Battery Balancing ... Do it - Don't do it ... who's right?

[Hangman]

I am so confused!  As I am waiting patiently for enough money to buy a new EUC, I have been reading posts & web sites, white papers & official documents regarding batteries & balancing.

There are numerous electric scooters on the market now & apparently most of them have some technology built into their system to automatically balance the battery while charging.  They are stating in their User Manuals (you can download the pdf versions from numerous sites) & technical white papers that you do not want to leave it on the charger after it is completed as it degrades the battery & does nothing for balancing as balancing is done while charging.

Then I read of numerous posts & web sites, youtube videos & other listings that we should leave the charger on the EUC for several hours after the charger has completed to balance the batteries.

InMotion makes both EUC's & Scooters.  KingSong makes both EUC's & Scooters.  The list goes on ...

My question(s) here is if these manufacturers make both, why aren't they putting the technology in the EUC that they have in the Scooter so the charger doesn't have to sit for 'hours' after a full charge to balance the battery?

Or perhaps they are putting the technology in the EUC's & these guys who keep posting to leave the charger on the wheel after it's complete are stuck in the old days with unbalanced wheels?  You know, like dinosaurs & fossil fuels.

Albeit I cannot find any documentation of such a 'balancing while charging' with an EUC.

Why?

Is the technology so drastically different that they can't put a balancing effect in the EUC system while it's charging?

[Dave Wood]

i personally try not to leave charger on after its turned green, i figure the green is there for a reason.. i have experience with balance chargers from my diy eboarding days and they always do the  balancing towards the end of the charge not after, i know we don't use balance chargers with our eucs bur im presuming our bms's act in a similar way during the balancing process  so that's how iv always charged my eucs, charge until the green light shows and switch off iv never had issues with unbalanced cells or degrading cells

Edited February 27, 2021 by Dave Wood

[Rich Sam]

This topic talks about this subject to great lengths, my takeaway from reading it was, EUCs do not have smart battery circuitry to be able to balance the individual cells until the end of the charge phase.   This is something the community really wants, but no wheel to my knowledge has that.

 

[Chriull]

2 hours ago, Hangman said:

There are numerous electric scooters on the market now & apparently most of them have some technology built into their system to automatically balance the battery while charging. 

As our EUC have. The passive balancing used in the BMS happens more or less during the constant voltage phase.

Quote

They are stating in their User Manuals (you can download the pdf versions from numerous sites) & technical white papers that you do not want to leave it on the charger after it is completed as it degrades the battery

That's called afaik trickle charge and li ion cells do not want this at all!

Quote

 & does nothing for balancing as balancing is done while charging.

There still (can) happen balancing if one leaves the charger on the battery, but one stresses the battery. So i personally would prefer to repeat some constant voltage charging cycles for "intensified" balancing compared to "trickle charging"

Quote

Then I read of numerous posts & web sites, youtube videos & other listings that we should leave the charger on the EUC for several hours after the charger has completed to balance the batteries.

Yes, that's quite often stated. As written above it continues balancing while stressing the cells. So _I_ would not recommend it.

Full charging until the charging current drops down to the manufacturer recommended threshold current of about 50-60mA per paralleled cell should be perfect and enough for balancing. Without unnecessary cell torture.

If one wants/needs to intesify balancing one can make a short trip after charging (to ~90-95% charge shown after a ~20 min rest) and charge full again. Thats the above mentioned repeated CV charging.

Quote

y question(s) here is if these manufacturers make both, why aren't they putting the technology in the EUC that they have in the Scooter so the charger doesn't have to sit for 'hours' after a full charge to balance the battery?

I don't have any specific detail or knowledge of whats used for scooters, but i'd profoundly guess it's the same technology and there is no difference. Just the users believes and preferences.

And - very important - afaik scooters use less cells in series and thereby are less prone to cell misbalance. The more cells are in series the more unstable the system gets!

Quote

Albeit I cannot find any documentation of such a 'balancing while charging' with an EUC.

Take a look at 

There i tried to collect the battery related knowledge with many links to related discussions.

Edited February 27, 2021 by Chriull

[mhpr262]

I have never believed those stories about EUC batteries beginning to balance only after you have finished charging. There is no technical reason for that whatsoever. I have decades of experience with multi-cell  lithium batteries and chargers (from R/C sports) and they ALL get balanced while charging. The only case where it would perhaps make sense to leave the charger attached is if you have a battery pack with massively unbalanced cells.

