DIY Electric Car Forums banner
1 - 8 of 8 Posts

·
Registered
Joined
·
124 Posts
Discussion Starter · #1 ·
I'm pretty sure this is a bad idea, but would like to understand all the cons. If I had two packs, each made up with different manufacturers modules (but very similar if not identical chemistry), what would go horribly wrong. Presumably I'd need the different modules in different cell groups on the BMS, or two BMS (and all that terrifying complexity) for starters, but I'm sure there are other reasons this isn't done.
 

·
Registered
Joined
·
1,069 Posts
Key bit is voltage - if the chemistries are similar so the SoC vs V table is similar…

and only if paralleling to increase capacity, never try to stack serially to get higher voltage.

But yes sub optimal, more care, knowledge and gear needed to keep things working smoothly
 

·
Registered
Joined
·
1,069 Posts
For serial connections, each member of the string really should be as identical as possible in V vs SoC, capacity, wear level, ESIR.

The pack is only as good as its weakest link, and cell/groups in poor health will accelerate the aging of the pack as a whole.

Plus you must have one BMS control the whole string, not separate ones per group.

With paralleling only gross failure of one member impacts the whole. And each sub-pack can more easily have its own BMS if you want
 

·
Registered
Joined
·
221 Posts
simplest reason is that any capacity above the lowest can't be used and is dead weight. Once the lowest capacity cell is at its lowest acceptable voltage, the vehicle must stop, regardless of how much the others have left in them. I know of a company that rejects hundreds of thousands of dollars worth of cells just to get sets that are within a percent of each other. There is a very good reason for this.
 
1 - 8 of 8 Posts
Top