I need to power a number of tp4056's with the battery outputs in series (using tp4056's because I have a bunch already, and because I want to be able to add more cells in series later). If I do this normally It will short out the cells so my options are to either charge in parallel then have a 4-6pdt switch that can take >5 amps of current (pretty damn hard to find and wouldn't allow usage and charging at the same time, or as I saw in a tutorial use isolation on multiple tp4056's with some B0505S IC's. These apparently cannot take alot of current and I would like to charge quickly so I need another option.

-edit: I may be an idiot but I just thought about the fact that I could just use two B0505S 's in parallel for each battery and charger.

  • \$\begingroup\$ An isolated SMPS? \$\endgroup\$ – winny Jul 26 '18 at 21:06
  • \$\begingroup\$ I will be using 4 tp4056 modules, each needs an isolated power supply. I would have to use 4 smps. \$\endgroup\$ – 5hutd0wn R35tart Jul 26 '18 at 21:08
  • 1
    \$\begingroup\$ Trying to charge multiple cells in series with separate TP4056 from a single 5 V is complex so no one does this that way. There are modules / charge controllers designed to charge cells in series. You should get one of those instead of insisting on "hacking" a solution out of components that aren't designed to be used like you want. \$\endgroup\$ – Bimpelrekkie Jul 26 '18 at 21:19
  • \$\begingroup\$ You can owe one SMPS with multiple floating secondaries, but as Bimpelrekkie said, it’s not the best way. \$\endgroup\$ – winny Jul 26 '18 at 21:22
  • \$\begingroup\$ @Bimpelrekkie Ok... please find me a 4s charging board (note a normal BMS does not charge cells properly and actually reduces the life of the cells) and that doesnt cost upwards of 50$ \$\endgroup\$ – 5hutd0wn R35tart Jul 26 '18 at 21:26

While it is possible to charge battery cells connected in series individually by using isolated power supplies and chargers (or by using one supply and one charger and switching them between cells) it is not a good idea.

The engineers trying hard to ensure battery remains balanced, hence all those additional harnesses for measuring individual voltages and battery balancing devices.

Even small discrepancies (due to manufacturing process or component precision) in multiple chargers will lead to unbalanced battery. Using single charger switched between cells is much better in this regard but it will increase charging time, of course. And it creates a possibility for accidentally skipping some cell with drastic results.

There are really good charges for up to 6 cells that will not break your bank, like iMAX B6 and will make the whole process much safer. I'd suggest investing in one.

But if you really inclined on doing this your way, already suggested option of using SMPS (or even good old transformer) with multiple outputs is certainly plausible. It might cost you even more than the charger, though.

  • \$\begingroup\$ I unfortunatly have to have the charger integrated into my project. Do you know how large that Imax's actual circuit board is? I may be able to rip it out and fit it into my enclosure. \$\endgroup\$ – 5hutd0wn R35tart Jul 26 '18 at 21:28
  • \$\begingroup\$ What if I were to use a cheepo BMS along with my charging set up? \$\endgroup\$ – 5hutd0wn R35tart Jul 26 '18 at 21:30
  • \$\begingroup\$ The whole charger is only 105 x 85 x 30 mm \$\endgroup\$ – Maple Jul 26 '18 at 21:32
  • \$\begingroup\$ I kinda wanted to make it completely modular. with individual chargers, individual protection circuit, etc. Would it be safe if I used protected panasonic cells? \$\endgroup\$ – 5hutd0wn R35tart Jul 26 '18 at 21:37
  • \$\begingroup\$ Good BMS is certainly a right way to go for any integrated solution. But it is neither simple nor cheap. And I've never seen multi-cell BMS designed for individual cell charging. Using protected cells reduces chances for failure, for sure. But it does not prevent unbalancing the pack. \$\endgroup\$ – Maple Jul 26 '18 at 21:39

There is a good reason that people take the BMS approach and don't do it this way: cost, complexity. Anyway you asked so ....

The simplest way is get a bunch of standard DC-DC convertor modules. (beware they are not short circuit protected - thats up to you)

The best way is to use a transformer with multiple windings, as you get true isolation against static and dynamic voltages. You may get away with 1 winding per 2 cells. Note this transformer arrangement is actually the simplest possible charger/bms system, as no charger chip is needed per-cell. It requires very careful design of the transformer and the controller to work properly however.

You can also use capacitive coupling. You need to watch startup/inrush currents. It also assumes that the voltages throughout the stack are constant and even. This would be true for light loads, maybe not true in an electric car. (obviously teh drivers shown are placeholders for some sort of push-pull driver of appropriate voltage and current rating)

In this schematic, the "battery" is a placeholder for the OP's charger module.


simulate this circuit – Schematic created using CircuitLab

(You might be able to put another bridge and battery between C2,C3. I can't be bothered working it out.)

  • \$\begingroup\$ Sorry, but if you do not replace those batteries in circuits with something like capacitor + charger module this answer deserves severe down-voting. These are lithium batteries OP is asking about. \$\endgroup\$ – Maple Jul 26 '18 at 22:06
  • \$\begingroup\$ @Maple The "battery" is a placeholder for the OP's charger module. However your comment is not strictly true. With very careful design, of the drivers and transformers a simple system that fully complies with the requirements for lithium battery charging can be constructed with direct connection to the cells. \$\endgroup\$ – Henry Crun Jul 26 '18 at 22:12
  • \$\begingroup\$ I don't know how you can comply with requirements for lithium charging without feedback. Also, the OP charging modules won't be able to work on pulsed voltage, they don't have enough capacitance on input. \$\endgroup\$ – Maple Jul 26 '18 at 22:25
  • 1
    \$\begingroup\$ @maple Standard chargers do it all the time. Series charging is done assuming matched cells, without regard to individual cell voltages, and a BMS is relied on to prevent the gradual accumulation of imbalance over multiple charge cycles. \$\endgroup\$ – Henry Crun Jul 26 '18 at 22:34
  • \$\begingroup\$ I hope you realize that standard chargers have a feedback too? Whether it is from each cell or entire battery is not important. I was commenting on your "very careful design, of the drivers and transformers ... with direct connection to the cells" which implied no feedback whatsoever, as on your schematics. \$\endgroup\$ – Maple Jul 26 '18 at 22:43

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.