How normal is it for a battery to go badly out of balance?

Tony1951

Esteemed Pedelecer
Mar 27, 2016
362
243
73
I have a three year old 10S 6P battery made of genuine Sanyo cells which is not used that much and periodically goes out of balance. It mostly gets well out of balance over winter time when it is unused. I do give it a periodic re-charge to full about once every six weeks or so with the normal charger when it is unused to avoid it getting too low, but gradually this one group will end up at about 3.95volts after a full series charge with the 42 volt charger while the other groups end the charge at 4.2 volts.

The bike has only had one decent outing this year - a ride two weeks ago of about twelve miles with lots of heavy hill climbing. I can quite easily put that group back to comparable voltage with the other groups by selectively re-charging the weak cells with a 4.15 volt lithium charger.

Five days ago, I got it back to near the voltage of the other cells so that it was 4.17v while the others in the battery were at 4.2v. So the weak group was only 0.03volts lower than the rest.

By this morning it had reduced to 4.11 while sitting on the bench, while the rest were 4.17v, so the weak group had dropped 0.07volts while the good groups had only dropped 0.03 volts.

Can anyone explain what is likely to be the cause of this gradual decline which is out of kilter with the other groups?

I am about to take it out for some work and will measure the voltage differences when I get back before re-charging with the 42 volt charger. I will report back on the voltages after my ride to see if giving the battery some serious work to do makes the voltage drop on the weak cells worse than in the others or whether it is just a long term leakage problem for that group of cells.
 

saneagle

Esteemed Pedelecer
Oct 10, 2010
4,038
2,007
Telford
That's very normal for many batteries. You can solve it by unplugging the multi-pin connector between the cell-pack and BMS when you put your battery into storage.

The BMS itself needs a small amount of power, which is often taken from only one cell group, so that group would go down in time. The balancing uses the "little and often" principle, so, in normal use, that cell gets topped up each time you charge.

This is one of the reasons you should not charge your battery only to 80%. The cell-group that provides the power to the BMS will go down and down. An occasional full charge is not enough to restore it. It's designed to do it a bit each full charge.
 

Tony1951

Esteemed Pedelecer
Mar 27, 2016
362
243
73
Thank you Saneagle. I appreciate your input and experience.

I just took the bike for a hard ride - only 9.6 miles, but loads of hill climbing, and steep too. The route included 1004 feet of ascent and the steep climbing parts were done on high PAS setting, with pedalling and full throttle drawing 500 watts during the climbing stages.

At the end of this, the good cells were standing at 3.83 Volts and the dodgy group at 3.753 volts - a difference of 0.081volts on the weaker group.

I am charging the battery now and will report the final state when it is done. I think you are right Saneagle in implying there is not much to worry about here.

Cheers.
 

saneagle

Esteemed Pedelecer
Oct 10, 2010
4,038
2,007
Telford
Thank you Saneagle. I appreciate your input and experience.

I just took the bike for a hard ride - only 9.6 miles, but loads of hill climbing, and steep too. The route included 1004 feet of ascent and the steep climbing parts were done on high PAS setting, with pedalling and full throttle drawing 500 watts during the climbing stages.

At the end of this, the good cells were standing at 3.83 Volts and the dodgy group at 3.753 volts - a difference of 0.081volts on the weaker group.

I am charging the battery now and will report the final state when it is done. I think you are right Saneagle in implying there is not much to worry about here.

Cheers.
Often, you can improve it by leaving it on charge after the green light, otherwise it'll sort itself out in time if you charge it regularly - ALWAYS TO FULL, of course!!!!
 

Tony1951

Esteemed Pedelecer
Mar 27, 2016
362
243
73
Often, you can improve it by leaving it on charge after the green light, otherwise it'll sort itself out in time if you charge it regularly - ALWAYS TO FULL, of course!!!!
Thanks. I really appreciate the expert guidance here. The red light switched off at 4.17volts on the battery group which had been well out of balance and initiated the thread. Right now, I am leaving it on charge with the green light on to see if it balances and rises up a bit more. I expect it will, but it takes a while for balancing to work through.

