[PStechPaul]

While working on my prototype for a flying capacitor BMS with charging capability for the bottom cell using the entire (4 cell) pack, I got to thinking that, instead of charging that one cell and using traditional shunt current balancing, why not design a charger for a single LiFePO4 or LiPo (or even lead-acid), with integral BMS and daisy-chain communication?

Consider that a 12 kW charger, like the EMW unit I was working on, costs about $2000 and can handle something like 100 cells at 120 watts/cell, and also needs a BMS for safe charging, which may add about $5/cell, for a total of $2500, or $25/cell. I think it should be possible to design a 120 watt single cell charger for about the same cost, and it could be scaled down for a small 16 cell 48 VDC pack for about $400. A 120 watt 3.5volt charger will supply 34 amps or 300 A-h overnight. The same circuit could be scaled down to perhaps 15 watts (4 amps) for small packs as for electric bikes, for maybe $10/cell.

This would provide much greater economies of scale, where runs of 100 or 1000 boards would be feasible. Also, if any single board or cell should fail, it could be bypassed and the remaining cells will remain active with little effect on the overall performance. This is presently just a flash in the pan idea, but perhaps it has merit. Please let me know your opinions. Thanks!

 

[JeroenK]

Sounds like a good idea. I'm wondering where you will leave the single cell charger/BMS and what about the heat produced? The additional weight for cabling? (you need serious cable's for 34A, these need to go to a central power-supply which produced enough current for all the boards. Imagine a EV with 120 or 200 cells (pouches?) what would that do with the cable works and the charger?

 

[Kevin Sharpe]

Please let me know your opinions. Thanks!

With regards to economics don't forget that you can often pick up Tesla 10kW chargers for less than $500

 

[PStechPaul]

The idea was to provide a source of 85 to 265 VAC or 120-400 VDC to each board (which would be a switching PSU), so if each charging module is 100W (3.5V @ 30A), the input would be only about 0.5-1.0A. A #12 AWG would be able to handle 20 cells at 120V or 40 cells at 240V. For 100 cells (350V) the power would be 10kW, which would be split between two or more 20A circuits.

After more thought, I realize that it may be difficult to design an inexpensive and efficient PSU for 3.5V output, but there are 5V 20A PSUs for about $16, so it should be possible. Of course, at that price, it is $160/kw or $1600 for a modest 10kW charger/BMS. I even found a 5V 30A (150W) supply on sale for $10.63:
https://www.banggood.com/DC-5V-30A-...-p-1110689.html?rmmds=search&cur_warehouse=CN

Maybe it would be better to make modules for four or eight cells. A simple DC-DC converter could be made with four isolated DC outputs followed by separate buck current regulators for charging. That would make sense for lithium cells where a 4S pack would be 12-16 volts or so. For lead-acid packs, a single 12-15V charger would make sense. The main idea is to make a module having charger, BMS, and batteries such that they can be used individually or in series up to 350 VDC or so.

 

[JeroenK]

That would e a better solution I think. With a SMPS connected you are able to supply a reasonable current and you have a distributed system for smaller packs. With this method you can easily distribute the packs over the car, place bigger batteries in the car and place battery packs in places where currently no pack can be placed due to the large size. With 3 phase systems public charge connections you are able to charge 11 kWh on a normal charge connection. It's much easier to wire AC 2,5 mm2 trough a car than

The only thing you cannot easily do is fast charging. Maybe you can trough the DC cable's to the Motor control. Use contactors dto disconnect the motor control and connect the DC charging. The Master controller communicates with the distributed BMS boards and the DC (Chademo/CCS), so the battery can be charged safely.