Please stop charging ebike batteries overnight!

[spokewrench]

Using a timer with any lithium battery charger is always a good idea. I prefer mechanical over electronic timers since they are less likely to malfunction from AC power surges, blips or outages.

Mine is mechanical. Previously you said you use a double-pole to isolate the battery from the charger. That sounds complex. I don't know, but I imagine both the battery and the charger would have semiconductors to prevent "backflow."

I log the voltage before and after charging and use that database to set the timer. I also use a watt meter. If the green light comes on the charger and the meter says it's drawing more than 2 watts, I assume it's equalizing and leave it on. An abnormal charging wattage could mean trouble. In 4 years I haven't seen any. I assume that if my battery capacity deteriorates, it will take fewer watt hours to top it off from a given voltage level. So far I haven't seen it. While charging, I like to check temperature with my hand or an IR thermometer.

 

[gromike]

How do you do a balance charge on your batteries, providing that your batteries' bms does that.

 

[harryS]

How do you do a balance charge on your batteries, providing that your batteries' bms does that.

I won't charge over night when I'm sleeping. I plug it in during the day and if I'm on top of my game, I'll remember unplug it around midnight. That's a pre-sleep routine anyway. Walk out to the garage to make sure door is closed, batteries unplugged, car locked in driveway,

 

[spokewrench]

How do you do a balance charge on your batteries, providing that your batteries' bms does that.

I assume that when the BMS is supplying current to undercharged cells, it will show up in input (AC) watts. Normally, when the charger shows a green light, the input is about 1 watt. If the watt meter reads much more, I assume the BMS is topping off low cells.

 

[kevinmccune]

Lots of hate is being unfairly heaped on e-bike batteries these days. I'd like to see some statistics comparing lithium battery fires with those caused by smoking in bed, kids playing with matches, electric heaters and other preventable fires caused by human ignorance. You don't see much about those incidents right now. It seems the press would rather vilify the product rather than the user.

just like you dont see much about ICE vehicle fires the only thing that "cooks off" around here are usually overheated trucks or ICE vehicles with faulty fuel systems or exhaust leaks.

 

[Slaphappygamer]

How do you do a balance charge on your batteries, providing that your batteries' bms does that.

I have a watt meter at the mains and plug in the charger. I note the AC current when the charger is unplugged from the battery, about .033A. When the green light comes on the charger, I see the AC current at about .155A. This is when balancing takes place, the time between the green light comes on and when the AC current drops to .033A. When AC current drops to .033A, I unplug the charger now that I see it’s completed. I do this about once a month.

 

[il Munduato]

Are you sure about that?
I believe what you are witnessing is the basic function of a CC - CV charger/bms. As target voltage is approached, current trails off.

I believe if balancing was actually taking place you would see a current increase or remain steady as the target voltage is reached and resistors are implemented to burn off current on the full cells allowing the others to catch up and the full cells to not overcharge.
Or you would need some pretty sophisticated circuitry for cell isolation and/or current redistribution.

y bms mv

 

[il Munduato]

Not fully charging when unnecessary to accomplish your ride is probably the best thing you can do to keep your battery healthy and safe.

I've gone automated!

 

[Slaphappygamer]

Are you sure about that?
I believe what you are witnessing is the basic function of a CC - CV charger/bms. As target voltage is approached, current trails off.

View attachment 185559


I believe if balancing was actually taking place you would see a current increase or remain steady as the target voltage is reached and resistors are implemented to burn off current on the full cells allowing the others to catch up and the full cells to not overcharge.
Or you would need some pretty sophisticated circuitry for cell isolation and/or current redistribution.

y bms mv

Yes. That’s what I’m seeing. When charging is first started, I see constant AC current at 1.54A. Then it slows gradually until the green light is illuminated. That’s when I see the current slowly drop from .5~A to finally to .033A.

 

[spokewrench]

Are you sure about that?
I believe what you are witnessing is the basic function of a CC - CV charger/bms. As target voltage is approached, current trails off.


I believe if balancing was actually taking place you would see a current increase or remain steady as the target voltage is reached and resistors are implemented to burn off current on the full cells allowing the others to catch up and the full cells to not overcharge.
Or you would need some pretty sophisticated circuitry for cell isolation and/or current redistribution.

y bms mv

How could balancing cause an increase in watts?
Suppose I have a battery with 3 columns of 13 cells in series. I'll round the fully charged voltage down to 52, or 4 volts per cell. A 2-amp charger would use 156 watts (nominally) to deliver 2/3 amp per column. (For AC I prefer to measure watts because power factor can make amp readings misleading.)

