Best and safe battery maintenance/care/charging practices

[Critzlez]

From my understanding, when it turns green it's always going to be at the same level of balance and that level is adequate , but if you want it to be more balanced you can charge it for longer after green but it isn't necessary.

I'd want someone to double check if my assumption is correct

Edited July 13, 2023 by Critzlez

[Chriull]

2 hours ago, Critzlez said:

when it turns green it's always going to be at the same level of balance

The led turns green once charging current dropped below some certain, often adjustable threshold. According to manufacturer specifications this should be roughly about 50 mA (per paralleled cell).

This has just some very minor correlation with level of balancing. But at thus stage together with some hour(s) after the charger is unplugged the battery is about as balanced as it is possible for the bms. If thisi s not enough one has to cycle this once again - but one already reached the limit if the bms, so checking if the cells are good enough or should be replaced could be recommendable.

[Critzlez]

2 hours ago, Chriull said:

This has just some very minor correlation with level of balancing

Could you explain more about what you mean by this?

Edited July 13, 2023 by Critzlez

[mrelwood]

On 7/11/2023 at 11:33 PM, Funky said:

But what about the beeps?

KS probably just measures the charging current, and beeps and cuts off the charge with a very different threshold than where the charger goes green.

 

On 7/11/2023 at 11:56 PM, The Brahan Seer said:

That quote is copy and pasted straight from the link at Battery University. 

… which doesn’t always give information related to EUC BMS features.

On 7/12/2023 at 12:07 AM, The Brahan Seer said:

@mrelwood  Different wheels use different systems.

Wheels’ passive top balancing systems are nearly always very similar.

On 7/12/2023 at 12:07 AM, The Brahan Seer said:

Some use proper BMS others passive balancing, many none at all*.

There are only a few wheels that might not have a balancing feature. All others do.

On 7/12/2023 at 12:07 AM, The Brahan Seer said:

How many people have problems with failing cells and hence balancing under normal use?

Quite a few actually, especially with the latest wheels. I suspect second grade cells or bad balance from the factory.

 

23 hours ago, Critzlez said:

Yeah from my understanding, it should be fine to not balance an extra few hours after green if it reaches the max voltage when it changes to green ,right?

No, voltage is not a sufficient meter for the state of charge, let alone balance. I suggest reading about core and surface voltages at the beginning of this thread.

 

On 7/12/2023 at 1:44 AM, The Brahan Seer said:

100% SOC is when the battery is 100% full.

You might still be thinking of a battery as a gas tank. When is a glass 100% full? When it reaches the advertised capacity? When water level reaches the edge? When the surface tension can barely hold the excess water in? All different amounts. “100% full” is often an inprecise statement.

Batteries can be charged to 150% voltage. (Many would fiercely burn though.) Even the max charging voltage in the battery datasheets is sometimes 4.25V. And when the charging voltage is kept there, the SoC still grows. We’ve just decided to stop charging Li-ions around 4.2V, not because it wouldn’t hold a higher charge level.

And KS 100% is different from LeaperKim’s 100%. Percentage doesn’t represent even the exact voltage, let alone SoC.

 

On 7/12/2023 at 1:44 AM, The Brahan Seer said:

So you are saying EUC's never utilise their whole capacity?

Absolutely they don’t. The end of the capacity already has so little power that it’s insufficient for self-balancing vehicles.

On 7/12/2023 at 1:44 AM, The Brahan Seer said:

  So  it stops it reaching this figure using bleed resistors?

The bleed resistors act very slowly, slower than the charge current. They only slightly slow down charging, and need hours of time to bleed the excess after the charger is unplugged.

On 7/12/2023 at 1:44 AM, The Brahan Seer said:

So balancing extra is only required if you don't hit 83.8V or above by the time the red light goes green?

Balancing is not directly related to the pack’s maximum voltage. Balance is what happens within that voltage.

On 7/12/2023 at 1:44 AM, The Brahan Seer said:

charging within the guidelines

Whose guidelines? Because they vary tremendously.

 

11 hours ago, Critzlez said:

Could you explain more about what you mean by this?

The charger turns green when the charging current gets lower. That’s all. It’s not directly related to balancing. Only if the pack is balanced very badly, will it show up in the pack’s charging behavior.

