Any harm in using a higher balancing current?

Klauts · Feb 21, 2023

My Monstrous Beast

(14S10P, 140 Samsung 18650 26j cells) pack is back in actions and sufficiently reinforced and shrink wrapped I do believe.

Now I'm trying to get the settings on my Daly Smart BMS dialed.

I can set the active balancing current anywhere from 0.050ma to 1a and it can be set to start balancing at any voltage from 2.75v - 4.2v

The cells charged at pretty sporadic rates the first charge and I had set the balancing activation at 3.6v initially for that reason. After a gentle discharge cycle and then putting it back on the charger the cells all charge at identical rates and the balancing hasn't even kicked in once... Screenshot of the typical voltage difference between cells below if it matter.

I'm assuming it was just the fact that the cells were brand new from manufacturer/never been discharged or charged and we're also 5 years old based on manufacturing date. Had to break them in or something?

Cheers!

Screenshot_2023-02-21-10-28-27-13_f89e2b5b019766820adc0c9448dea3af.jpg

amberwolf · Feb 21, 2023

Klauts said:
The cells charged at pretty sporadic rates the first charge and I had set the balancing activation at 3.6v initially for that reason. After a gentle discharge cycle and then putting it back on the charger the cells all charge at identical rates and the balancing hasn't even kicked in once... Screenshot of the typical voltage difference between cells below if it matter.

I'm assuming it was just the fact that the cells were brand new from manufacturer/never been discharged or charged and we're also 5 years old based on manufacturing date. Had to break them in or something?

As I'm pretty sure has been previously noted in one thread or another for this pack / project, if the cells didn't start out the same, they won't ever be the same; their properties and characteristics are different from each other and will always get worse over time, never better.

Balancing only makes them all the same voltage at that moment in time, it does not fix anything about the differences in the cells themselves. You'd have to replace all the cells with new identical cells (that all have the same characteristics), to do that.

Regarding balance current, that is supposed to be just the rate at which the BMS is able to drain the cells at during balancing. But you will have to verify this is actually what this means for this BMS, as being able to change this via programming means they must be able to adjust the resistances of each of the balancing circuits (assuming resistive bleeder balancing), and that is a much more complex design than most BMSs have (and more expensive to produce, and potentially more failure-prone). (adjusting the resistances could be as complex as actual voltage-controlled-resistors, or as simple as PWMing a single low resistance to allow more or less current thru, on average--but the typical BMS only turns the balancer on, or turns it off).

If it is just the balancing current, this affects how high you can allow the charge rate to continue to be while cells are being balanced (normally a BMS is so low in balance current that charging must basically stop entirely during balancing, so the BMS usually turns off the charge port FETs during balancing; this is unnecessary with a high enough balancing current ability, since the balancers on cells that don't need draining will just be "bypassed" or "shunted around" for the charge current, while charge current continues to flow thru the lower cells and keep charging them up).

But...any pack that *needs* balancing, especially one that needs high-current balancing (relative to it's capacity), has what amount to bad cells in it that will just get worse over time--cells are not matched and will never be the same, only growing more different over time.

Klauts · Feb 21, 2023

"As I'm pretty sure has been previously noted in one thread or another for this pack / project, if the cells didn't start out the same, they won't ever be the same; their properties and characteristics are different from each other and will always get worse over time, never better."

You are completely misunderstanding me... The cell Initially charged at different rates now they never differ more than 0.016v from eachother with the balancing turned off...

Even at 25 amp discharge they stay balanced. I was merely commenting on the oddity I noticed on the first initial charge.

I've since spoken to Samsung and apparently it is completely normal for cells to behave erratically on first charge after being unused 5 years passed manufacture and never processing an active load...

As I also noted previously all cells were at the same initial voltage, resistance and max capacity.

The only variable was charge rate which fixed itself after the first charge/discharge cycle. I was simply hoping for a more.elaborate explanation for the behavior than Samsungs.corporate mumbo jumbo

amberwolf · Feb 21, 2023

Oh, ok. Sorry about the misunderstanding. I must be remembering a different pack / project; I see so many here that they probably are all mixed up in my head at this point, and I am always so tired and dozing / waking that I confused things again.

Klauts · Feb 21, 2023

amberwolf said:
Oh, ok. Sorry about the misunderstanding. I must be remembering a different pack / project; I see so many here that they probably are all mixed up in my head at this point, and I am always so tired and dozing / waking that I confused things again.

No problem! Sorry if I sounded snippy

I'm going through some crazy life stuff. Breakup/moving cross country with me dog in 6 days. Didn't mean to be harsh, not normally so standoffish. You've been incredibly helpful seem like a genuinely kind human; I really feel bad about how I went about that...

Klauts · Feb 21, 2023

amberwolf said:
As I'm pretty sure has been previously noted in one thread or another for this pack / project, if the cells didn't start out the same, they won't ever be the same; their properties and characteristics are different from each other and will always get worse over time, never better.

Balancing only makes them all the same voltage at that moment in time, it does not fix anything about the differences in the cells themselves. You'd have to replace all the cells with new identical cells (that all have the same characteristics), to do that.