I don't see how balancing could draw more than the charger's limit of 2 amps. The column voltage will be the sum of the voltages of its 13 cells. Each cell's voltage will be the sum of its state of charge and the product of its current and its resistance. At some point the combination of state of charge and resistance voltages will reach the nominal 52. As the state of charge continues to climb, current will have to diminish to stay at a total of 52.

I usually see something close to that. The problem is self-discharge, which varies widely between cells. Suppose the worst cells in a battery had a rate 1ma faster than the best. In two months, when the best cells finished charging, the worst would be 1.5 AH low. If they show less than the fully charged voltage, which I'll call 4.00, the charged cells in the column will show more. That can cause oxygen venting, which shorten's cell life, and heating that could be disastrous. I believe an important function of a BMS is to give each cell a bypass circuit to prevent overvoltage while other cells in the column continue to charge.

A couple of years ago I took out a bike to check the odometer on a measured track. I hadn't ridden it lately. The display read 48 volts, so I didn't charge it. When I used the throttle even slightly, the battery would cut out. I took that to mean some cells were so low that they impeded the flow. The display showed that other cells must have been much better charged. I hadn't charged it in months, and before that I'd followed the fad of charging only to 80%.

The first part of the charge was normal, but as the wattage tapered, the process was slow. I think it took days because I used the timer to charge only when I was present. A green light doesn't always mean a battery is balanced. If I let it rest and turned on the timer an hour later, it might charge at 20 watts. In chemical batteries, a buildup of ions can fool a charger. A rest will let them dissipate. I was rewarded with a battery that after 2 years and 6,000 miles seemed to have the same range and performance as new.

These days I think the best way to be sure a battery is balanced is to turn on the charger after an hour's rest to be sure it's still pulling only a watt or so.

 

[harryS]

Have built more than a few batteries for my ebikes, I've watched them charge/discharge carefully and make the following observations.

When I monitor the charging current between charger and the battery, it drops to zero as the pack hits full charge, unless it's a balance BMS. Then I'll see a very low current as the cells are equalized. By low current, I mean around 100 ma or less. Balance time can vary, Hours to days to forever, As for the green charger light, it depends on the charger and BMS.

Most chargers will keep the charger voltage present when the light turns green. That's all that is needed for a balance BMS to complete balancing, If you bought the bike/battery/charger as a package, you'll have the right charger for your battery,

I'll also say that only the higher quality bikes and batteries have balance BMS circuits. The rest just use a BMS that checks for safe cell voltages. I've an opinion that a balance BMS installed with cheap cells may actually be a liability. If a battery is going to fail because cheap cells drifted apart, it's better to scrap that battery than keep it limping along with balance circuitry. I think those are the fire risks.

 

[il Munduato]

Have built more than a few batteries for my ebikes, I've watched them charge/discharge carefully and make the following observations.

When I monitor the charging current between charger and the battery, it drops to zero as the pack hits full charge, unless it's a balance BMS. Then I'll see a very low current as the cells are equalized. By low current, I mean around 100 ma or less. Balance time can vary, Hours to days to forever, As for the green charger light, it depends on the charger and BMS.

Most chargers will keep the charger voltage present when the light turns green. That's all that is needed for a balance BMS to complete balancing, If you bought the bike/battery/charger as a package, you'll have the right charger for your battery,

I'll also say that only the higher quality bikes and batteries have balance BMS circuits. The rest just use a BMS that checks for safe cell voltages. I've an opinion that a balance BMS installed with cheap cells may actually be a liability. If a battery is going to fail because cheap cells drifted apart, it's better to scrap that battery than keep it limping along with balance circuitry. I think those are the fire risks.

Though there are more than a few variations to skin this cat... You're observations are the same as mine.
The Passive balancing offered by eBike BMS's use resistors to bleed current to keep the full cells from over charging. This happens at the very end of the charge when current has already fallen in the CC-CV process. So there will be current increase in the mA range. Depending on how out of balance the cells are (and it should not be much for a quality pack) and your test equipment it may be hard to detect.
Since most mid to low end batteries do not have a balance BMS... what most witness is what I described above... Just a toping off of cells at low current as the CC - CV process ends.
Bottom line... No Balance BMS = No Balancing.

 

[il Munduato]

and before that I'd followed the fad of charging only to 80%.

I draw my conclusions and implement best practices based on technical documentation produced by professionals.

BU-808: How to Prolong Lithium-based Batteries

BU meta description needed...

batteryuniversity.com

Not anecdotal observations found on the internet.

