Help! Battery Conundrum

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
A strange one this.

I have a Pinnacle Mercury (Evans one brand 36v 250w Bafang G020 hub motor) which I’ve owned from new since late March. I’m careful with charging so the battery comes off the bike and is charged inside below 10°C otherwise I prefer to leave it in situ and charge on the bike. It’s around 11 Ah/360 Wh and I usually do 20 miles a ride and charge it just before the next ride.

Anyway, last week I did 20 miles which ran down to 3/5 bars and I charged it mid week ready for a ride today. I use a timer to make sure the power is turned off after a couple of hours or so (depending on battery level). For some reason, I didn’t check the in built power meter but the charger had turned green which indicates that charging is complete.

Switched the bike on today and zilch. Checked in built meter and nothing at all on the indication light so I plugged in the charger. Checked after 90 minutes (nowhere near full charge time) and it was fully charged according to the battery indicator, charger, and handlebar controller. Went for a ride using various power levels, figuring max power would stress a duff cell, and it behaved perfectly with battery depletion in line with previous range etc.

So my assumption is that something put the battery into hibernation mode and the charger powered up the BMS and revived it. The question is what ? The battery clearly wasn’t fully depleted as it initially looked. Only thing I can think of was that the battery was charged at around 10°C warmer than today but it’s done at the back of a brick built garage so max temp is only ever around 25°C. The difference was enough for a 8 or 9 PSI tyre pressure difference.

Is this likely to be a one off or something I should be concerned about ?

Chicken madras.
 

vfr400

Esteemed Pedelecer
Jun 12, 2011
9,822
3,931
Basildon
You guys make simple things difficult. Plug the charger into the battery, wait for the green light, unplug it and ride your bike. You don't need to worry about temperatures or anything else.
 

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
That’s precisely what I did. 72 hours later and what looked like a completely flat battery. No lights on the battery self test and the handlebar power unit wouldn’t switch on. Plugged it in again and after 90 minutes normal service was resumed. A fully flat battery would have taken 3+ hours.

So what did I do wrong this time that I haven’t mucked up on the previous 10 or so recharges ?
 

Tony1951

Pedelecer
Mar 27, 2016
198
95
70
It sounds as if the BMS shut down. The battery clearly wasn't flat even though nothing was coming out of it. I can't shed light on why it went into this state. Obviously they are designed to shut down short circuits on the output; under voltage cells; or in certain cases where thermistors are included in the battery, batteries which have become too hot or cold.

Monitor what happens in future and tell us whether or not it recurs and in what circumstances.
 
  • Like
Reactions: Barnsleyrob

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
It sounds as if the BMS shut down. The battery clearly wasn't flat even though nothing was coming out of it. I can't shed light on why it went into this state. Obviously they are designed to shut down short circuits on the output; under voltage cells; or in certain cases where thermistors are included in the battery, batteries which have become too hot or cold.

Monitor what happens in future and tell us whether or not it recurs and in what circumstances.
Will do. I hope it’s a one off as I don’t really want to have to deal with Mike Ashley’s finest over a battery warranty issue.
 

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
Full charge on Thursday with fairly even temps and it’s all looking good so far. Unhelpful comments from 400 therefore ignored. Plick.
 

Tony1951

Pedelecer
Mar 27, 2016
198
95
70
"I use a timer to make sure the power is turned off after a couple of hours or so (depending on battery level). For some reason, I didn’t check the in built power meter but the charger had turned green which indicates that charging is complete."

Just thinking about your original post, and wondering if perhaps the charger plug was not quite pushed home and the charge was interrupted by a half made contact. It is difficult o imagine why a charging system would terminate early like your problem one seems to have done. Just a wild guess, but the issue seems to have been a one off. You also say that you had not checked the charge state when you came back to use the bike. Could be a one off 'fluff' up of the plugging in or a mess up with the timer.

As for VFR400's remark, I don't think he meant anything ill by it. It is easy to get a bit 'anal' about these matters. My thoughts on this are that communities like this tend to get obsessed with the minor details of what you might call 'best practice' when the reality is that in our case, ebikes are a consumer item and don't need a specialist engineering approach to normal use, as long as you don't do anything really stupid (not that you would). What he said is right: Plug it in, wait for the green light, and ride it. So, you might by stopping the charge at 41 volts get 1000 charge cycles at 80% capacity rather than 900 at 65%, but would that be worth all the trouble? I don't know, people are different.
 
  • Like
Reactions: artspeck

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
Well, it sort of let me down again.

After a full charge around 7 days before my ride, I got into the habit of checking the battery status each day and all looked good. However, the day before I was due to go out it was dead as the proverbial so I got out the charger and gave it a tickle.

The battery has 3 indicated charge levels, Blue - 100%, Green > 25%, Red <25%. A few seconds after plugging in the charger, the battery LED briefly went blue then started flashing green, which is the usual charging status. After 15 minutes the LED went out followed shortly by the charger status turning from Red to Green.

The following day, the bike was fine and delivered 15 trouble free miles with 5/5 bars battery power still showing on the Bafang control unit.

