I am planning on building an ~60V, ~50A, 18650 battery pack to power a 2 kW electric motor. I've called some electronics stores and it seems like I can get a decent amount of old laptop batteries for free. The obvious problem is that each cell is going to have different lifetimes, capacities, resistances etc. so ill need some way to balance them.

How can I make what is probably going to be the most hack battery ever, as reliable as possible? Should I buy a few BMS's and try and balance each individual cell? should I balance each parallel group? should I not even bother trying this and shell out money for real batteries?

  • 5
    \$\begingroup\$ Of course you should not even bother trying this etc. But ignoring that, you should match each cell in every parallel group as closely as possible, when it comes to internal resistance, capacity and C rating. Match every parallel group to each other as closely as possible. Balance each parallel group. Overrate everything by at least 100%. Keep a fire extinguisher at hand. Pray to your deity of choice if you are so inclined. Call the fire service and ask them to stand by. Then grab your blast shield and press IGNITION. It'll be glorious. \$\endgroup\$ – Dampmaskin Sep 26 '17 at 15:11
  • \$\begingroup\$ I plan on getting enough batteries that I can pick and choose and hopefully match them well enough. If I do everything right do you think that there is still a really high chance of me blowing myself up? I was hoping to use this to power my daily driver DIY ebike. \$\endgroup\$ – teepers Sep 26 '17 at 15:18
  • \$\begingroup\$ the balancer must absorb any mismatch in capacity and ESR \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Sep 26 '17 at 16:00
  • 3
    \$\begingroup\$ Don't do it. Laptop batteries are not optimized for high drain, and the fact that you will be picking over discarded packs means most of the batteries you get will be old. The cells used in power tools and E bikes are different from the ones in laptops. Also, you will be investing a lot of effort and money building a pack that may have a very limited lifetime. \$\endgroup\$ – mkeith Sep 26 '17 at 16:12
  • \$\begingroup\$ Electrical Vehicle industry strives for reliable and efficient batteries. The industry and academia has some substantial research on this. Which articles did you search and study before asking this question? \$\endgroup\$ – Ale..chenski Sep 26 '17 at 17:57

You certainly can have some fun doing this project. Dead Laptop batteries still might have few fairly good cells to harvest. Keep in mind however that the internal cells don't have any protection, so be careful not to short anything at a risk of "rapid venting with fire".

There are some hackers who do this kind of projects and offer their advice, like this one. In most cases take their advice at your own risk, since usually their methods are not backed up by any test nor reliability/safety research. For example, they would boldly connect 18650 cells in parallel in large numbers without any reservations, which doesn't make them overly reliable and long lasting.

For a overview of some challenges in the area, I just found this nice article, which I highly recommend for a start.

Finally, I am afraid you can't just "shell out money for real batteries" and expect you EV bike to fly. One battery doesn't make a EV. You would need a full EV solution for power management, braking recuperation, etc, all sorts of fine engineering.

| improve this answer | |
  • \$\begingroup\$ Batteries in parallel are the most stable configuration. It's the least voltage across a maximum number of cells in a self-balancing architecture. As long as one of the batteries didn't start with a short before adding it to the parallel set, it's much more difficult to go wrong vs series connections. \$\endgroup\$ – horta Sep 26 '17 at 18:29
  • \$\begingroup\$ @horta, if you plan to charge-discharge your battery pack 10-20 times, then yes, cells in parallel are the simplest hack. However, if you plan to ride on this battery for a bit longer, here are some studies on the subject., sciencedirect.com/science/article/pii/S0378775313019447 \$\endgroup\$ – Ale..chenski Sep 26 '17 at 18:45
  • \$\begingroup\$ Very nice read. Thanks! They do mention there that if you ensure you aren't pushing the cells to the very limit when you're charging or discharging, then you'll mitigate this issue. \$\endgroup\$ – horta Sep 26 '17 at 20:00
  • 2
    \$\begingroup\$ E bikes are always high-discharge applications, because nobody is willing to carry a large enough battery pack to cruise for hours. Laptop batteries are pretty much always NOT high-discharge packs. High discharge packs have much less margin for error. What the OP is proposing to do is, in my opinion, time poorly spent. Because a lot of work will be required to make a battery pack which may not last very long or perform very well. I think it would be much better to buy some new high-discharge cells and build a pack with them. \$\endgroup\$ – mkeith Sep 29 '17 at 4:14

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.