Battery Balance, or weak cell?

[Martijn970]

Hello experts!
I am new. Do you think with the below data that my next step in determining why my LVC keeps tripping the BMS while biking using my old battery I picked out of storage is because of a damaged/old cell, or because it needs to balance longer? Should I let it balance longer before riding again? Should I let the battery sit off the charger and measure again tomorrow to see if any cells drop significantly?
Battery: Panasonic 37v 9Ah, from a e-Moto Velocity 2.0.
BMS: KL3688CL3
Volt at battery output: 42V
Charger output: 42.3V
Days on charger: initially 2 days, then a bike ride where system shut down several times under load but rode 12 miles, and then another day on the charger (24hrs). The following is the result after the 2nd session on the charger.
Cell volts:
1: 4.27
2: 4.27
3: 4.17
4: 4.17
5: 3.89
6: 4.27
7: 4.27

Any input is welcome, I want to see if I can fix it vs. spending hundreds... obviously group 5 is pretty low...

 

[E-HP]

How are you measuring the voltages; meaning do connect to the BMS via bluetooth, or is the pack opened up?
If you have access to the balance leads, you could charge just that one group and see if it holds a charge or can charge at all. You need a 4.2V supply.

EDIT: I would ride it around the block to get the other cells down to 4.2, and see if you can get groups 3 and 4 up to 4.2 as well.

 

[DogDipstick]

7 cells? You said it iis a " 37v" batery? Charges at 42v?

36v is 10s... and 42v fully charged. Not 7s?

7s is 25.2v... and 29.4v fully charged.

So is is a 7s battery or a 10s?

 

[Martijn970]

I was wondering about that too, but the large white BMS connector that I measure through only has 9 wires to it to measure from. I measure by taking the battery off the charger, open up the case, get to the BMS without removing the shrink-wrap, and measure across the connector. I am not sure how to measure additional cells if there are any. Also don't know how to individually bring up the bottom 3 groups? Can that be done through BMS connector?View attachment 1

 

[Martijn970]

7 cells? You said it iis a " 37v" batery? Charges at 42v?

36v is 10s... and 42v fully charged. Not 7s?

7s is 25.2v... and 29.4v fully charged.

So is is a 7s battery or a 10s?

I was wondering about that too, but the large white BMS connector that I measure through only has 9 wires to it to measure from. I measure by taking the battery off the charger, open up the case, get to the BMS without removing the shrink-wrap, and measure across the connector. I am not sure how to measure additional cells if there are any. Also don't know how to individually bring up the bottom 3 groups? Can that be done through BMS connector?

 

[DogDipstick]

Ok. I think it is a 10s. Those are EXACTLY the pictures we need.

Gimmie another straight on, so I can trace the PCB tracks out. This might help me diagnose the circuitry more. I need to see the board head on..... Esp. the writing RIGHT above the big header.


.....there are 10 bleed resistors on that board ( and room for 12,: two are unpopulated) .

10s, 3.7v nominal cells, 37v nominal battery. 42v charged. About 30-32v completely discharged.

That would explain the output voltage of both the charger and the battery ( 42v full) ... and they are populated. Looks like a BMS capable of 12s, but only setup for 10s. I see the ten bleeders right there. You must be missing a plug to the batteries string tap(s)... somewhere in there. They are not all on the same header.. YES the one you have unplugged ( rainbow) is PART of the whole TAP input for the system...... But you are not seeing some. I betcha there are 2 wires you are not measuring but should be. What plugs into the header right next to the rainbow header? Those two little pins next to the "sw"?

maybe the taps ( for the first and last cells in the string ) are taken from the main leads.... ? IDK. We need some help from some OG BMS people to input opinions.


YES you can absolutely level charges slowly through the BMS header manually.... BUT you must know what pins to charge/discharge. We have to find, absolutely, the 11 pins that goto the taps. Verify the (nominal) 3.6v - 7.2v - 10.8v-14.4v....~ 36v... The first will be the system ground and the last will be the 36v... ( when @ nominal V.. not when charged.. adjust accordingly..)


