BMS or DC conversion? Which is better?

[tigertoo]

Hi All,

I currently have a project and looking for some advice on the best way to approach battery management?

The project is creating a li ion battery from 50 x 18650 cells with the desired voltage of 36v.

The current method is to wire 10 cells in series (5 batteries in each cell) giving a total of between 30 and 41v dc (depending on charge state) and approximately 10Ah of current. The problem with this method is keeping all cells in balance and the typical way of getting round this is using a battery management system. I believe there could be a better alternative but I am unsure of the feasibility. What if instead of using a BMS to balance each cell to the nominal voltage have all cells wired in parallel to achieve the following:

3.0v to 4.1vdc and approximately 100Ah of discharge current available then via a DC to DC converter to convert the voltage to a constant 36v with a maximum output power of around 500 watts.

The benefit of doing this is that all 50 cells would naturally be balanced as they are wired in parallel and would not have to reply on a cheap or possibly inferior BMS to manage the voltage of multiple cells. I believe that the battery life could be enhanced by looking at the problem this way.

The drawbacks of creating such a DC to DC converter could be
-conversion inefficiency
-cost
-bulky size.

Now I am unsure if any of these issues can be overcome but there would be advantages if they can.

Typically the DC to DC converter would have to be around 90% efficient as this would match the inefficiencies of a BMS.
Extra cost could be warn by battery life extension and less requirement for a specialised charger.

So I am mostly looking on advice on wether going down the DC to DC conversion route is a viable route or if there is anything I might have missed?

Regards!!!

tigertoo.

 

[FreddyNurk]

What's the expected short circuit current of 50 batteries in parallel like?

 

[itsmoked]

Interesting problem.

I think its performance would not be what you expect. I have a fist full of 18650s I run a flashlight with (one cell) and they have surprisingly different voltages and charge to different voltages. I think that would translate to quick cell damage and lower than expected pack capacity.

Perhaps if the batteries could be rigorously qualified for voltage balance it would work a little better but that's very time consuming and doesn't work all that well with pack aging.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[tigertoo]

@ itsmoked

All the cells I have are the same make, age and same rated voltage. i.e. min 2.5v max 4.1v When cells are wired together in parallel they naturally balance to the same voltage. It is true that there are slight differences in voltage between different li ion chemistries but as long as you didn't charge past the maximum voltage of the lowest voltage cell then it should not be an issue. This would also be the same of discharge.

@ freddynurk

I am unsure to be honest but it would be quite high. max constant discharge for each 18650 cell is around 2.2A @ 3.6v which would give 50 cells in parallel a total of 110A @ 3.6v constant discharge but they could handle a temp peak much higher. For the DC to DC converter I would limit the max current to 10A @ 36v.