 

[spokewrench]

I draw my conclusions and implement best practices based on technical documentation produced by professionals.

BU-808: How to Prolong Lithium-based Batteries

BU meta description needed...

batteryuniversity.com

Not anecdotal observations found on the internet.

Next time you find some anecdotal smart phone research on the internet, you should put on your thinking cap and ask yourself if applies to e-bikes.

Cadex tested cells, not batteries. A battery is cells in series. If my bikes were powered by single cells like smart phones, smart phone research would apply because a cell can't get out of balance with itself. Unfortunately, my bikes are powered by batteries. As I discovered a couple of years ago, a battery that hasn't been charged above 80% in months will be useless until balanced.

 

[spokewrench]

Though there are more than a few variations to skin this cat... You're observations are the same as mine.
The Passive balancing offered by eBike BMS's use resistors to bleed current to keep the full cells from over charging. This happens at the very end of the charge when current has already fallen in the CC-CV process. So there will be current increase in the mA range. Depending on how out of balance the cells are (and it should not be much for a quality pack) and your test equipment it may be hard to detect.
Since most mid to low end batteries do not have a balance BMS... what most witness is what I described above... Just a toping off of cells at low current as the CC - CV process ends.
Bottom line... No Balance BMS = No Balancing.

If passive balancing simply used bypass resistors, the self-discharge would be horrendous.

If bypassing charged cells caused any increase in current, then the low cells, not being bypassed, would see an increase in charge current as they approached full charge. That sounds like a disaster in the making.

I don't know what voltage my chargers generate. It must be at least 55 because my batteries come off at 54.5. I have monitored charging watts many times and never seen an increase of even 0.1 during the taper. An increase of 0.1 W input would be an increase of 1.8 ma charging current, and I haven't seen it.

No Balance BMS = No Balancing? How much have you had to drink?
You might be more coherent if you used the term "active balancing." In my experience, passive balancing works fine both for keeping batteries in balance and for restoring batteries badly out of balance.

How Does A BMS Balance A Lithium Battery? - Cell Saviors

BMSs balance lithium batteries by two main process which vary from bms to bms, read more on this here!

cellsaviors.com

 

[PCeBiker]

When I monitor the charging current between charger and the battery, it drops to zero as the pack hits full charge, unless it's a balance BMS.

The Passive balancing offered by eBike BMS's use resistors to bleed current to keep the full cells from over charging.
Bottom line... No Balance BMS = No Balancing.

No Balance BMS = No Balancing? How much have you had to drink?

Most ebike batteries do not have a balancing BMS.

If they do have a balancing BMS, they either have an Active or Passive Balancing function.

 

[il Munduato]

Thing is I'm not sure how many implement the more expensive active balancing.. especially those not of the most expensive brands. Though recently prices are dropping and with the UL regulations being enforced we may see battery builders willing to use higher end components.

Next time you find some anecdotal smart phone research on the internet, you should put on your thinking cap and ask yourself if applies to e-bikes.

Cadex tested cells, not batteries. A battery is cells in series. If my bikes were powered by single cells like smart phones, smart phone research would apply because a cell can't get out of balance with itself. Unfortunately, my bikes are powered by batteries. As I discovered a couple of years ago, a battery that hasn't been charged above 80% in months will be useless until balanced.

If you actually read the document you would know it wasn't singe cell smart phone exclusive and referenced packs for laptops, EV's and satellites.
But the last thing I'm going to do is argue or try to educate a clueless tool that doesn't understand passive balancing at all, any of my statements... and thinks facts are a fad... especially one that found himself struggling for years with loose Schrader valve cores

 

[harryS]

That Cell Saviors article needs peer review. , They started out several years ago taking apart surplus ebike batteries and selling the cells to hobbyists. I bought some lightly used LG cells from them, and they worked out fine, I am sure they lost money selling those cells, so now it's a public service site. I think they know their stuff, but could write better,
-------------------------------------------------------------------------------

Does a BMS balance cells when not charging?​

Yes. In most cases, a BMS will continue to balance the cells when the battery is not charging. There are some really nice BMS that give you the option as to when balancing occurs. In those BMS, they can be set to only balance when the cells are charging, or only balance when they a re discharging. In those fancy BMS, lithium battery balancing can even be set to occur or not occur depending on the voltage level of the cell groups.

In contrast, the most basic, low-cost BMS will always balance the cells regardless of the state of other factors such as cell voltage, discharge or charge state, etc.
------------------------------------------------------------------------------------
There are some BMS where you can set them to balamce at lower voltages. This is great for someone who only charges to 90%. Still, all the basic BMS that have balance are set to do it at the end of the charge cycle, And it seems to imply that all BMS have balance when of course, many do not.