That said it doesnt harm the battery either if you keep the charger attached after the battery has finished charging. Li-ion batteries are charged with the CC/CV principle, that means that after a phase where the current is restricted to avoid  too much current flowing through the battery the charger switches to providing the max voltage of the battery pack and then just lets the pack soak up as much current as it wants. As the battery keeps getting fuller the voltage difference dwindles and so does the charging current, until it naturally stops charging when the battery pack has reached the same voltage as the charger. There is no such thing as "trickle charging" with a li-ion charger.

It works like filling a container of water by connecting it with a tube to another container of water in which the water is always kept at the same level. The water level in the second container can never rise higher than to the level in the first container, no matter how thick you make the connecting tube or how often you turn the tap on and off.

What does degrade the battery though is being kept at 100% state of charge for a long time, but that has nothing to do with the charger staying connected to the battery. the battery pack will degrade just the same even if the charger is not connected to it.

Edited February 28, 2021 by mhpr262

[Seba]

46 minutes ago, mhpr262 said:

I have never believed those stories about EUC batteries beginning to balance only after you have finished charging. There is no technical reason for that whatsoever. I have decades of experience with multi-cell  lithium batteries and chargers (from R/C sports) and they ALL get balanced while charging. The only case where it would perhaps make sense to leave the charger attached is if you have a battery pack with massively unbalanced cells.

Trust comes from knowledge  As your knowledge comes from R/C smart chargers like Graupner or Redox, you may think that all balancers work in the same way. No, they don't. R/C charger measures voltage on each cell, so when it detects that there is some imbalance between cells, will stop charging and start discharging selected cells to match the voltage of the least charged cell (cell with lowest voltage). Then it will resume charging and entire process can be repeated as often as needed. This way battery can be charger partially and still get balanced during entire charging process, regardless of state of charge.

But most balancing circuits built into BMSes used in our EUC batteries are completely different. In fact, calling these a BMS is an exxageration. They are just protection circuits. There are not smart and in fact, they don't manage entire battery (that's why they shouldn't be called BMS). They don't measure voltages on each cell group. Instead, each paralell cell group have its own, independent circuit called voltage comparator that checks if cell voltage is at or above 4.22~4.25 V. Only two possible states - voltage above or below predefined level. If above, it will connect a small resistor (usually 100 ohm) that will bleed excessive charge from this cell group and convert it to heat. And that's all. Nothing less, nothing more. There is no smart algorithm that manages entire battery, there is no balancing during charging. But to make this working, cell voltage must reach 4.2V and you get it only when battery is fully charged.

13 hours ago, Dave Wood said:

i personally try not to leave charger on after its turned green, i figure the green is there for a reason..

Green LED goes on when charging currents drops below certain limit - something between 100 - 300 mA. Usually this signals that charging is (almost) complete, but in fact it will also signal any drop in charging current. This is why this LED goes green when you just unplug charger from wheel. More, you'll see green LED when battery protection circuit will disconnect charging ciurcuit after sensing single-cell overcharge, but entire battery will be far from full charge.

[Chriull]

3 hours ago, mhpr262 said:

There is no such thing as "trickle charging" with a li-ion charger.

Batteryuniversity.com (and afaik many others) have a different knowledge:

"Li-ion cannot absorb overcharge. When fully charged, the charge current must be cut off. A continuous trickle charge would cause plating of metallic lithium and compromise safety. To minimize stress, keep the lithium-ion battery at the peak cut-off as short as possible."

From https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries

[.....]

20 minutes ago, Chriull said:

Batteryuniversity.com (and afaik many others) have a different knowledge:

"Li-ion cannot absorb overcharge. When fully charged, the charge current must be cut off. A continuous trickle charge would cause plating of metallic lithium and compromise safety. To minimize stress, keep the lithium-ion battery at the peak cut-off as short as possible."

From https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries

This seems to agree with the quoted statement of @mhpr262. Good information here. I dont see them addressing the issue of charging in 'packs' tho. I guess the problem arises when the charger can't 'see' the voltage of a cell, as there are many cells in a pack. THIS may be where the imbalance comes from? After months of reading this stuff, it's just so hard to comprehend exactly. Perhaps this is because our PACKS of batteries are all hidden behind a bms. Even tho it isnt a 'smart' bms, does it not change the method or safeties a little bit? Truth be told, I'd have been much happier if I never began to even wonder about it... If i fully charge and 'balance' my packs, does it not drop a few voltage points at rest? If I dont keep topping it off, wouldnt that mean it is resting at a voltage lower than 100% and thus I neednt fear of damage from storage?