I think you have nailed the issue, and in fact, you mentioned that to me a week or two back when I raised this matter elsewhere. Maybe it is indeed down to the BMS running itself off one group of cells and gradually putting the battery out of balance.

It is now showing 4.18v. Even if I don't get any further balancing action, at 4.18 volts, I would not be loosing any range because it is pretty near the full charge voltage of 4.2v.
 

guerney

Esteemed Pedelecer
Sep 7, 2021
10,168
2,989
Are you charging with the battery switched on or off? Contrary to the instructions supplied with my battery, I discovered recently that mine only appears to balance charge when the battery is switched on.
 
Last edited:

saneagle

Esteemed Pedelecer
Oct 10, 2010
4,038
2,007
Telford
Are you charging with the battery switched on or off? Contrary to the instructions supplied with my battery, I discovered recently that mine only appears to balance charge when the battery is switched on.
Not sure about that. I've never seen it. how are you judging when balancing is happening?
 

guerney

Esteemed Pedelecer
Sep 7, 2021
10,168
2,989
Not sure about that. I've never seen it. how are you judging when balancing is happening?
The resulting voltage of my battery pack is higher if I leave the battery charging for an hour, after the red light on the charger goes green, charging while it's switched on. The battery charges to a lower voltage if charging while the battery is switched off, and left for an hour after the charger light goes green. Also battery case surface temperature is higher when it's charged while switched on, which I assume is due to bleed resistors bleeding heat into the case during the balancing phase.

Some measurements here, on my GuerneyHome Ruination by Ebike Battery Fire Expulserating PreTerminator™®©℠ thread:

https://www.pedelecs.co.uk/forum/threads/home-ruination-by-ebike-battery-fire-expulserating-preterminator%E2%84%A2%C2%AE%C2%A9%E2%84%A0.46934/post-708768
 

saneagle

Esteemed Pedelecer
Oct 10, 2010
4,038
2,007
Telford
The resulting voltage of my battery pack is higher if I leave the battery charging for an hour, after the red light on the charger goes green, charging while it's switched on. The battery charges to a lower voltage if charging while the battery is switched off, and left for an hour after the charger light goes green. Also battery case surface temperature is higher when it's charged while switched on, which I assume is due to bleed resistors bleeding heat into the case during the balancing phase.

Some measurements here, on my GuerneyHome Ruination by Ebike Battery Fire Expulserating PreTerminator™®©℠ thread:

https://www.pedelecs.co.uk/forum/threads/home-ruination-by-ebike-battery-fire-expulserating-preterminator%E2%84%A2%C2%AE%C2%A9%E2%84%A0.46934/post-708768
I suspect that when it's switched on, something else, like USB, is taking a bit of power, or, if the battery is still in the cradle, something else is taking a bit. Do you have a DC?DC 12v converter?
 

guerney

Esteemed Pedelecer
Sep 7, 2021
10,168
2,989
I suspect that when it's switched on, something else, like USB, is taking a bit of power, or, if the battery is still in the cradle, something else is taking a bit. Do you have a DC?DC 12v converter?
There's no USB port on my battery case, just the BMS inside with no DC converter attached. Plus the battery pack charges to a higher voltage, when charged with the battery switched on.
 
