I'm learning about batteries and I'm really fascinated by this topic.
I saw some online videos on how to build some "Voltaic Pile - like" batteries with coins and other kind of cells in different ways.
The hosts then usually use this batteries to power up simple elctronic devices ( such as flashlights or calculators).
I then saw some videos showing how to replace computer batteries using industrial cells(batteries) in order to rebuild the laptop battery pack.
I was wondering if it's possible to join this 2 processes in someway. Essentially, building some cells (single batteries) from scratch and arranging them together in order to create a laptop battery pack.
I don't know if anyone knows about someone who did it and can link them to me.
I did not find anything like that, and I'm starting to think that If it's not impossible, it is at least very complicated.
Anyway, I don't have the skill and the experience to understand where the complexity actually is, so that's why I'm asking it.

  • \$\begingroup\$ Would it really be worth the time and effort? Or is this just a science project? \$\endgroup\$ – Tyler Mar 21 '18 at 15:45
  • 2
    \$\begingroup\$ @Tyler Honestly, very few projects here will be worth the time and effort. :) \$\endgroup\$ – pipe Mar 21 '18 at 15:58
  • 1
    \$\begingroup\$ So, use lemons and copper strips, then you can make some G&T’s as well... \$\endgroup\$ – Solar Mike Mar 21 '18 at 16:10
  • 1
    \$\begingroup\$ A raspberry pi is still quite wasteful, as it lacks the kind of fine grained power control that an MCU or even a faster-than-pi mobile chip has. Running one on a potentially flakey power supply is also very unwise, as it is a system which requires orderly shutdown to avoid corruption. With enough effort it is do-able, but it's going to be at least an order of magnitude harder project than getting a little ARM or ATtiny to do some proof-of-concept computing on a few lemons and coins. \$\endgroup\$ – Chris Stratton Mar 21 '18 at 16:49
  • 1
    \$\begingroup\$ That would likely require more lemons than you can keep reliably operating at one time. \$\endgroup\$ – Chris Stratton Mar 21 '18 at 16:52

Yes, but it would likely be neither cheap nor compact.

Essentially you would have to

  • pick a chemistry
  • make a string of sufficient cells in series to provide acceptable input voltage
  • make sufficient "strings" to provide adequate output current

You would almost certainly want to pick a rechargeable chemistry, which suggests either lead-acid or NiFe ("Edison"). NiCad is too poisonous and the higher-density lithium ones are too flammable.

There's a community of people building DIY NiFe batteries for use with off-grid renewable energy. Google will show you plans. Or you can look at Edison's original patents.

  • \$\begingroup\$ Thank you for your answer! If it eliminates some complexity, it would be enough if they're able to make the computer run even for only a cylce, just to see it working. Probably not the most intelligent approach, but It would be very helpful to me to eliminate some chemistry related complexity :) \$\endgroup\$ – Gabriele Scarlatti Mar 21 '18 at 15:59
  • \$\begingroup\$ I would also suggest NiFe if you want to make some vaguely useful battery. A more practical project might be NiFe+solar panel+leds = home made solar light. The whole thing lives and works outside, which is where you really want to keep something with wet alkali (or acid). The amount of energy, and required #cells is fairly low. It might be able to run for many years, with just a water top-up now and then, which would be long term, more satisfying that a ran-a-raspberry-pi-once-then-threw-it-away type project. earthineer.com/blog/29533/the-idiot-edison-battery-diy \$\endgroup\$ – Henry Crun Mar 21 '18 at 21:40
  • \$\begingroup\$ One of the good things about the basic lead-acid and NiFe chemistries is that they have relatively good current delivery for reasonable-sized electrodes. So as long as your computer is a low-power model, you only need to worry about getting sufficient voltage and reliability. \$\endgroup\$ – Chromatix Apr 22 '18 at 14:39

The first step is to consider what countsa as the minimal "computer" for this demonstration. I suggest the Raspberry Pi zero as its power consumption can get down to 400mW with a bit of effort. If you back up the SD card before you start you don't have to worry too much about an ungraceful shutdown.

Then you have to build your battery. I'd go for a vinegar battery (probably with copper and zinc) as it lends itself more to big electrodes for higher current than some other designs (though you can always connect in parallel). I'd then power it through a switching regulator like OKI78SR. This would mean enough cells in series to get more than about 6V loaded; you can test this with a suitable resistor before buying anything expensive.

There are lower power processors out there -- you've probably got one in you pocket -- but this gives you a general purpose computer that does things perople understand a computer to do. You'd need power for your monitor though (unless you went for a PiTFT or similar, but even then your battery wouldn't last long

  • \$\begingroup\$ While this is better than a desktop or laptop, it's still unlikely to be practical - you'd need enough cells that it may be hard to even just have them all working at the same time. \$\endgroup\$ – Chris Stratton Mar 21 '18 at 17:04
  • 1
    \$\begingroup\$ @ChrisStratton a vinegar battery could easily be scaled up to cells the size of buckets, with large electrodes, hopefully avoiding the need for parallel connections. I'd actually use something like paint roller trays and sheet electrodes with plastic spacers. 4 in series would be enough to provide sufficient headroom for a switching regulator. My Pi is a 2B, which draws quite a bit more, and I only have linear regulators, otherwise I might be tempted to have a go. \$\endgroup\$ – Chris H Mar 21 '18 at 17:21
  • \$\begingroup\$ You seem to be either overestimating the potential of a copper zinc cell, or assuming you can run a pi at a lower voltage than the manufacturer's specification. \$\endgroup\$ – Chris Stratton Mar 21 '18 at 17:31
  • \$\begingroup\$ @ChrisStratton I'd misremembered it as 1.6V. Having checked you'd need 5 or 6 in series as they're apparently 1.1V \$\endgroup\$ – Chris H Mar 21 '18 at 17:34
  • 1
    \$\begingroup\$ @ChrisStratton Which is the limit of this cells you can have working together at the same time according to you? and What limits this number? \$\endgroup\$ – Gabriele Scarlatti Mar 26 '18 at 15:22

Voltaic Pile-like batteries

The classic Votaic pile was a stack of many small, low-capacity cells. Relatively high voltage, but not much power.

I'd look to the sort of batteries that powered early Telegraph systems. I built something like a gravity cell (a.k.a., "crow's foot cell") when I was a kid in fifth grade. The chemicals are mostly harmless. Never measured how much energy I could get from it, but check out these pages: https://mysite.du.edu/~jcalvert/phys/gravity.htm , http://w1tp.com/mbatt.htm .

You can Google for other examples.


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.