Solar charging your car, camping or emergency batteries



Huffing Boing Boing

FYI: Instead of the pigtail connectors, the amateur radio community commonly uses Anderson Power Pole connectors…

It seems pretty decent to me, and being able to just power some amateur’s radio gear in an emergency seems like a nice bonus.

Great write-up. Thanks. I might have to build something like this myself.


The pigtails just kinda happened to me but they really work well. They are simple and rugged. Some of the BBS readers suggested they are also useful as a power source for things like heated vests and seat covers on the motorcycle. I don’t trust the bikes charging systems enough for that but…


Solar panels are an excellent thing to have around. They however tend to be costly-ish and the power rapidly falls when the sun is not optimal. If money allow, go for more panels rather than less; the bigger the panels the less sun you need for satisfying performance. Worst case, the excess power gets unused. (Balance this with cost and size/area/weight of the assembly.)

(Solar charging is simple when you use off-the-shelf DC-DC conversion modules. Designing your own will lead you to the murky territory of buck-boost converters and finding the optimum power point, where the voltage on the panel times current taken from it is the highest. Go off that and the efficiency falls. Nothing that a microcontroller couldn’t handle, though. But the programming will be a bitch.)

Standardization of interconnects is definitely The Must. I mod my electronics to charge from microUSB for 5V and 5.5/2.1mm barrel jack for 12V or 19V (the laptop power supply standard). For charging portable power I use a Hobbyking-style microcontroller-equipped charger, which runs up to 18V (and then claims overvoltage), so a switch with some voltage-dropping diodes in series was built in to happily run from a laptop brick. (The laptop was also modded to have the same connector.) Mix and match for the win.

Also, beware of broken solar cells. They are nice and handy if you want to make yourself a cheap panel and have some startrek reruns to watch. But the caveat is, you can solder only to the solder pads on the bottom and the collector strips on the top. Makes it difficult to assemble higher voltage panels from smaller cell fragments without these features. But definitely possible.

The thin “fingers” over the top side are likely aluminium or something else; no solder alloy nor flux (even aluminium-grade that gets cooking foil wetted by tin-lead) sticks on it. The conductive backing, exposed by scratching off the protective varnish at the bottom, won’t stick as well. I don’t trust adhesive copper tapes to maintain conduction over time, through the sticky layer (maybe a prejudice).

So conductive paint it is, with epoxy overcoat for mechanical strength, and that will be tested later, together with using diamond-wheel dremel to score and cut the wafers to smaller pieces; they are too fragile to use conventional glass score-break approach. They also sorely need some support; a choice is either gluing them front-to-acrylic, or back-to-circuitboard. The mount plate material has to not flex because the buggers WILL crack when you look at them wrong.

Expect results in couple weeks to few months, it will be a slow-mo project set for long winter nights. (I thought it will be more straightforward. It never is!)

That said, they are fun!

Is there a chance to find somewhere a conductive hot-melt adhesive? If the PVA kind is filled with high enough proportion of metallic particles, would it conduct through volume while keeping surface stickiness? Hot melt, unlike epoxy, is better for repairs/reworks and overall serviceability.

Another “one weird trick” for battery systems is using the Polyswitch fuses. They have the disadvantage of higher series resistance than a regular fuse (so can insert a significant voltage loss), but the advantage of reclosing after cooling down. (Can be also used as a crude overtemperature shutdown. They act as a self-heating nonlinearly heat-dependent resistor, the warmer the higher resistance. If they self-heat above limit, their resistance skyrockets from few ohms to orders of magnitude more, and is kept there by the increased power loss on the part. So you can combine the effects with external heating by e.g. a transistor heatsink, to make them more sensitive to overcurrent/trip as temperature of the critical part rises.) The Polyswitches have some quirks, so testing that the particular choice won’t trip under sustained correct load and will under sufficiently low overcurrent (so the wires won’t act as a fuse instead!) in controlled conditions (e.g. not-camping) is strongly advised. (A test a day keeps Murphy away.) But once set, it will work and you won’t end up in the middle of nowhere with burned fuse and no spare.


I remember those… The robots in FIRST competitions used them for the hookup between the battery and the onboard electrical system.


I’m an ex-solar cell designer, still in the business but in a different (albeit still technical) capacity. I can explain a bit what you’re seeing here.

The fingers on the front side are silver, made of exactly the same stuff as the busbars/solder pads/collector strips (or whatever you want to call them – I find that people in different segments of the industry have different names for them…). They’re applied at the same time using the same stuff and processed identically. I suspect you have trouble soldering to them because they are so narrow that they are probably difficult to wet. Just a theory – I have no proof for it – but on modern cells their width is about 1.5-2X the diameter of a human hair, and on reject cells (which you will often find amongst broken cells) defects like discontinuities in these fingers are common. Even if you were to get a connection, I think the slightest stress on the ribbon would break it.

On the backside, the strips you can solder to are also silver, though with a bit of aluminum mixed in. The area around this is almost entirely aluminum. The “varnish” you mention is actually just oxidized aluminum – nothing special is applied to the back of the cells. When you scrape it away, you are exposing more-or-less pure aluminum, which is why you cannot solder to it.

All of the above assumes you’re using the typical sort of cell that is produced by almost every solar cell company on the planet. If you get your hands on something else, what you see can range from slightly different to totally different.

I’d encourage you to experiment with the conductive adhesive copper. Nobody in industry is using it, but honestly speaking, you probably don’t have the capability to hermetically seal your cells at home – it literally took decades for teams of professional equipment designers to get it right (and plenty of low-end manufacturers still screw it up). I suspect that if you get good conduction to begin with, the adhesive will last as long as a homemade panel. On the other hand, it will probably never be as conductive as a well soldered connection, so you might maintain performance over the life of the panel, but start at a lower performance level. Again, just a hunch – I haven’t tried it myself.