A 84.0V pack has 20 cell groups. To reach a total of 84.0V, some can be 4.2V, some 4.15, some 4.25V. The chargers don’t know this, and the wheel would still seemingly charge up to full normally.

[Critzlez]

1 hour ago, mrelwood said:

The charger turns green when the charging current gets lower. That’s all. It’s not directly related to balancing. Only if the pack is balanced very badly, will it show up in the pack’s charging behavior.

A 84.0V pack has 20 cell groups. To reach a total of 84.0V, some can be 4.2V, some 4.15, some 4.25V. The chargers don’t know this, and the wheel would still seemingly charge up to full normally.

Would it be detrimental to the battery pack to stop charging immediately after green for its entire lifetime? Or a better way to word it, at what current is fine to stop charging during the CV phase?

Edited July 14, 2023 by Critzlez

[mrelwood]

4 hours ago, Critzlez said:

at what current is fine to stop charging during the CV phase?

There are differing views on this. Some battery cell manufacturers say 50mA, which would be a total of 200mA for most 1500-2500Wh wheels. This doesn’t however take balancing a large series cell pack into consideration, which is why I think it is better to charge further in the long run. Or at least every few charges. When I monitored the charging current (between the charger and the wheel), I usually went down to 10-50mA (total). That system never went below 9mA though, so I do recognize that the low end of my range is very low.

[rcgldr]

Stock charger for V8F (20S2P pack) changes the LED to green at around 400 mA, but continues to charge. Soon after this (about 8 minutes on my V8F), one of the cell pairs will reach an overcharge voltage, and the V8F will halt the charge, and the V8F will slowly discharge over the next 8 hours or so until it drops below about 82.6 volts, at which point it re-enables charging. Doing a short ride after V8F has shut down charging will also re-enable charging.

The eWheels charger changes the LED to green at 300 mA, after which it shuts off charging, which in turn causes the V8F to shut down rather than slowly discharge as it does with a stock charger. If I charge with stock charger until V8F stops charging, then switch to eWheels charger, it will supply some voltage at a very small current (display shows 0.00 amps), which I assume is due to V8F sensing of voltage while it slowly discharges the battery pack.

If the 1k ohm resistors on the V8F BMS are used as bleeding resistors, that represents 4.2 mA of current, but I doubt the charger current ever gets down that low, which explains why my V8F ends up detecting overcharge and shuts off charging.  If a charger only gets down to about 200 mA current, then the bleeding resistors only reduce this to 195.8 mA, so hardly any difference than the cell groups charging at 200 mA. It's possible that the V8F does some balancing during the very slow discharge after it shuts off charging, but doesn't seem likely and I have no way to test this.

Based on what I've read, a typical e-bike BMS will have true passive cell balancing with bleeding resistors chosen so discharge current is greater than charger current when in constant voltage mode. In this case, any groups of cells that have reached the maximum voltage will be discharged while the rest of the pack continues to charge. Eventually all cells will reach a set voltage and all of the bleeding resistors will be enabled. This is what is shown in Wrong Ways' video about the lack of passive cell balancing in some Inmotion EUCs.

[skunkmonkey]

I really wish that EUC companies would step up their game when it comes to this sort of thing. It's a major safety issue. It doesn't help that their manuals give different advice. For instance, my InMotion V11 manual explicitly says not to let the charger continue for more than an hour or two after reaching green or you risk a fire. There's no mention of the rest period indicated in this thread. If such a rest period is needed, then the companies really should have a tri-state light. Red for charging, Yellow for balancing, Green for unplug.

[mrelwood]

4 hours ago, skunkmonkey said:

I really wish that EUC companies would step up their game when it comes to this sort of thing.

I absolutely agree. Unfortunately it seems that Li-ion battery practices aren’t well known (or at least not shared) in other industries either. For example laptop manufacturers don’t tell you that having the laptop continuously in the charger will completely kill the battery in a year or two.

 The information we’ve been able to gather as a wonderful EUC community that we’re are is extremely valuable to the users, and I don’t think the EUC manufacturers have any similar sources for the information that we do. But it seems that they can’t directly put the community tips in their manuals either, since it’s not necessarily a solid enough source for them. That’s my guess.

[rcgldr]

4 hours ago, skunkmonkey said:

my InMotion V11 manual explicitly says not to let the charger continue for more than an hour or two after reaching green or you risk a fire.