Once again, the only thing of any significance was a noticeable drop in temperature on the day it failed.

I’ve contacted the manufacturer (rather than Evans Cycles) so will await their comments.
 

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
This very helpful explanation from the manufacturer.

“ Sorry, , after internal discussed and clarify with our RD, the symptom you mention is battery normal behavior , this action just remind you the battery is almost full, thanks.. “

Which I assume means that the battery shuts down completely to remind you that you charged it.
 

Nealh

Esteemed Pedelecer
Aug 7, 2014
14,540
5,638
58
West Sx RH
I have never yet had a battery act in this way, me thinks you are being fobbed off with nonsense from a salesman.
 
  • Agree
Reactions: flecc

flecc

Member
Oct 25, 2006
49,290
27,001
Shutting down at full charge is rather like a premature overvoltage cut off. One bike I owned had a very tight overvoltage limit and could cut out in that way at the peak of charge and the controller refusing to operate on that battery. But would once the voltage dropped off slightly.
.
 
  • Like
Reactions: Nealh

Nealh

Esteemed Pedelecer
Aug 7, 2014
14,540
5,638
58
West Sx RH
Yes can agree with flecc there, a remote possibility is (an unusual one at that) that the BMS allows over charge and as mentioned the slight over charge will not allow the BMS to switch the battery on until it has self bled down.

Most likely scenario is a cell group is/was out of balance and the others or some charged to full first hence the BMS tells the charger it is charged and turns green and the battery indicator not agreeing as it measures over all voltage and not single cell group voltage as the BMS does. Once the single cell group over charge bleeds down to a suitable voltage the BMS allows it to switch on.

The BMS carries out top end balancing but if the voltage disparity is a lot then it won't do so and may take several charges to right, with older cells it won't balance at all.
 
Last edited:

Nealh

Esteemed Pedelecer
Aug 7, 2014
14,540
5,638
58
West Sx RH
You will have to see if this happens time after time or is just a one off .
 

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
So far it’s happened twice out of 18 charge and ride cycles. Both were similar circumstances where it was fully charged and self tested for several days before it shut down seemingly as the temperature dropped.

As far as the charge cycles were concerned, they both took the same time as on previous charges from roughly the same levels (we tend to use 2 or 3 regular routes) and the ride cycles were exactly the same once the battery was recovered.

One charge was undertaken on the bike and the other off the bike, but again that’s not unusual in that my early charge cycles were off the bike due to the outside temp being below 10°C.

At least now I can charge the bike “just in time” to prevent further shutdowns and, if my theory is right, it should likely behave more predictably as the weather cools. In fact, I’m currently leaving the battery as it finished the last ride and don’t expect any issues.

Out of interest, does a BMS constantly monitor the cells or does it go into some sort of suspended state ?
 

Nealh

Esteemed Pedelecer
Aug 7, 2014
14,540
5,638
58
West Sx RH
No it doesn't constantly monitor or balance after charging, the common denominators are hvc during charging and lvc during discharging

If a battery is way out of balance then the BMS stops play during charge or discharge, the first to hit 4.15 - 4.2v will stop any charge and during discharge the first to hit lvc will cause drive to cease. Lvc varies depending on the BMS spec it could be anywhere between 2.8v - 3.2v, often more so above 3v.

Should BMS balancing fail during charging then there is a secondary back up and the BMS will cease charging at a higher voltage of 4.25v and one hopes the bleed resistors open and bleed off excess voltage.
 

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
No it doesn't constantly monitor or balance after charging, the common denominators are hvc during charging and lvc during discharging

If a battery is way out of balance then the BMS stops play during charge or discharge, the first to hit 4.15 - 4.2v will stop any charge and during discharge the first to hit lvc will cause drive to cease. Lvc varies depending on the BMS spec it could be anywhere between 2.8v - 3.2v, often more so above 3v.

Should BMS balancing fail during charging then there is a secondary back up and the BMS will cease charging at a higher voltage of 4.25v and one hopes the bleed resistors open and bleed off excess voltage.
Useful explanation. Thanks.

My simplistic explanation to myself has been that the battery holds slightly more charge at 25°C than it does at 18°C and that that is sufficient to trip the BMS when the temperature drops (Although I’ve got no clue if that has any factual basis). Charging in the house wont help cos it’s hotter than the garage most days.
 

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
Hmmm. After my last ride on 15 August, I left the battery at around 75% rather than recharging it and was hopeful it would remain operation but on Saturday 21 August it had shut down again having responded to status checks on each of the other days.

I believe the EzeGo e-bikes use the same battery and that the BMS puts the battery into sleep mode after around 40 minutes, but on that bike a press of the battery status button wakes it up again unless it was allowed to discharge fully.
 

Barnsleyrob

Pedelecer
Jul 20, 2020
144
22
The latest explanation is that the battery goes into sleep mode after 6 days of no power being drawn. Holding the LED status indicator/button for 5-10 seconds will wake it back up without the need to plug in the charger and, to my eternal scepticism, it works.

A manual may have avoided the confusion though.
 
  • Like
Reactions: Tony1951