Edited February 28, 2021 by ShanesPlanet

[Chriull]

1 hour ago, ShanesPlanet said:

This seems to agree with the quoted statement of @mhpr262.

yes, just looking at the quote - seems i quoted inaccurately . The sentence before missing:

 

5 hours ago, mhpr262 said:

As the battery keeps getting fuller the voltage difference dwindles and so does the charging current, until it naturally stops charging when the battery pack has reached the same voltage as the charger.

As i understand this trickle charge should not be a problem as it cannot exist? Which is wrong.

Quote

Good information here. I dont see them addressing the issue of charging in 'packs' tho.

No - this statement is for one cell.

If all cells are perfectly balanced pack charging has the same effect fir each cell.

In our BMS we have cell overvoltage protection - so during cv/saturation phase or trickle charge the bms will cut off once due to imbalance one cell reaches too high voltages.

Within this "cell overvoltage protection" threshold some cells are charged to a bit higher voltages, some a bit lower voltages. So some are during a trickle charge more stressed and some a bit less.

3 hours ago, Seba said:

But most balancing circuits built into BMSes used in our EUC batteries are completely different. In fact, calling these a BMS is an exxageration. They are just protection circuits. There are not smart and in fact, they don't manage entire battery (that's why they shouldn't be called BMS).

They are not smart, but reducing them to solely protection is not true, too.

Passive balancing is not much of "intelligence" but a little bit more than simple protection!

Quote

They don't measure voltages on each cell group. Instead, each paralell cell group have its own, independent circuit called voltage comparator that checks if cell voltage is at or above 4.22~4.25 V.

The idea of passive balancing is that the shunt resistor is activated at 4.2V - at full charge voltage. So that after the constant current phase the the first cell group to reach this 4.2V during constant voltage phase gets a bit of their charging current taken away.

From most bms/datasheets i have seen till now the threshold for the balancing/shunt tesistors is ~4.2V.

Problem seems to be here the tolerances of widely used comparator circuits - mostly +/-0.05V which would give 4.15-4.25V as range...:( Some bms datasheets specify only +0.05V as tolerance, but i suppose the chinese reseller just forgot the "/-"?

On the other side ic's like the LTC6804 has total cell voltage mesuarment error of less than 1.2V. And (some of) these kind of ic's have programmable thresholds - or are "just" measurement and switching devices and the algorithm can be self implemented.

Quote

And that's all. Nothing less, nothing more. There is no smart algorithm that manages entire battery, there is no balancing during charging. But to make this working, cell voltage must reach 4.2V and you get it only when battery is fully charged.

Considering tolerances, you are partly right. But looking at the other end of the tolerances some bms have to start during constant current stage...

And looking in detail at this, admittetly not very smart algorithm, if _all_ cells are at or below 4.2V no balancing has to and will happen as the cells are balanced!

Only cells with higher voltages need balancing...

Considering the charger voltage tolerances and misadjustments happening over time, the comparator tolerances and their till now not mentioned temperature drift i assume you are absolutely right that in (too) many cases no balancing can happen (or just once it's way to late).

But from design, if peoperly implemented it's a nice and simple measure for our packs.

 

Edit: btw - maybe @RagingGrandpa still has the ild gw bms and/or some detail photos were one can identify the used conparators so one could look up threshold voltages, hysteresus and tolerances?!

 

[Nostris]

So, everyone has their own theory, and still we have no definitive answer. The people who make these things should come forward and let us know. So speak up INMOTION, Kingsong, Gotway and Veteran.....we are all ears!

[pico]

Well, well, well...

The manufacturers...

Think about it. The BIGGEST element of uncertainty is Cell matching.

In an Ideal world all the cell would be closely matched and would age together gracefully with no special care, like a BIG single cell. And the packs would last a long time. 500 complete cycles nominally is a LOT of kms! (over 15000kms for a meager V8F!). 