In case some missed this in my prior comment, the eWheels rapid charger shuts off when the LED goes green. In the case of my V8F, this occurs when charging current drops below .300 amps (300 ma), and after the charger shuts off, the V8F also shuts off.

Even with the stock charger, which does not shut off, the V8F will stop charging once any cell pair reaches an overvoltage state. It does detect voltage from the stock charger and stays awake, but draws less than .001 amps to sense voltage, while the battery pack slowly drains down to about 82.5 volts in about 8 hours before it starts charging again, only to repeat the cycle: charge for a few minutes, detect overvoltage, shut off charging for 8 hours or so.

Edited August 31, 2023 by rcgldr

[alcatraz]

2-3 year old wheels aren't thrown away. They're sold. Even 5y old wheels, soon 10y old wheels.

My point is that all this tinkering counting hours on the charger, and following the manual is not going to guarantee much. People need to ultimately check if cells are balanced and for that they need to open their packs up. Simple as that. Then they can rest assured.

Then it doesn't matter that the wheel is 10y old. A 2y old wheel can catch fire by using the manual instructions. You just have to be unlucky to have a bms malfunction and deterorating cells.

The easiest thing we can do is to ensure we can charge to the pack's end voltage. There's one drawback though. This method requires overvoltage protection to work. If it doesn't, the pack will probably still catch fire.

If it's time consuming to open the pack, then install balance wires and check that it's balanced every year.

Or get a wheel with a smart bms.

[Mike.P]

On 7/11/2023 at 11:33 PM, Funky said:

But what about the beeps? They happen normally 1-3 hours after charger has turned green. (More ridden, over multiples days not charging. It takes more time to get the beeps to happen. VS when i have done 40km in on day and charge again.) My guess more unbalanced batteries - takes more time to get them balanced and the beeps.

There are something happening on mainboard.. Otherwise there would be no beeps. And they only happen when meter drops to 2.3Watts from wall. Maybe they simply happen when wheel doesn't get any charge incoming? Or drop to low.

 

As a new member to the forum i would like to say hi to everyone

I bought  a new KS 18 xl in August, i have done around 5k khm as a new rider and i really love it

I have been charging this wheel since day one with a shelly plug and euc world because you can set your own charging settings

my wheel is fully charged at 84.00v when it reaches 84.1 the overvoltage/overcharge circuit is enabled and my charger is pulling 0 amps and 1.6 watts at this time

after one hour or something my wheel is back to 84.00v and beeps 5 times after that if i dont manually restart the wheel i can connect to any app Bluetooth is open and  it losses around 0.1v every hour 

these are my observations but i could be wrong

 

please excuse my grammar im not a good English speaker

 

[Chriull]

1 hour ago, Mike.P said:

As a new member to the forum i would like to say hi to everyone

Welcome!

1 hour ago, Mike.P said:

I have been charging this wheel since day one with a shelly plug and euc world because you can set your own charging settings

That's a really great feature from @Seba! Try to keep some charge log saved on the phone and/or computer so you can compare over time the cell degradation.

Although just following EUC Worlds computed battery internal resistance is a nice indicator, too.

1 hour ago, Mike.P said:

my wheel is fully charged at 84.00v when it reaches 84.1 the overvoltage/overcharge circuit is enabled

That's in theory true, but bms and motherboard have independent voltage measurement which can be quite off! Also if it works now it could drift over time and in the worst case prematurely cut off charging so that no balancing happens at all!

1 hour ago, Mike.P said:

and my charger is pulling 0 amps and 1.6 watts at this time

That's the right indicator! EUC World offers charge cut of based on charger power consumption! That's much safer than 0.1V surplus for theoretical max battery voltage!

1 hour ago, Mike.P said:

after one hour or something my wheel is back to 84.00v and beeps 5 times after that if i dont manually restart the wheel i can connect to any app Bluetooth is open and  it losses around 0.1v every hour 

One should turn the wheel of once it is charged. This loss seems to be the consumption of the mainboard?

1 hour ago, Mike.P said:

these are my observations but i could be wrong

Thanks for sharing!

1 hour ago, Mike.P said:

please excuse my grammar im not a good English speaker

Me neither - i don't care or even realize ! Hope my english is about ok.

