Parallel Nissan Leaf Modules

[cricketo]

Originally Posted by izzybird

no one should put all their faith in a BMS to make sure there is no overcharge state.

What should we put the faith into then ? :)

[izzybird]

Originally Posted by cricketo

What should we put the faith into then ? :)

Well, like most dribble on the interweb it’s just a opinion. I just think it’s a safe precaution to physically observe voltage of the cells and/or have a few fail safes built in to make sure you don’t get over 4.2v. You know, spread a little faith.

[cricketo]

Originally Posted by rjz

I contacted the vendor (techdirect) when 3 of modules began to swell. I think I received bad batteries. The vendor is indicating the issue is my pack configuration. I have my modules in one set, with every 2 modules paralleled. The vendor indicated that format is not safe and that I should assemble two 5 packs and hook them up in parallel. This seems to me more likely to cause balancing issues and I am unclear why it would be necessary. Can someone let me know if they are correct or if I am overlooking something?

I've dealt with TechDirect multiple times. Their support isn't all that great, probably because they aren't really a vendor, just a reseller of salvaged goods and some custom stuff.


Paralleling the modules is fine as long as you also parallel the center taps. Swelling can happen with old modules, and can also happen due to overcharging. One concern with paralleled modules is balancing - BMS boards typically use fairly small current to balance the modules, and having very large capacity (132Ah) can cause for BMS to be unable to keep up with the charger.

I went back to see if you gave a description of your setup, but there wasn't much. So how are you protecting, charging, balancing the modules ?

[rjz]

Thanks for the feedback. The swollen packs were taken out of service immediately. All batteries were pre-balanced and I had checked the individual voltages and pretty quickly saw that some were falling out of balance. I was taking the pack apart to re-balance the order of the cells to see if it would improve the behavior. I am fairly convinced the modules had already been over discharged before I received them. The pack is dissembled so nothing to see currently, but it was bus barred and compressed using nissan plates and threaded rod as I have seen in many builds on this site. The cart is un-upgraded 98 ezgo txt. Motor has been replaced, but it was just a stock replacement. Wires swapped out for 2 gauge and rotary FW changed to a reversing contactor.

[Sergio]

A BMS can only balance the cells at the Top so if You measure the voltage after use the only way the cells will stay balanced is if they are exactly the same capacity.

You should typically allow enough headroom above the "charging voltage" before the "max voltage" for the BMS and shunts to do their work.

It also helps the longevity of the pack (charging to 85%) since the cells are under the most stress when fully charged.

[cricketo]

Originally Posted by Sergio

A BMS can only balance the cells at the Top

Depends on the BMS. I use TinyBMS for my builds, and it has a programmable threshold.

cells will stay balanced is if they are exactly the same capacity.

If I'm not mistaken, variance in internal resistance will also result in cells going out of balance.

It also helps the longevity of the pack (charging to 85%) since the cells are under the most stress when fully charged.

That's extremely conservative. Additionally, how are you determining 85% without tracking pack's state of charge ?

[Sergio]

Good questions, I will add more detail in case other members have similar questions.

1 & 2) The concept of "balancing" is to bring cells to the same known and returnable SOC level.

The analogy is that the cells are like water cylinders of the same height but different diameters and therefore hold different amounts of water.

If You stack them side by side and remove 1/2 of the water from each, the water will be at the same level in all of them, but the amount of water you removed from each cylinder was different.

Lithium cells connected in Series have the same amount of energy removed from each of them.

In the water cylinder example above if some of the cylinders have 60 gallons capacity and some 40 gallons capacity, removing 20 gallons from each of them will result on the 40 gallon cylinders being at 50% level/SOC and the 60 gallon cylinders being at the 66% level/SOC.

If you then add 20 gallons back to each cylinder (charging), they all go back/return to the 100% level/SOC.

That is why it only makes sense to use a BMS to top balance since we know if the cells have different capacities their SOC and voltage will differ as You discharge them in Series.

Some folks prefer to bottom balance the pack or bring the water level to near empty 0%level/SOC and not use a BMS at all.

As you add 40 gallows to each cylinder, You will fill up the 40 gallons but only reach the 66% level/SOC of the 60 gallons cylinder.