A solar powered turtle pond has been on my to do list for a couple years now…
It has been interesting watching the efficiencies rise (a bit) and the prices drop quite a bit on the panels – my procrastination has paid off to some degree. I started looking at the kits, but as @jlw pointed out, one can find better solutions for about the same price by getting the separate pieces. I still have some time for the solar panel part, since I plan on running it off of a 12 volt supply for the first 6 months or so until I am certain that I have the motor/etc mix that I need, although I would like to start using battery backup during the night right away (I think that I should be able to use the solar charge controller for this). The hardest part is finding decent 12 volt pumps – I am currently planning on using bilge/livewell pumps. I have been running my test-bed turtle-tub with a bilge pump for about 3 months now and it seems to be working ok. My current plans would have a power draw of ~100 watts during the day and ~50 watts during the night, so I was planning on ~250 watts of solar capacity.
I will hopefully be breaking ground sometime in the near future (whenever temps start staying below the 90-100 F range for a while).


Just mount some panels on the bike.


Yes, 12V is easy. By the way, Jason, your “400 watt battery” isn’t a 400 watt battery, it’s a 12V, ?? amp-hour battery connected to a 400W inverter.

Amp-hours is the key factor here. Unfortunately, solar panels are sold by watts rather than amps, which is silly, as they are intended to charge 12V batteries, so the amps flow directly into the battery with no voltage conversion (just voltage loss!).

Most solar panels come with a current rating at the typical sunny day exposure. If it’s tilted directly at the sun at the meridian, then 6 hours a day is a good estimate of the total exposure you’ll get (varies summer to winter).

I’ve put solar panels on everything from pirate radio transmitters in the mountains to my Loud Bike at Burning Man. They rock! But I live in Sunny Arizona.


Yes, thank you! @thaumatechnicia also got me straight. I knew there was something wrong with that number. Its all about Cold Cranking Amps with car batteries, however the inverter is the most interesting part of that rescue battery, imho.


Loud Bike was cool, but that video coat! OMG!


Looks nice! What’s the thrust-to-weight ratio at full volume?
[ducks and covers]


Thrust to weight is very low today, as it’s been raining too much to charge the batteries. Oh, well. Next week.


Did you consider solid-fuel boosters? That thing could FLY!

(…brain, stop. No going off on tangent, making up The Legend of the Space Bike!)


Speaking of that…

Both a speaker and a solid fuel rocket function by producing a suitably crafted series of pressure levels over time at their output point.

With a speaker, control is relatively easy, because electromagnets are handy like that and you can just feed them an electrical signal.

With solid fuel propellants, control is relatively tricky; because burn rate can vary according to propellant geometry(which itself changes as the propellant burns) and any irregularities in propellant composition, and the actual pressure at the output will vary according to burn rate and the degree of expansion provided by the chemical and thermal changes.

However, in principle, if you could exert fine-grained control over the composition of a solid fuel propellant slug, rather than the more usual homogeneous solid or powder-fill, you could thereby control the output pressure at each point in time over the duration of the propellant burn.

In addition to be absurdly foolish this would allow you to ‘encode’ the sounds of your choice(realistically, probably the low frequency sounds of your choice) into a slug of solid fuel and deploy a (one time use) solid fueled speaker, quite possibly one of the few audio devices more Totally Metal than tesla coils.

In terms of implementation, combining a controlled mixing system(to deliver fuel slurry with the demanded mixture of more and less energetic fuels, inert filler/binder, etc.) with an extrusion type 3d printer might allow you to get the necessary control over fuel slug geometry and composition (ideally without introducing enough voids to have the thing blow itself apart).

First implementation to produce a recognizable rendition of a well chosen musical work shall be awarded the sum of 100 internets.

First implementation to achieve escape velocity by sufficiently energetic rendition of Ride of the Valkyries shall be awarded the sum of 1000 internets.


Or you can go the hybrid way, use a solid fuel (without oxidizer) and inject oxidizer from a separate tank. Modulating the flow of the oxidizer allows throttling the engine, and should provide control over low frequencies.

High frequencies could be added by modulating the exhaust gas temperature outside of the nozzle, possibly by focused microwaves while it is still ionized and absorbing, resulting in a form of a plasma speaker. There would be a lot of underlying noise from the operation of the engine itself, though, which can not be easily regulated.

A proof of concept of non-electronically-assisted sound/music generation should be doable as not a thrust-producing motor but as a quick-match. Vary the combustion speed by e.g. varying the thickness of the black powder/binder layer on the carrier filament. Could be possible to produce using e.g. modified 3d-printing extruder (there is a plenty of variants for different materials). (Thought… a low-melting (~120’C) hot-melt adhesive composition, if doubling partly or wholly as a fuel, would it be a suitable binder for quick match? Could be then extruded in varying thickness/width easily using constant-speed extrusion and regulated-speed carrier, whether a filament or a tape.)

Also a thought for more energetic composition than quickmatch: burn as a series of tens to thousands eplosions of discrete propellant injections per second. That itself should provide a recognizable melody. But the load on the chamber would be very significant. (Actually, those weird baffles in bigger rocket engines serve to suppress combustion instabilities that cause resonances that can be so bad they cut the chamber off like an ultrasonic knife.) Thought: use hypergolic fuel (N2H4/N2O4, perhaps?) and car-style fuel injectors? Probably not enough for significant thrust, the fuel injectors have rather pathetic average flow rate, but enough for a demo.


This topic was automatically closed after 5 days. New replies are no longer allowed.