Edited January 7 by Chriull

[Funky]

2 hours ago, Mike.P said:

As a new member to the forum i would like to say hi to everyone

I bought  a new KS 18 xl in August, i have done around 5k khm as a new rider and i really love it

I have been charging this wheel since day one with a shelly plug and euc world because you can set your own charging settings

my wheel is fully charged at 84.00v when it reaches 84.1 the overvoltage/overcharge circuit is enabled and my charger is pulling 0 amps and 1.6 watts at this time

after one hour or something my wheel is back to 84.00v and beeps 5 times after that if i dont manually restart the wheel i can connect to any app Bluetooth is open and  it losses around 0.1v every hour 

these are my observations but i could be wrong

 

please excuse my grammar im not a good English speaker

 

 

Hey.

It beeps 6 times. When done fully charging. 

After you have removed charging cable press power button once - it will power on/off really fast. Afterwards it will power off completely. 

Simply removing charging cable and not powering it up (Pressing the power button once.) - motherboard/bluetooth connection stays powered on. That's why the power draw..

You don't need to charge it every time.. Do it when you gonna ride or around 40-55% battery. I normally charge mine to 100% and 3hrs extra connected, when battery drops down to 50% battery. (~73V) As i don't ride big distances, i charge my wheel once every 2 weeks. When battery drops down to ~50%. But every time i do charge it - i do it to 100% +3hrs extra connected. I hear beeps around 3hrs later, that's why the extra 3hrs. It takes longer for me because the battery cells need more balancing over those 2 weeks not charging once.

Edited January 7 by Funky

[Chriull]

1 hour ago, Funky said:

You don't need to charge it every time

Depends about how one wants to ride. Less charge means less voltage which means less maximum speed before overlean/faceplant. And less range.

For

1 hour ago, Funky said:

As i don't ride big distances

that's great. Especially for lower speed/accelerations.

For others it could be a very bad recommendation.

[Funky]

36 minutes ago, Chriull said:

Depends about how one wants to ride. Less charge means less voltage which means less maximum speed before overlean/faceplant. And less range.

For

that's great. Especially for lower speed/accelerations.

For others it could be a very bad recommendation.

I think most will know how much range they need.. Meaning they will figure it out themselves. Speed/Range/Power needed..

I just gave an example.. If he doesn't need big ranges. He can choose not to charge it every time to 100%. If he goes for 4-10km ride and charge wheel every time. The battery would be sitting at 90-100% all the time, which is also kind of bad.

I'm 280lbs and have been riding 40km/h speed be battery 100% or 40%.. Still had 20%+ safety margin. 18xl starts to speed limit at around 30%. Meaning you can easily go 40km/h till then. > See i didn't mention anything about max speed, which is 50km/h on this wheel. I personally don't ride max speed. And beginner should not also..

If he rides close to 50km/h then sure.. It's better to have full battery. And as it get's lower one should go slower. Still you can go 40km/h any time. (If you got flat roads and don't accelerate like mad man.)

My wheel has never given me any warnings/alarms. Simply because i don't accelerate fast enough to get one.

Edited January 7 by Funky

[rcgldr]

2 hours ago, Funky said:

I normally charge mine to 100% and 3hrs extra connected,

Leaving the charger connected does nothing for my V8F (and probably V8 V8S V10 V10F), as my V8F shuts off charging once a cell pair reaches maximum voltage. Once my V8F has shut off charging, voltage has to drop to 82.6 volts before it re-enables charging, which can take about 8 hours. Since there is no passive cell balancing, it might be slowly balancing while disconnected from a charger just after a full charge. The pattern that may work is to charge to 100% a few hours before a ride, disconnect from charger or use eWheels rapid charger which self-disconnects and goes into standby once charge current drops below .3 amps. My pack is charging to about .3 volts less than it did in April 2023. I used the pattern and it may have gained .1 volt back. I'll continue with this pattern to see if it helps and post another comment if I see a change.

