Battery charging puzzle

[Cisco-man]

Hi. I’ve inherited a 10S 2P battery pack that won’t charge to 42v via its BMS.
The charger is giving 42.5v at the charge port of the BMS, and thinks that the battery is fully charged (green light on).
The cell pairs measure 3.9, 3.8, 4.1, 4.2, 3.9, 3.9, 4.1, 3.9, 3.6, & 3.8, with a total value of 39.2v when not under charge conditions, and that’s as high as it goes.
I’ve successfully charged the pack to 42.5v by charging via the output leads.
I’ve just changed the BMS and still no joy.

Any thoughts?

 

[Cisco-man]

I’m just going to discharge the pack down to 35v - to see if the BMS will charge from that voltage. Maybe something’s open-circuit?

 

[Nealh]

There is a cell group at 4.2v so the BMS won't allow further charging as the other cells are too far out of balance.
The pack is out of balance , the BMS won't balance such a large discepency of voltage. The 4.2v cell group needs discharging down 3.8/3.9v as do the pair at 4.1v, the 3.6v group needs raising similarly .

Then the BMS may balance out the cells if they are with in 0.1v or acheive it over two or three charge cycles..

Reverse charging is not recommended for a full charge as one is bypassing the BMS safety features, if wanting to do so simply don't bother with a BMS. The only other safe method of charging via the discharge is with a same port BMS or lipo style where cells can be closely monitorered cell by cell.

 

[saneagle]

I’m just going to discharge the pack down to 35v - to see if the BMS will charge from that voltage. Maybe something’s open-circuit?

The logic is very simple: As soon as the first cell reaches 4.2v, the charging is switched off. To get 42v, all cells need to be balanced at 4.2v.

When you charged through the output, that was a really dangerous thing to do to an unbalanced battery because some cells would have to be way above the 4.2v maximum. You might have even damaged them.

Before going any further, you need to get all the cells to the same voltage by either charging the low ones individually or discharging the high ones down to the lowest.

If it doesn't stay balanced, it's either knackered or not suitable.

 

[Nealh]

You guys who like playing with fire and this is not the only time it has been mentioned or tried on the forum. If you want to carry out stoopid techniques then carry on, if you are lucky you may have the kudos of being the forums first recognised battery fire.

The BMS is fitted for a reason and that is for safety , they though do not prevent stoopidity with wrongful charge techniques.

 

[Cisco-man]

Fully accept what you say re safety. I work on my batteries using an ammo box, and frankly I charged via the output by mistake.
So the mechanism coming into play here, is a cell group has reached 4.2 - and the BMS then shuts down - cool. I assure you I am no fool - I’m a 50 years plus fully qualified engineer in both mechanical and electrical systems. Everything I do is with total safety in mind - but we all make mistakes!

 

[Nealh]

It is not only the 4.2v cell group that is the issue, the pair at 4.1v will also act the same . If one were to discharge the 4.2v and not the 4.1 v then once they also reach 4.2v ahead of the othe rcell groups , said BMS will cease charging as any balance will still be too far out.
One will need to deal with four out of the ten cell groups to get them within 0.1v - 0.2v of each other.

 

[Cisco-man]

Hi Neal. I’ve got load resistors pulling down the “high” banks at the mo. I should have them all within 0.1v by late tomorrow. I can then try a pack charge again.
I’m still assuming the pack is not viable because something put it out of balance (partial charging maybe). Once I get it in balance I’ll see how it fairs over a few cycles to see if it’s worth the effort. For me, it’s good learning stuff though.

 

[Nealh]

It's only a 2p pack so depends on the cells used and the previous users current draw dependant on the cell discharge rating.
It may be a bit abused/knackered or as you say may not have been balance charged properly , only time will tell once one can get it to balance charge.

I use 12s 2p self welded packs but only use higher current rated 20a rated or higher cells .
Currently use HG2 and P42a in 2p configs.
HG2 are great 25 miles from a 44v 6ah bottle battery , actual range would be far more if discharged to LVC so nearer 32 - 36 miles.
I recharge at circa 44v or 3.65v per cell.
Next cell I would like to try is P28A to compare them to the HG2 but it may be a while yet before that occurs.

 

[Cisco-man]

So the “charge completed” light on the charger is not exactly a charger function - even though it’s on the charger. It’s triggered by a cell group reaching 4.2v, the BMS therefore turning off it’s charge input, and when the charger senses no load it turns the green light on? Hence the green light is on if you power up the charger without a pack attached.

 

[Nealh]

So the “charge completed” light on the charger is not exactly a charger function - even though it’s on the charger. It’s triggered by a cell group reaching 4.2v, the BMS therefore turning off it’s charge input, and when the charger senses no load it turns the green light on? Hence the green light is on if you power up the charger without a pack attached.

