perform properly. A 12V battery that has dropped to 7V is long past dead! :>

I.e., you want to watch the battery voltage, not the converter's output. Then, arrange to shut down the system before the battery voltage gets to a point where it WON'T be able to hold up the load as seen through the converter.

See above.

Note that there are other issues that you've glossed over. E.g., how long will the battery sit with the load still hanging off it (even if powered "off") in the worst case scenario (probably only a day or two).

Also, consider how temperature is likely to affect your circuit/components. Will all this stuff be buried below grade? Or, sitting under that very same Sun? Will the site see periodic maintenance/inspection? Or, is this a "set and forget" installation?

[Uninsulated garages, here, routinely see temperatures of 140F in the Summer. Some components sitting under a box out in the Sun would likely see worse. Batteries don't like heat. The characteristics of electronic components change with temperature. etc.]

Does the system need to stay on continuously? Can it, instead, wake up every minute (or five), make some observations, report them (logging them as well) and then shut down, again?

[One advantage to this type of operation is that it ensures the system is periodically reset instead of hoping that it will run properly indefinitely]

Completely not true. There are any number of reasons why a spike can happen. The battery does have a finite impedance and 20 volts is not unheard although I would be unusual. There are ways to protect a load if that is important, but I would expect nearly any 12 volt powered DC/DC converter to handle 20 volts on the input. Just check the data sheet.

--

Rick C

There's nothing in the OP's design afaik that would cause a huge battery voltage spike. It's not like a car with a massive inductive alternator and very heavy loads being switched off without warning.

Of course that will be different once the battery is totally & utterly dead, and is effectively o/c. so if not maintained one should still design for it.

NT

Charging a lead acid accumulator it is unlikely that the internal resistance of a modest solar panel will allow it to stress the battery at anything beyond 14.4v or C/10 for a nominal 12v lead acid battery.

You can allow the solar panel to always provide at least the current drawn by the weather station plus a little bit extra. The simplest crude solution would be a network of resistors and diodes to allow a fast charge from 11v up to 14v and then tailing off over 14v to 14.5v.

Since a nighttime partial discharge is inevitable we don't have to worry too much about trickle charging for too long and too high.

A DIY lead acid charge controller is still quite likely to kill the battery it is used with unless temperature compensated. See for example:

The thing that needs defending against is continuing to charge it beyond the maximum permitted voltage at the possibly high ambient temperature in summer when the sun is shining and high in the sky.

The other failure mode at least in the UK is in midwinter when the solar panel gets only 4 hours of feeble low sun and usually just thick cloud. The result is that in winter active bend warning signs are never working on cold frosty mornings when they might actually do some good. They trash their batteries every winter almost without fail.

Last time I thought solar power would do what I wanted off grid I found two 7Ah SLA's turned out to be a way more economic solution than solar.

--
Regards, 
Martin Brown

You aren't clear. How do you use rechargeable batteries to replace an energy source? Batteries run down. I assume you had some way of recharging them? That is the important part.

--

Rick C