[Funky]

19 minutes ago, rcgldr said:

Leaving the charger connected does nothing for my V8F (and probably V8 V8S V10 V10F), as my V8F shuts off charging once a cell pair reaches maximum voltage. Once my V8F has shut off charging, voltage has to drop to 82.6 volts before it re-enables charging, which can take about 8 hours. Since there is no passive cell balancing, it might be slowly balancing while disconnected from a charger just after a full charge. The pattern that may work is to charge to 100% a few hours before a ride, disconnect from charger or use eWheels rapid charger which self-disconnects and goes into standby once charge current drops below .3 amps. My pack is charging to about .3 volts less than it did in April 2023. I used the pattern and it may have gained .1 volt back. I'll continue with this pattern to see if it helps and post another comment if I see a change.

Yeah - it's well known that those wheels (Some..) the charger turns off as soon as first cell gets to 4.2Volts.. Doesn't matter that some weaker cells are still 3.99V and some 4.1V.. (Just given "crazy" numbers.. The imbalance can be quite smaller.)

Edited January 7 by Funky

[rcgldr]

On 1/7/2024 at 2:15 PM, Funky said:

Yeah - it's well known that those wheels (Some..) the charger turns off as soon as first cell gets to 4.2Volts.. Doesn't matter that some weaker cells are still 3.99V and some 4.1V.. (Just given "crazy" numbers.. The imbalance can be quite smaller.)

From mrelwood's video about Inmotion EUCs not having passive cell balancing, the issue is the 1 k ohm resistors, That's a maximum of 4.2 milliamps balancing | discharge load on a cell group. The pack voltage would have to be almost exactly the same as the charge voltage for the charger current to be reduced to 4.2 milliamps or less. This isn't going to happen and instead one or more cell groups reach maximum voltage and charging is shut off.  The other issue is the V8F doesn't allow charging once the highest voltage cell groups drop below some specific voltage, but waits for pack voltage to drop below 82.6 volts before it re-enables charging.

Edited January 12 by rcgldr

[alcatraz]

If the charger shuts off, does the bleed resistor too?

[techyiam]

1 hour ago, alcatraz said:

If the charger shuts off, does the bleed resistor too?

Nope.

At least not on my S22.

The BMS keeps continue to bleed those cells that are above 4.2 V or maybe even at a little lower voltage.

[The Brahan Seer]

This might be why we are seeing EUC manufacturer's improving their safety all of a sudden (and it might be little to do with influencers)...

https://www.exponent.com/article/mandatory-chinese-safety-standards-e-bikes-soon-going-effect

[teoman]

Hello everyone,

I am a Fin living in Istanbul and the proud owner of an inmotion V11 which is at 800km at the moment. My driving conditions are probably very different to a lot of you out there constantly having to dodge scooters, e-scooters, pedestrians that move without purpose, crazy drivers, inattentive drivers, hills, potholes, uneven sidewalks, stray cats and dogs to name a few of the challenges. 

I had a kaabo skywalker 10C e-scooter, but the battery is pretty much toast after 2.5k km. I also have Zero x10 that I think could be having the beginning of battery issues at 1500 km, hence my desire to investigate best practices. 

I came to this forum after trying to connect with mrelwood on youtube after watching his fantastic youtube video discussing the battery balancing.

My initial idea after watching his video was setting up a power supply that would deliver the battery charge voltage but at a very limited current, i.e. 4.2 mA as discussed above. Charge to 82V is (for a 84V inmotion) and then hook up the limited power supply which is weak enough that it could be overcome by the balancing resistors. Charging this way would take a very very long time, but it would ensure that the cells are completely balanced. 

I am very in to home automation and have installed homeassistant (Home automation server that does not need to connect to the cloud) and I have some shelly plugs (a plug that has a relay inside it that also monitors power and energy). So, I will be able to monitor the power consumption of the charger. This will enable me to calculate the energy that is fed to the euc. In addition I would like to implement the wisdom in this thread:

1) Monitor the power consumption, once it reaches the minimum, trigger an automation to cut power to the charger after 2 hours.

1.1) Monitor power consumption, cut charging once it reaches 0, then restart the charging after 2 hours for 1 hour giving time to the balancers do their job.

2) Setup the charging according to my schedule, so that the above automation will complete just in time for when I am ready to leave the house (So the euc does not needlesly wait with a full charge longer than necessary)

3) If i manage to detect that the batter is out of balance, i could program the charger to come on for 5 minutes every hour, hopefully giving the balance resistors time to bleedoff the excess voltage before the next micro-charge.

 

Any ideas for further improvements? Or how this could fail spectacularly? 

 

Willing to share all data and implementation later on if anyone is interested. 