Last edited:

saneagle

Esteemed Pedelecer
Oct 10, 2010
4,038
2,007
Telford
There's no USB port on my battery case, just the BMS inside with no DC converter attached. Plus the battery pack charges to a higher voltage, when charged with the battery switched on.
Are you measuring with the charger switched on or off, because the charger will hold the voltage higher when the battery is switched on? Basically, you're measuring the charger voltage, not the battery voltage.
 

guerney

Esteemed Pedelecer
Sep 7, 2021
10,168
2,989
Are you measuring with the charger switched on or off, because the charger will hold the voltage higher when the battery is switched on? Basically, you're measuring the charger voltage, not the battery voltage.
I measured battery voltage with the battery switched on, after charging, with the charger disconnected. Battery voltage is higher after charging with the battery switched on, than when it's been charged with the battery switched off. Plus, measured voltage of the charged battery (when the battery was switched on while charging) lowered after a time, but not as much as when I charged the battery while it was switched off.

https://www.pedelecs.co.uk/forum/threads/home-ruination-by-ebike-battery-fire-expulserating-preterminator%E2%84%A2%C2%AE%C2%A9%E2%84%A0.46934/#post-708768
 
Last edited:

Tony1951

Esteemed Pedelecer
Mar 27, 2016
362
243
73
Often, you can improve it by leaving it on charge after the green light, otherwise it'll sort itself out in time if you charge it regularly - ALWAYS TO FULL, of course!!!!
Update:

I left the battery on the normal serial charger after the red light went out and the green light came lit. In time, the suspect cell group arrived at 4.2 volts - the full charge voltage. I disconnected and left it for an hour and it declined very slightly to 4.19volts, so the pack is now completely balanced.

My prediction is that this group will very slowly decline a bit over time, and go a bit lower.

I expect that Saneagle is correct and that if the pack is used regularly and charged to full, the suspect group of cells in the 10S 6P pack will be balanced by the BMS while the green light is showing.

I'm kind of dubious about leaving the pack connected to the mains for longer periods, in case of circuit failure in the BMS and the charger - both of which appear to have some safety related cut off circuitry, but the price of using a serial charger is that some cell groups can go low unless you do allow extensive balance charging on a regular basis..... I think that is right. Others may comment if not.

I think it might be better if the charging mechanism charged each group separately to 4.2 volts. It would need more complexity, but I think a BMS could be made to do that. Then each cell group would receive an optimum charge. I'd be surprised if electric cars don't charge that way. We have to balance complexity and cheapness though, so we have what we have. I know it is perfectly possible to charge an individual cell pack separately from the others because I had to do that when the suspect group became too far out of balance. I just connected a 4.2v lithium charger to the positive and negative end of that particular group.
 
Last edited:

saneagle

Esteemed Pedelecer
Oct 10, 2010
4,038
2,007
Telford
Update:

I left the battery on the normal serial charger after the red light went out and the green light came lit. In time, the suspect cell group arrived at 4.2 volts - the full charge voltage. I disconnected and left it for an hour and it declined very slightly to 4.19volts, so the pack is now completely balanced.

My prediction is that this group will very slowly decline a bit over time, and go a bit lower.

I expect that Saneagle is correct and that if the pack is used regularly and charged to full, the suspect group of cells in the 10S 6P pack will be balanced by the BMS while the green light is showing.

I'm kind of dubious about leaving the pack connected to the mains for longer periods, in case of circuit failure in the BMS and the charger - both of which appear to have some safety related cut of circuitry, but the price of using a serial charger is that some cell groups can go low unless you do allow extensive balance charging on a regular basis..... I think that is right. Others may comment if not.

I think it might be better if the charging mechanism charged each group separately to 4.2 volts. It would need more complexity, but I think a BMS could be made to do that. Then each cell group would receive an optimum charge. I'd be surprised if electric cars don't charge that way. We have to balance complexity and cheapness though, so we have what we have. I know it is perfectly possible to charge an individual cell pack separately from the others because I had to do that when the suspect group became too far out of balance. I just connected a 4.2v lithium charger to the positive and negative end of that particular group.
Its quite complicated to charge each group seperately. I have such a charger, but it can only do a max of 6 cells at a time. I use it for balancing when I build or repair a battery. You have to charge through the multi-pin connector, which can be complicated because there are several types and they're wired in many different ways.
 
  • Like
Reactions: Tony1951