View Single Post
Old 12th Aug 2020, 4:38 pm   #14
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: "Battery Manager" for a Li ion Battery

Quote:
Originally Posted by Boulevardier View Post
Quote:
Originally Posted by Bazz4CQJ View Post
My 3-cell 12V battery will have the cells permanently wired in, and that's why I was hoping that the BMS would take care of optimising the use of each cell.

The design team has handed over the drawings to production, so I should have it for testing in a week or three .

B
I was wondering about that -the optimising of charging for each cell. What happens when one cell is fully charged, but the others need to continue? Surely this would require bypassing of the charged cell, and a reduction of charging voltage to the remaining ones.
Yes, good balance charging does need this, and the board in the original post doesn't seem to support it. Lots and lots of simple Li-ion BMSes just don't bother, and terminate the charge cycle when the first cell is 'full', so they don't get the maximum capacity and cycle life from the cells. They just assume that the cells are closely-enough matched not to worry about it. I have a 'dead' Dyson battery pack in which the individual cells appear to be OK but their states of charge are really unbalanced: one cell always charges first and the BMS gives up long before the others are charged. This is after hundreds of cycles, though. One day I'll manually balance it and it'll probably be OK for a few hundred more!

There are various ways of doing 'proper' balancing, getting every cell to full charge. You can charge each cell individually from its own supply, which is inconvenient if they're in a series pack. You can have a BMS with a reasonably beefy bypass capability, so it can dissipate most of the charge current for those cells which are getting full. Or you can have a BMS which interacts with the charger, telling it to reduce the overall charge current once bypass starts on one or more cells, so those cells are guaranteed not to get overcharged and the others can slowly catch up. Or, there are some really clever (too clever, in my experience) BMS chips out there which work out each cell capacity during discharge, then try to bleed a small current from those with lower capacities during charge, so they all stay pretty much in sync all the way through charging and should reach full capacity at the same time. It's possible, though unlikely, that this module contains one of those.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline