Be cautious ordering generic blue wrap LFP cells
- Thread starter hour
- Start date
dreadlocks
Well-known member
the best way to practice is with someone else's battery 
luthj
Engineer In Residence
I would hope a new pre-made pack would not need significant commissioning work. Though there will be an initial charge.
While some folks like to use a balance charger to do the commissioning charge in the final series config. My thoughts are to use the parallel method with a calibrated power supply. Basically get the batteries near 90% with a bulk charge, the use the 20A bench power supply to push the parallel cells up to the 100% full mark. No need to push past 3.6V. Sitting at this voltage for extended periods (weeks), will cause some chemical aging. But for the few hours it takes to top balance it is insignificant.
If you mean a sealed up drop-in, often no further balancing is even possible, no access to individual cell wires or even voltages, just trust and send it back when something goes wrong.
But easy on C-rates the first couple months helps with longevity
and getting a SoH benchmark for 100% capacity at the pack level will help as a diagnostic later.
Yes, a powerful enough lab-style adjustable PSU for 1S voltage charging in parallel is ideal.
Some "hobby" chargers can be adapted to do so.
Just sitting in parallel is very slow as V gets closer and closer, but pushing up to 3.6V or the vendor spec max voltage should be the final step below that.
Yes not healthy to go past that, IMO just holding CV until current drops to zero, is too much, even past 0.03C or so, better to stop before that.
Becomes your definition of "vendor Full", used for benchmarking.
Different from the gentler 100% Full defined for normal usage cycling, resetting BMV.
For that the minimum required for getting BMS to do balancing will often be the main factor, that can take days sometimes.
But easy on C-rates the first couple months helps with longevity
and getting a SoH benchmark for 100% capacity at the pack level will help as a diagnostic later.
Yes, a powerful enough lab-style adjustable PSU for 1S voltage charging in parallel is ideal.
Some "hobby" chargers can be adapted to do so.
Just sitting in parallel is very slow as V gets closer and closer, but pushing up to 3.6V or the vendor spec max voltage should be the final step below that.
Yes not healthy to go past that, IMO just holding CV until current drops to zero, is too much, even past 0.03C or so, better to stop before that.
Becomes your definition of "vendor Full", used for benchmarking.
Different from the gentler 100% Full defined for normal usage cycling, resetting BMV.
For that the minimum required for getting BMS to do balancing will often be the main factor, that can take days sometimes.
shade
Well-known member
I'd think not, but I figure there must be a way to exercise it a little the first few cycles.
RC has a few thoughts on the matter, of course: https://marinehowto.com/lifepo4-batteries-on-boats/
When I was digging out a piece of phenolic sheet recently, I think I saw a bench power supply that might be worth bringing home.
shade
Well-known member
Will do, but I doubt there's much of a difference.
I scanned that article a few months ago, and almost forgot about how much information he packed into it.
hour
Observer
Second charge up is nearing completion. Victron BMV shows that the pack has been recharged (from 151ah consumed to 0). Things did perform differently this time - I woke up to find three of the cells at 3.6v and current on smart hobby charger a miniscule 1a. One cell is at 3.35v.
On the last (first full) charge, two cells were at this low voltage reading while the other two sat at 3.6v and charge current had been reduced to peanuts. So uhh, maybe this last weird cell will be onboard with the program in another few cycles.
Connected active balancer again to rectify this, with smart charger running during still trying to get everyone to 3.6v. This should be fixed in a few hours and then I'll do a partial discharge, charge up (active balancer on whole time) and see if we're getting there..
On the last (first full) charge, two cells were at this low voltage reading while the other two sat at 3.6v and charge current had been reduced to peanuts. So uhh, maybe this last weird cell will be onboard with the program in another few cycles.
Connected active balancer again to rectify this, with smart charger running during still trying to get everyone to 3.6v. This should be fixed in a few hours and then I'll do a partial discharge, charge up (active balancer on whole time) and see if we're getting there..
For this testing period, only use that active balancer
as necessary to speed up the balancing process
and only at the top part of the charge cycle.
It may well interfere with the top balance, if deployed at lower SoC, only while charging OK?
The hobby charger is stopping charging as designed, while trying to bleed/burn off Ah from the higher cells, then going back to charge them all, then stopping to bleed (loop)
a super inefficient method, only used because cheap to implement. Same as most cheap BMS.
Make sure your connections are all the same, and well torqued.
Swap positions maybe to test.
Diagram the intra bank wiring?
as necessary to speed up the balancing process
and only at the top part of the charge cycle.
It may well interfere with the top balance, if deployed at lower SoC, only while charging OK?
The hobby charger is stopping charging as designed, while trying to bleed/burn off Ah from the higher cells, then going back to charge them all, then stopping to bleed (loop)
a super inefficient method, only used because cheap to implement. Same as most cheap BMS.
Make sure your connections are all the same, and well torqued.
Swap positions maybe to test.
Diagram the intra bank wiring?
Maybe just pack-level charge, disable balancing with the hobby charger.
Just use the Deligreen balancer, and watch the cell voltages manually to ensure not going too high.
If balancing still takes a long time, tune the current input lower so it's just enough, the Deligreen can keep up redistributing.
Just use the Deligreen balancer, and watch the cell voltages manually to ensure not going too high.
If balancing still takes a long time, tune the current input lower so it's just enough, the Deligreen can keep up redistributing.