This is again expected since all you are concerned is that when You again remove 40 gallons they will all return to 0% level at the same time.

Differences in manufacturing or changes in internal resistance as cells age take a long time to develop and it is only a concern for folks that bottom balance their pack since a BMS is always top balancing and easily corrects those differences.

Sometimes the differences the BMS is correcting are introduced by the BMS setup itself due to differences in the ADC converters or wiring and connections resistance.

If You have positive experience doing some kind of "Mid Balancing", please share with us.


3) As far as increasing the longevity of Lithium packs by not charging them to 100% SOC, that is a very well documented fact due to stress caused to the cells.

Just look at this one article for example, Table 4 indicates how much longer You can expect your pack to last if You don't charge to 100%.

https://batteryuniversity.com/learn/...ased_batteries

You can also look at pretty much all electric cars on the Market and their "Daily charge" vs "Long Range" charging recommendations.

[NCPW]

Sergio;

Very good analogy and I think that comment should be a sticky in itself under The 'All Things Lithium' subforum. This should help anyone wanting to convert to lithium understand the importance of using a BMS and the reason behind it.

[cricketo]

Originally Posted by Sergio

1 & 2) The concept of "balancing" is to bring cells to the same known and returnable SOC level.

That is not quite correct. Balancing brings the cells to the same voltage, not SOC. The concern is with some cells going over
their lower or upper voltage limits during charge or discharge. There is some relationship between SOC and voltage, but given
it's not direct, no manufacturer states anything other than voltage limits in the specs.

Lithium cells connected in Series have the same amount of energy removed from each of them.

Again, what I stated above about internal resistance - cells with lower internal resistance will provide more current (energy) than
the ones with higher internal resistance. Battery packs produced by large vendors will have closely matched cells, but even then
over time they will develop variances. Salvaged cells are even more likely to be mismatched if they don't come as original packs.

That is why it only makes sense to use a BMS to top balance since we know if the cells have different capacities their SOC and voltage will differ as You discharge them in Series.

And that's the part I highlighted not to be true at all. You can balance the cells as soon as you start charging (let's say from 20% SOC). Depending on the amount of current the BMS can bleed via shunts, your charging current and the variance in voltages between the cells one may need to start balancing sooner or later, at the risk not being able to charge all the cells to the same desired level.

3) As far as increasing the longevity of Lithium packs by not charging them to 100% SOC, that is a very well documented fact due to stress caused to the cells.

I did not suggest charging to 100%. I suggested that charging to 85% is very conservative (as opposed to 90-95). Additionally you still didn't elaborate on how you're estimating 85%, since relationship between SOC and voltage isn't all that direct for most Lithium Ion chemistries.

If You have positive experience doing some kind of "Mid Balancing", please share with us.

Positive as in whether I haven't set my house on fire ? :D Like I mentioned earlier, BMS I use is programmable. Quote from the docs :

Currently Tiny BMS device supports only generic CC/CV chargers, which is connected directly to Tiny BMS
charging port. Tiny BMS controls the charging process only by turning on or off charging port (internal
CFET or external relay / contactor). During charging process cells groups connected in series are
balanced. Tiny BMS charging and cells balancing process starts when charger is detected, Charger
Connected (0x64) event is generated, charger switch is turned on and BMS state becomes Charging.
During battery charging process, if at least one cell voltage reaches Fully Charged Voltage, Charger Switch
(Dual Port mode) or Single Port Switch (Single Port mode) is turned off (battery charging is stopped) and
only cell balancing is performed. When the cells imbalance becomes less than the Allowed Disbalance
setting, the Charger Switch (Dual Port mode) or Single Switch (Single Port mode) is restarted and the
battery charging continues. This cycle continues until the battery fully charged conditions are met.

[Sergio]

cricketo, it seems like you are more interested in arguing nuances than contributing to the site.

Mid-balancing to me just seems like a waste of time even if you are absolutely sure the cells will behave exactly the same since different capacity cells will change in voltage according to their own SOC.

If the cells are wired in series they cannot have different current going through them, that can only happen in parallel arrangements.

If You look at the table in the link I provided, the voltages recommended are shown and there are many sources like evbattercenter that have plots of measured SOC vs cell voltage for the Leaf cells.

Do you have any recommendations for members on the best way to configure their BMS?