A passive BMS wil continue to balance when the charger is removed, but only for a brief period. As soon as the bleed reistors take the voltage under the balance trigger point, they shut off.

So no, our ebike balamce packs only balance when charging, And they use passive balance if they ise balance at all, Hobbyiists can buy active balamcers for their batteries, This one costs $125.

 

[il Munduato]

That Cell Saviors article needs peer review. , They started out several years ago taking apart surplus ebike batteries and selling the cells to hobbyists. I bought some lightly used LG cells from them, and they worked out fine, I am sure they lost money selling those cells, so now it's a public service site. I think they know their stuff, but could write better,
-------------------------------------------------------------------------------

That's an understatement.

Does a BMS balance cells when not charging?​

Yes. In most cases, a BMS will continue to balance the cells when the battery is not charging. There are some really nice BMS that give you the option as to when balancing occurs. In those BMS, they can be set to only balance when the cells are charging, or only balance when they a re discharging. In those fancy BMS, lithium battery balancing can even be set to occur or not occur depending on the voltage level of the cell groups.

In contrast, the most basic, low-cost BMS will always balance the cells regardless of the state of other factors such as cell voltage, discharge or charge state, etc.
------------------------------------------------------------------------------------
There are some BMS where you can set them to balamce at lower voltages. This is great for someone who only charges to 90%. Still, all the basic BMS that have balance are set to do it at the end of the charge cycle,

Balancing at lower voltages would be great but realize that you'd be at a higher current point of the charge and this wound cause more heat, especially in the case of a passive balance bms. This could be solved with a matched intelligent charger but additional safety measures would also need to be implemented. And using a generic charger could make things interesting.

And it seems to imply that all BMS have balance when of course, many do not.

This is where I believe our tool is stuck

A passive BMS wil continue to balance when the charger is removed, but only for a brief period. As soon as the bleed reistors take the voltage under the balance trigger point, they shut off.

So no, our ebike balamce packs only balance when charging, And they use passive balance if they ise balance at all, Hobbyiists can buy active balamcers for their batteries, This one costs $125.

I concur

 

[spokewrench]

There are some BMS where you can set them to balamce at lower voltages. This is great for someone who only charges to 90%. Still, all the basic BMS that have balance are set to do it at the end of the charge cycle, And it seems to imply that all BMS have balance when of course, many do not.

A passive BMS wil continue to balance when the charger is removed, but only for a brief period. As soon as the bleed reistors take the voltage under the balance trigger point, they shut off.

So no, our ebike balamce packs only balance when charging, And they use passive balance if they ise balance at all, Hobbyiists can buy active balamcers for their batteries, This one costs $125.

What e-bike BMSs lack balancing? My Radmission cost $500 new. If the BMS didn't balance, two years of differing cell self-discharge would have greatly reduced my range. According to my log, it's the same as it was new.

Radpower's balancing instructions were the same as for my Radrunner in 2020. Leave the charger on 12 hours. Ride even a short distance. Leave it on another 12 hours. Ride again. Leave it on 12 hours a third time.

It said you would need to do the balancing process again someday but didn't say when. That was daffy to me, so used a watt meter. I think it was in 2023 that Radpower's instructions changed, saying to turn off the charger as soon as the green light showed, and no other balancing was necessary.

I think that came from their legal team as charging 12 hours would entail leaving it unattended.

I didn’t understand when you said passive balancing can continue briefly after the charger is removed. Now I see, thank’s to @PCeBiker 's chart. Suppose the MOSFET trips when source-drain voltage is exactly 4.20. Source-drain voltage will immediately drop by the product of resistance and bypass current. It will switch off and on. With a voltage slightly over 4.20, the cell will trickle charge. When the charger is removed, the gate will continue to switch on and off until the cell voltage is down to 4.20.

That can explain what I've observed. Suppose the BMS will shut off charging current at 54.6 V (13 cells x 4.20 V.) Suppose 1 cell is undercharged, and trickle charging has brought the other 12 to 4.28 V. The BMS will stop charging when the low cell is at 3.48. After the current is shut off, battery voltage will drop as the 12 overcharged cells come down to 4.20. If the BMS shuts off, that will turn the charger light green and my AC meter will show a watt or less. If I give the charged cells an hour to come down to 4.20 and then turn on the charger, a current draw will show that there are still cells that aren't fully charged.

When this happens, the charging light may be red or green, depending on how much current the BMS is drawing.