Yes exactly.

 

[AntonyC]

Wear can be down to age or usage, the cell resistances and self discharge increase and their capacities reduce. The BMS optimises use of available capacity by balancing the voltages but it can't reverse actual capacity loss, so an old pack ought to charge to full but discharge too quickly in use, which balancing can't fix, and sag badly under load.

3.6 to 4.2 seems an unusually large imbalance and there's a hint in "I’ve inherited". Abuse is another wear mechanism and mistakenly bypassing the BMS when charging may have snapped this battery's walking stick - not a "success"

Like you I store batteries in vented metal boxes, although I sometimes wonder why. If one catches fire that's _in the order of_ 20 cells x 60kJ x (1 + 3 chemical) / 3 mins = a 27kW fire. It's not like I could pick up the box or anything if I were around at the time but it might add a few seconds to get out.

 

[Cisco-man]

Like you I store batteries in vented metal boxes, although I sometimes wonder why. If one catches fire that's _in the order of_ 20 cells x 60kJ x (1 + 3 chemical) / 3 mins = a 27kW fire. It's not like I could pick up the box or anything if I were around at the time but it might add a few seconds to get out.

I’m trusting that I’ve vented the ammo boxes enough so that IF a pack ever went up, then I’ve got it so the gases can escape rapidly enough not to explode the ammo box! I keep them in the garage with nothing too combustible close by. I must try and not to be somewhere that people can get rid of a battery to! I use the ones I’ve got to take camping - for lighting, radio, and charging my phone/ipad with.

 

[Cisco-man]

… and before you say it - No, I don’t keep the packs in the tent (or car)!

 

[Cisco-man]

I’ve taken two things away from my attempt to resurrect this defunct battery pack.

1. Once the “charge complete” indicator comes on on the charger, then absolutely no further charging is going to take place (not even a trickle).

2. The final voltage that the pack achieves when the ‘charge complete” comes on, is a clear indication as to the balance (and probably) health of the pack.

Obvious to the experts here, but not so clear to us learners!

 

[AntonyC]

1. Many BMS start cell balancing when the battery's nearly full so it's useful to leave a 'full' battery on charge for an hour or two every few weeks, some say more often.

2. Overall battery voltage doesn't say much about balance because some cells can be under and others over (4.2V is chosen for safety, it's a bit under the max that's possible). Nor does it tell you about health: it's as though each cell's a cup that shrinks as it ages, it stores less energy and provides a weaker flow (ahem) but it still fills to the brim (4.2V) like a new one.

 

[Sturmey]

1. Once the “charge complete” indicator comes on on the charger, then absolutely no further charging is going to take place (not even a trickle).

Some/many chargers continue to trickle charge even though the green light is on, so I think its best to plug out when charging is complete.

 

[Ghost1951]

Some/many chargers continue to trickle charge even though the green light is on, so I think its best to plug out when charging is complete.

Mine balances the cells when the green light is on. I have checked this empirically by measuring the voltage of individual cells while it is plugged in to a charger with a green light on it. This can take a good number of hours. Now and then - maybe once a month, I leave mine plugged in with a green light and the cells balance up. I only do this when I am supervising it and from time to time just grabbing a feel of the battery case to see if any part of it is at all warm.

Nobody wants a battery fire anywhere near their house, let alone in it. That said. Mine seems to benefit from being left on charge occasionally, but I never leave it alone and unattended like this.

 

[Cisco-man]

Mine balances the cells when the green light is on. I have checked this empirically by measuring the voltage of individual cells while it is plugged in to a charger with a green light on it. This can take a good number of hours. Now and then - maybe once a month, I leave mine plugged in with a green light and the cells balance up. I only do this when I am supervising it and from time to time just grabbing a feel of the battery case to see if any part of it is at all warm.

Nobody wants a battery fire anywhere near their house, let alone in it. That said. Mine seems to benefit from being left on charge occasionally, but I never leave it alone and unattended like this.

OK, but this must be a function of the BMS allowing balancing to continue once a cell-group has reached 4.2v. The charger isn’t in control, the BMS is. Admittedly I’ve only had an evenings read of bms design though, but I still believe that the charge complete light only comes on when there is no demand from the bms.

 

[Ghost1951]

OK, but this must be a function of the BMS allowing balancing to continue once a cell-group has reached 4.2v. The charger isn’t in control, the BMS is. Admittedly I’ve only had an evenings read of bms design though, but I still believe that the charge complete light only comes on when there is no demand from the bms.

Yes - the bms drains high cells and the charger seems to trickle so they all get back to the same voltage.