 

Teoman

 

[Chriull]

3 hours ago, teoman said:

My initial idea after watching his video was setting up a power supply that would deliver the battery charge voltage but at a very limited current, i.e. 4.2 mA as discussed above. Charge to 82V is (for a 84V inmotion) and then hook up the limited power supply which is weak enough that it could be overcome by the balancing resistors. Charging this way would take a very very long time, but it would ensure that the cells are completely balanced. 

Charging with 4.2mA to 82V is a uhm, very weird idea. It will take centuries (literally about 2 centuries). I assume you mean this as "trickle charger" once the battery is charged with a normal charger? Imho not necessary, as with a normal charger the charge current will decrease automatically as battery voltage increases!

Only charging to 4.1V per cell hinders charging. The bleed resistors are only activated for cells reaching 4.2V and more. So every unbalance below this ~0.1V is ignored.

3 hours ago, teoman said:

Monitor the power consumption, once it reaches the minimum, trigger an automation to cut power to the charger after 2 hours.

Li ion cell manufacturers specify charge stop once charge current drops below ~50-60 mA per cell - so ~200-240 for V11's 4 paralleled cells.

Leaving the battery on the cell longer will balance the cells just a little bit more (with your 82V most likely not) and stress the cells a little bit more by keeping it at "high" voltage.

3 hours ago, teoman said:

1.1) Monitor power consumption, cut charging once it reaches 0,

Power consumption never reaches 0. The charger has losses and feeds the motherboard which consumes power.

3 hours ago, teoman said:

then restart the charging after 2 hours for 1 hour giving time to the balancers do their job.

Trickle charging is bad for li ion cells.

3 hours ago, teoman said:

Setup the charging according to my schedule, so that the above automation will complete just in time for when I am ready to leave the house (So the euc does not needlesly wait with a full charge longer than necessary)

A main part of balancing happens after the charger is turned off. If the battery is out of balance and one (or more) cells are above 4.2V they are, after the charger is turned off discharged down to 4.2V again. Maximum single cell voltage is somewhere about 4.25. So ~0.05V of 4 paralleled 5Ah cells are to be discharged max. That's roughly about 5% of their capacity - 4*5Ah*5% = 1Ah. Discharging this with some roughly about ~100 Ohm bleeding resistor takes about 24 hours...(*)

3 hours ago, teoman said:

If i manage to detect that the batter is out of balance,

Detectable is the premature charger cutoff once a single cell group reaches the above mentioned threshold of ~4.25V. Once such a unbalance happened best is to just wait this ~24 hours.

Activated balancing resistors while the cells are about fully charged could be undetectable aside the motherboard current consumption and other cells still consuming small charge currents.

3 hours ago, teoman said:

Any ideas for further improvements?

Another manufacturer safety recommendation is to set a fixed maximum charge duration after which no further charging can happen.

I'd personally let the battery rest after charging current dropped below this ~200mA to give the resistors a chance to bleed of cells with voltages above 4.2V instead of additional trickle charging.

Once premature bms charge cutoff happened i'd plan such ~24h rests with charging inbetween again until no premature cutoff happens again. Best with separated packs as likely only one of them is unbalanced.

If premature charge cutoff does not stop after some cycles think about replacing/repairing the unbalanced pack.

 

(*) The bleeding resistor value of the V11 is unknown to me. The capacity difference around 4.2V is extrapolated from table two at https://batteryuniversity.com/article/bu-409-charging-lithium-ion. The fir the V11 used cells could have different characteristics. So the 24h are just a very very rough estimation.

[teoman]

20 minutes ago, Chriull said:

Trickle charging is bad for li ion cells.

I was not intending for trickle charging like with Lead acid cells, my intentions are to give the balancing resistors time to do their job. (I noted the 24H and am very appreciative for the calculation). 

 

25 minutes ago, Chriull said:

Power consumption never reaches 0. The charger has losses and feeds the motherboard which consumes power.

Yes, by 0, I intended on measuring the charging current or power after several hours of charging and using that as the base line. (1.6W as indicated previously in this thread, measured using the same equipment I have). 

 

 

So would it be better to implement my setup to charge fully once I get home and cutoff once the power goes back to baseline OR is there any value in timing my charging end time to coincide with the next use of the wheel?