Regarding balance current, that is supposed to be just the rate at which the BMS is able to drain the cells at during balancing. But you will have to verify this is actually what this means for this BMS, as being able to change this via programming means they must be able to adjust the resistances of each of the balancing circuits (assuming resistive bleeder balancing), and that is a much more complex design than most BMSs have (and more expensive to produce, and potentially more failure-prone). (adjusting the resistances could be as complex as actual voltage-controlled-resistors, or as simple as PWMing a single low resistance to allow more or less current thru, on average--but the typical BMS only turns the balancer on, or turns it off).

If it is just the balancing current, this affects how high you can allow the charge rate to continue to be while cells are being balanced (normally a BMS is so low in balance current that charging must basically stop entirely during balancing, so the BMS usually turns off the charge port FETs during balancing; this is unnecessary with a high enough balancing current ability, since the balancers on cells that don't need draining will just be "bypassed" or "shunted around" for the charge current, while charge current continues to flow thru the lower cells and keep charging them up).

But...any pack that *needs* balancing, especially one that needs high-current balancing (relative to it's capacity), has what amount to bad cells in it that will just get worse over time--cells are not matched and will never be the same, only growing more different over time.

C$ 45.47 38%OFF | Daly Smart BMS LiFePo4 16S 48V BMS 3S 12V 8S 12S 24S 12V 24V 36V 60V 30A 60A 80A 100A 150A 200A 250A 300A 500A Li-Ion Battery

https://a.aliexpress.com/_mqGm4Ts

This is the BMS I'm using, I spent the money for one that was full featured and could handle my needs. You seem to know much more than me though so if you think this one is going to be prone to failure due to the advance balancing modulation I will consider an alternative.

The 40a Bluetooth version with 1A active balancing...

Klauts · Feb 21, 2023

amberwolf said:
As I'm pretty sure has been previously noted in one thread or another for this pack / project, if the cells didn't start out the same, they won't ever be the same; their properties and characteristics are different from each other and will always get worse over time, never better.

Balancing only makes them all the same voltage at that moment in time, it does not fix anything about the differences in the cells themselves. You'd have to replace all the cells with new identical cells (that all have the same characteristics), to do that.

Regarding balance current, that is supposed to be just the rate at which the BMS is able to drain the cells at during balancing. But you will have to verify this is actually what this means for this BMS, as being able to change this via programming means they must be able to adjust the resistances of each of the balancing circuits (assuming resistive bleeder balancing), and that is a much more complex design than most BMSs have (and more expensive to produce, and potentially more failure-prone). (adjusting the resistances could be as complex as actual voltage-controlled-resistors, or as simple as PWMing a single low resistance to allow more or less current thru, on average--but the typical BMS only turns the balancer on, or turns it off).

If it is just the balancing current, this affects how high you can allow the charge rate to continue to be while cells are being balanced (normally a BMS is so low in balance current that charging must basically stop entirely during balancing, so the BMS usually turns off the charge port FETs during balancing; this is unnecessary with a high enough balancing current ability, since the balancers on cells that don't need draining will just be "bypassed" or "shunted around" for the charge current, while charge current continues to flow thru the lower cells and keep charging them up).

But...any pack that *needs* balancing, especially one that needs high-current balancing (relative to it's capacity), has what amount to bad cells in it that will just get worse over time--cells are not matched and will never be the same, only growing more different over time.

I haven't been skimping on any components price wise as I've got a healthy fear of house fires (lived through 3 in 31 years....)

I got the highest quality BMS in my budget and tested each cell three times before I even thought about connecting them in series (parallels were already made...) Made sure all bus bars and connections were rated at least 3 times what my current and voltage would be at max, completely insulated all exposed metal. Melted tracks for the wires into the hardcases on each cell groups to prevent shorts and throughly Coated them in rubber.

Applied those fuses at the base of each balance lead and shrinked wrapped each layer of cells individually to prevent vibration and the hardcases rubbing on each other...

I feel like I've done my due diligence to build a safe pack even using old cells...

All my research shows that there shouldn't be any added risk to using 5 year old cells that have never been charged or discharged and we're stored in ideal conditions?

A 5%ish capacity loss is all I've seen and have been led to expect?

E-HP · Feb 22, 2023

Klauts said:
tested each cell three times before I even thought about connecting them in series (parallels were already made...)

What does this mean? Seems that if the parallel groups weren't deconstructed, then there's no way that you could have tested each cell three times. Do you mean you tested each parallel group three times? What was the nature of the test?

Klauts · Feb 22, 2023

I mean I deconstructed all series connections in the initial 2s10p packs so I was left with 14 1s10p packs and then tested the resting voltage and resistance of each 1s10p group.

Then I charged each parallel group and monitored and noted the charge rates of each. Initial charge rate was the only discrepancy I noticed between packs. I did a small load discharge to see if there was any strange voltage sag happening and then I constructed my 14s10p pack using the BMS mentioned above.

I then went to finish the full charge cycle as I'd only done a partial when they were still 1s10p packs and saw that they were still charging at slightly different rates but nothing too drastic.

Finished charge cycle did a full discharge on the bike at a 12 amp peak draw and then when I went to charge it again the discrepancy was gone.

All cells charge at equal rates +/- 0.016mv now and discharge equally as well.

I was hoping for an explanation of the physics or chemistry that might be causing the strange initial differences in charge rate.

Any harm in using a higher balancing current?

Klauts

10 W

amberwolf

Administrator

Klauts

10 W

amberwolf

Administrator

Klauts

10 W

Klauts

10 W

Klauts

10 W

E-HP

1 TW

Klauts

10 W

Similar threads