Cell Balancing 101

[Stoaty]

Is 4 hours prolonged? I don't want to "leave" my car sitting at 100% for any length of time.

It depends on the ambient temperature. If you are at 5 temperature bars, no worries. More--not sure. If you don't need the range (I don't) you can balance every 3-4 months. That is what I do (and I don't balance in the hot summer). Overly conservative??? Maybe, but given what has transpired so far with the battery pack, perhaps not.

 

[Nubo]

Worst-case the charger has a hard time recognizing the termination point and overcharges.

Any evidence of that? I find that doubtful, as voltage detection and termination is not all that difficult.

My understanding is that there are multiple termination criteria, and that trying to charge a charged cell can mask some of them -- for example, Delta-V. I'm sure Nissan has implemented enough criteria so that charging still stops before things get dangerous. But that doesn't mean the battery doesn't suffer for whatever short period before one of the criteria is finally met. As jkirkebo notes putting an an additional 3 GIDs into the battery. Evidence enough for me.

I didn't mean to suggest an overcharge to the point of "venting with flame", but disregarding the first termination is something I would call an overcharge nonetheless -- more energy went in.

 

[DaveEV]

Yes, the "balancing" happens after the LEAF's charge is complete.

Balancing should start near the end of the charge when one sees that the charge current starts dropping. It may start sooner - hard to know exactly. Typically charge current starts ramping down about 30 minutes before a 100% charge finishes. The maximum rate at which balancing occurs is slow, so balancing continuous after charging has stopped. Would be interesting to know what limits the BMS uses internally when balancing.

 

[sethr]

Thank you for talking about this... Nissan refused to answer my questions (two different reps.) as to whether cell balancing takes place during an 80% charge, saying "that is proprietary information". This really annoyed me, since I cannot imagine what negative outcome there could be for Nissan by answering the question. Just what car company do they think wants to know about their bottom of the heap battery management anyway? (Sorry, had to vent a bit.) I love my Leaf, but I am NOT loving Nissan.

 

[gbarry42]

My understanding is that there are multiple termination criteria, and that trying to charge a charged cell can mask some of them -- for example, Delta-V.

Delta-V is a technique used with NiMH chemistry. We're not talking about a Prius More to the point, there's a tendency to lump all battery chemistries together, so a particular idea may not apply.

Given the prevailing theory behind the LEAF cell balancing, the associated "bleed" circuitry can only handle small currents. From that we know that the balancing can only take place at the end of the charge when the charge current has dropped to a very low value.

When Tony says, "You'll get a "top off" charge" during the next 4 hours, it should be, "You may get a top off charge". I've done this multiple times and have never had the charge come back on. Most of my 100% charges have been the "sits all night afterwards" variety. Maybe the cells never get very far out of balance. They do try to match them at the factory, so I expect sometimes they'll get lucky.

 

[KJD]

When Tony says, "You'll get a "top off" charge" during the next 4 hours, it should be, "You may get a top off charge". I've done this multiple times and have never had the charge come back on. Most of my 100% charges have been the "sits all night afterwards" variety. Maybe the cells never get very far out of balance. They do try to match them at the factory, so I expect sometimes they'll get lucky.

The harder you work the batteries, the more likely they will drift out of balance. On the other hand if you keep the SOC between 20% and 80% that is good. If you drive like an old lady so the amp draw is really low that is also good for the battery and that helps keep them in balance.

On my car I only do the 100% charge once in a great while and the balance procedure only runs a short time ( 0 to 30 minutes typical )

As they say your mileage may vary

 

[Stoaty]

The harder you work the batteries, the more likely they will drift out of balance. On the other hand if you keep the SOC between 20% and 80% that is good. If you drive like an old lady so the amp draw is really low that is also good for the battery and that helps keep them in balance.

On my car I only do the 100% charge once in a great while and the balance procedure only runs a short time ( 0 to 30 minutes typical )

How can you tell that the balance procedure is running?

 

[KJD]

The harder you work the batteries, the more likely they will drift out of balance. On the other hand if you keep the SOC between 20% and 80% that is good. If you drive like an old lady so the amp draw is really low that is also good for the battery and that helps keep them in balance.

On my car I only do the 100% charge once in a great while and the balance procedure only runs a short time ( 0 to 30 minutes typical )

How can you tell that the balance procedure is running?

The amp draw is very low and all 3 dash lights are on solid ( not blinking )

 

[TomT]

That doesn't necessarily indicate that balancing is underway. Normal charging to 100 percent will give the same results toward the end of the charge regardless of any balancing.

How can you tell that the balance procedure is running?

The amp draw is very low and all 3 dash lights are on solid ( not blinking )

 

[DaveEV]

That doesn't necessarily indicate that balancing is underway. Normal charging to 100 percent will give the same results toward the end of the charge regardless of any balancing.

Yep. The BMS could start draining voltage from a cell at any time during the charge process should it decide to. L1, L2, QC - doesn't matter.

 

[smkettner]

I am not convinced you need 100% charge for balancing to take place.

 

[Stoaty]

I am not convinced you need 100% charge for balancing to take place.

Reference or reasoning?

 

[TonyWilliams]

I am not convinced you need 100% charge for balancing to take place.

I'm not convinced its not 100%

 

[garygid]

Since the determination of which cell-pairs are "higher" is probably a
millivolt comparison, it is unlikely that the "bleeding" for balancing
happens when the car is driving or actually during charging.

The process probably waits for there to be no current flow from the
Battery Pack, measures and compares the "resting" cell voltages,
makes the decision of which cells to "bleed", and go about doing it.

It is likely that the bleed current is less than an amp, so the time
needed for equalizing can be significant.

After the highest cells are "bled" down a bit, the car might try to
charge the car a bit more, to bring the whole pack up a bit, but
this probably only happens if the pack is down some significant
amount, but not when the pack is just down a little bit.

Since it is more work to program the car to equalize at two different
states of charge (100 and 80), and the voltage differences are more
detectable near "full" charge, and equalizing is not so important if
one only charges to 80%, my guess is that the car only equalizes
after a "full" charge.

 

[smkettner]

I am not convinced you need 100% charge for balancing to take place.

Reference or reasoning?

Because range varies significantly every 5 to 8 days. I have not logged it. Initial range seems to creep down and after several days it gets restored on a repetitive basis. Whether 80% or 100% the cycle seems to repete. Don't read too much into it. Just not sure you need 100%. I don't think there is documentation either way.

Lots of posts seem to take 100% as required as gospel. I say the 100% requirement is unproven.

 

[ebill3]

Well, as I see it, a big reason for cell balancing is to prevent a cell (actually, a pair of cells) from becoming over charged. With the LEAF that is over 4.1 volts.

IF that is the case, then balancing at 80% seems a waste of time, as it would seem very unlikely that a cell pair would exceed 4.1 volts when the charge is stopped at 80%.

 

[smkettner]

What is 80% voltage? I still think there is an incentive for Nissan to have all cells not too far above 80% long life mode.

 

[Herm]

What is 80% voltage?

4.05V per cell, IIRC.

 

[vegastar]

In my car the OCV of a 80% charge is usually 386V or 4.02 V/cell.

 

[garygid]

Equalization is not used to keep cell (pairs) from overcharging.

During charging, the highest voltage cell (pair) is normally used
to stop the charging, not the pack voltage. If the pack voltage
was used, it would be easy for one "high" cell (pair) to get
overcharged and damaged.

Instead, bleeding off the highest cell(s) allows the many lower-voltage
cells to continue charging (usually later).