The built-in safeguards are already there. (no under voltage, no over voltage, no over current, primitive balancing)

Unfortunately with the consumer products we are dealing with, a lot of those manufacturers don't seem to control the batch number of the cells that are used.

To make things worse we have (a crime IMHO) resellers, (and hobbyists) that join in with (homebrewed) add-on packs!

This is what I think is the origin of such diverse opinions and experiences...

Some interesting reading for us nerds...

https://www.sciencedirect.com/science/article/abs/pii/S0378775313019447

 

[RagingGrandpa]

I'm having trouble identifying the actual question being asked in this thread...

If it's "how does cell balancing work in my <gotway> EUC?" - we've covered that in detail, with definitive conclusions.
 

On 2/27/2021 at 11:41 AM, Hangman said:

I am so confused! 

if these manufacturers make both, why aren't they putting the technology in the EUC that they have in the Scooter so the charger doesn't have to sit for 'hours' after a full charge to balance the battery?

It seems like you're requesting design feedback from the manufacturers directly.
Sorry to say- it's very unlikely that you'll encounter actual engineers from Gotway/Inmo/Kingsong in this forum.

Edited March 1, 2021 by RagingGrandpa

[Chriull]

15 minutes ago, RagingGrandpa said:

that in detail, with

Ah - just seen that you nicely listed the used comparators in your first post!

HY2213-BB3A ASIC for balancing:

detect 4.2V+/-0.025V, release 4.190+/-0.035V

HY2113-KB5B ASIC for cell overvoltage charge cutoff:

Detect 4.25V+/-0.025V, release 4.19V +/-0.05V

[alcatraz]

If you buy your euc new or nearly new you will have the advantage of experiencing how the battery pack ages.

You'll get an idea how long it should take to charge. How long it takes for one bar to disappear from the meter. How many km you get on one charge. You want these variables to change very very gradually over the years. If you notice a larger change over a relatively short period of time, then that's a red flag.

If you want to keep using a red flag battery it's wise to open it up and measure the voltages. Only then can you be 99% sure not to burn down your home.

[Planemo]

On 2/28/2021 at 8:06 AM, Seba said:

There is no smart algorithm that manages entire battery, there is no balancing during charging. But to make this working, cell voltage must reach 4.2V and you get it only when battery is fully charged.

This is one of the best quotes I have seen, and worth knowing. This is what makes EUC charging completely different to say RC lipo charging.

In answer to @mhpr262 the fundamental problem we have is 'parallel cells'. This isn't a problem in RC, as nothing is paralleled. In RC pretty much everything is done in serial, whether that be 2, 3, 4, 5 or 6 cells. All that changes is voltage. So each cell in a battery pack can get it's very own charge/sensing circuit, and as a result, far better control over charging.

EUC's use parallel (and serial) cells so there are always multiple cells hanging off one charge wire or 'string'. So we never know the individual cell data (nor can we charge individual cells), we can only deal with whats on each string. If we could do away with parallel cells, we could employ truly smart BMS's and chargers, very much like the RC world. The problem is that we need parallel to get the current requirements for EUC's...

The answer? We need big cells capable of big current outputs. In the case of the MSX (6P - 2 battery packs of 3P each) with each Sanyo cell being good for 10A/3500Mah, we would need single cells capable of 30A discharge and a capacity of 10500Mah. Each pack could then still be 24S (for 100v) with a smart BMS and full control over each individual cell. Run 2 packs (each pack with it's own 24S BMS, or even 1x 48S BMS) to make it capable of 60A and voila. The issue is that AFAIK these magical cells aren't available, or at least I'm not aware of them. The nearest I have seen is the Boston Swing 5300mah 'twin' cells but even they are only rated at 13A.

That said, we could still do better even with parallel cells if the manufacturers at least gave up on passive charging. A smart BMS could still balance throughout charging, even if it did 3 cells at a time, which is better than what we have now where everything tries to balance at the very end, not always successfully as you might imagine.

I do wonder how Tesla have engineered it. They used to (maybe still do) use huge quantities of 18650 cells so they must be paralleling them as well but I have no doubt that their cell management is far superior to anything we have seen in EUC's.

[mhpr262]

I cant wait for the Tesla 4680 cells to become publicly available, those puppies can deliver six times the power of a standard 18650 cell. No more parallel cells will be necessary then.

[Planemo]

Yeah and at 46mm diameter we could get them to fit in our wheels as well! It would be awesome.