diy solar

diy solar

Balancing cell capacity in DIY LFP batteries by adding smaller cells in parallel

Sverige

A Brit in Sweden
Joined
Oct 8, 2020
Messages
733
Location
59.5N, 15.5E
Ok, this may constitute heresy even asking this question, but here goes anyway! :LOL:

Given that the weakest cell in a series pack dictates the overall battery capacity (as it reaches LVC soonest), is there anything to prevent the DIY LFP battery builder balancing out cell capacities by adding small capacity LFP cells in parallel with the lowest capacity cells in a series pack?

An example - I buy an 8 cell set of 280Ah prismatic cells which are sold as matched grade A, but on initial testing I find my cells have capacities ranging from 285Ah down to 278Ah. What if anything stops me using some 2.2Ah 26650 size A123 cells (also LiFePO4 chemistry) in parallel with the weakest cells to get them all up to 285Ah so my battery maintains not only top balance, but is closely bottom balanced also?

Apart from the practical difficulties of arranging the cabling so as to avoid any shorts, etc, I can’t see a drawback. Maybe you could argue just the cost of the small cells isn’t justified by the benefit of the additional capacity to the overall pack, but if just one prismatic cell is lower than the others, then adding a few small cells in parallel might bring up the capacity of the whole pack.

Anyone care to comment on how feasible this is?
 
Given that the weakest cell in a series pack dictates the overall battery capacity (as it reaches LVC soonest), is there anything to prevent the DIY LFP battery builder balancing out cell capacities by adding small capacity LFP cells in parallel with the lowest capacity cells in a series pack?


I believe @Maast has been experimenting with something similar to this.
 
Ths approach has been common practice for buyers of used LifePO4 packs such as the BYD units.

In reality though, dont bother, its a lot of work to add such little capacity, its easier to set your max charge voltage a little lower to accomodate the smaller cell and spend the money you save on simply buying more full size batteries.

Im seeing this trend in the USA where new owners are getting more and more paranoid about this technology, due to weird conspiracy theories on groups like this.

Its a battery! Buy the corrrect equipment thats needed, like a good BMS and go ahead an use it.

Do you think the millions of users in China where they make this stuff do all these crazy rituals?
 
Im seeing this trend in the USA where new owners are getting more and more paranoid about this technology, due to weird conspiracy theories on groups like this.

Its a battery! Buy the corrrect equipment that's needed, like a good BMS and go ahead an use it.

This went over my head, can you elaborate on what you mean by 'conspiracy theories?'
 
Ok, this may constitute heresy even asking this question, but here goes anyway! :LOL:

Given that the weakest cell in a series pack dictates the overall battery capacity (as it reaches LVC soonest), is there anything to prevent the DIY LFP battery builder balancing out cell capacities by adding small capacity LFP cells in parallel with the lowest capacity cells in a series pack?

An example - I buy an 8 cell set of 280Ah prismatic cells which are sold as matched grade A, but on initial testing I find my cells have capacities ranging from 285Ah down to 278Ah. What if anything stops me using some 2.2Ah 26650 size A123 cells (also LiFePO4 chemistry) in parallel with the weakest cells to get them all up to 285Ah so my battery maintains not only top balance, but is closely bottom balanced also?

Apart from the practical difficulties of arranging the cabling so as to avoid any shorts, etc, I can’t see a drawback. Maybe you could argue just the cost of the small cells isn’t justified by the benefit of the additional capacity to the overall pack, but if just one prismatic cell is lower than the others, then adding a few small cells in parallel might bring up the capacity of the whole pack.

Anyone care to comment on how feasible this is?
This extremely feasible, and in fact I would even say required for grossly mismatched cells like are in the BYD modules. It has three main benefits:
- 1st and most obvious is that it increases the total available pack capacity.
- 2nd is that it helps prevent unexpected BMS cutoffs when a BMS detects a cell that has exceeded the voltage limitations. You can easily have a runaway cell and not know it from the overall pack voltage.
- 3rd and not immediately apparent is that it helps prevent cell drift. Internal resistance changes throughout the state of charge and if you have one cell thats not matched to the others it'll be out of phase and the effect is that that particular cell capacity will get more and more out of phase as the pack accumulates cycles. This why we have cell balancers at all, its just that they can't compensate with cell capacities too far out of whack.
 
This extremely feasible, and in fact I would even say required for grossly mismatched cells like are in the BYD modules. It has three main benefits:
- 1st and most obvious is that it increases the total available pack capacity.
- 2nd is that it helps prevent unexpected BMS cutoffs when a BMS detects a cell that has exceeded the voltage limitations. You can easily have a runaway cell and not know it from the overall pack voltage.
- 3rd and not immediately apparent is that it helps prevent cell drift. Internal resistance changes throughout the state of charge and if you have one cell thats not matched to the others it'll be out of phase and the effect is that that particular cell capacity will get more and more out of phase as the pack accumulates cycles. This why we have cell balancers at all, its just that they can't compensate with cell capacities too far out of whack.
Thanks (y) I see it as beneficial, and since the smaller cell would have plenty of time to equalise in voltage with the much bigger cell it’s paired with, I don’t think large currents would be expected to flow in the cabling joining the pair. I’m considering using a balance wiring harness to present the voltage from the smaller cell/s to the larger pack.
 
Thanks (y) I see it as beneficial, and since the smaller cell would have plenty of time to equalise in voltage with the much bigger cell it’s paired with, I don’t think large currents would be expected to flow in the cabling joining the pair. I’m considering using a balance wiring harness to present the voltage from the smaller cell/s to the larger pack.
Well like everything else in life, it depends... :) On my turds with a 60 amp load applied I saw 23A coming from my added 30AH of kiitokala cells when the 'host' cell was at 3.0V. I'm thinking it was because of the relatively high resistance of the leads being connected to an aluminum bus bar. It was a lot more than I expected.
 
Back
Top