Some references: THIS example shows this BMS in a 7s config.. different from yours... YOURS is a 10s. Definitely a 10s.
https://endless-sphere.com/forums/viewtopic.php?t=59634

 

[Martijn970]

Ok. I think it is a 10s. Those are EXACTLY the pictures we need.

Gimmie another straight on, so I can trace the PCB tracks out. This might help me diagnose the circuitry more. I need to see the board head on..... Esp. the writing RIGHT above the big header.


.....there are 10 bleed resistors on that board ( and room for 12,: two are unpopulated) .

10s, 3.7v nominal cells, 37v nominal battery. 42v charged. About 30-32v completely discharged.

That would explain the output voltage of both the charger and the battery ( 42v full) ... and they are populated. Looks like a BMS capable of 12s, but only setup for 10s. I see the ten bleeders right there. You must be missing a plug to the batteries string tap(s)... somewhere in there. They are not all on the same header.. YES the one you have unplugged ( rainbow) is PART of the whole TAP input for the system...... But you are not seeing some. I betcha there are 2 wires you are not measuring but should be. What plugs into the header right next to the rainbow header? Those two little pins next to the "sw"?

maybe the taps ( for the first and last cells in the string ) are taken from the main leads.... ? IDK. We need some help from some OG BMS people to input opinions.


YES you can absolutely level charges slowly through the BMS header manually.... BUT you must know what pins to charge/discharge. We have to find, absolutely, the 11 pins that goto the taps. Verify the (nominal) 3.6v - 7.2v - 10.8v-14.4v....~ 36v... The first will be the system ground and the last will be the 36v... ( when @ nominal V.. not when charged.. adjust accordingly..)


Some references: THIS example shows this BMS in a 7s config.. different from yours... YOURS is a 10s. Definitely a 10s.
https://endless-sphere.com/forums/viewtopic.php?t=59634

Thank you, I will open it back up. I think/wonder if the black and yellow plug are the additional cell groups. If I find an old cell phone charger, how do I use that/connect that to the BMS header leads to bring them up manually? Sorry, I know enough to get me in trouble, but not enough to not need assistance! EDIT: by the way,, nothing plugs into the empty spot next to the BMS main plug.

 

[E-HP]

did you ride it around the block yet? you're basically killing the good/overcharged cells the longer you leave them overcharged. pretty soon you won't have a pack to salvage. first things first, unless it's just a learning experience for you.

 

[Martijn970]

Here is another picture I took last night. Will open it up again after work today.

 

[Martijn970]

did you ride it around the block yet? you're basically killing the good/overcharged cells the longer you leave them overcharged. pretty soon you won't have a pack to salvage. first things first, unless it's just a learning experience for you.

Trying to salvage. Have not ridden it yet. Was hoping that the BMS balancing would prevent overcharging?

 

[DogDipstick]

Hey... Out of curiosity, doyou see those "5 legged SOT25 IC chips?" .. that there are 10 of them? They look like THIS ( link) chip here: The writing will be TINY but a good phone may be able to take that picture. The writing may be half a mm or less.



Could you tell me what is written on top of them? Perchance please? Just out of curiosity, but it may be relevant to what we study here. Tell me if there is anything that looks like a " D" or a "W" in particular.


They look similar to this: What is written on them? Or take a good pic, magnified, I know its tough.... But we can figure this out. For sure. If you could, I thank you in advance.


https://www.google.com/search?q=SOT25&client=firefox-b-1-d&sxsrf=AOaemvK03p78VWELjPqeLHi9zKs-ahvlow:1639679668642&source=lnms&tbm=isch&sa=X&ved=2ahUKEwjH7cbh-uj0AhVyUt8KHeV5Dv0Q_AUoAXoECAIQAw&biw=1280&bih=909&dpr=1

 

[Martijn970]

Hey... Out of curiosity, doyou see those "5 legged SOT25 IC chips?" .. that there are 10 of them? They look like THIS ( link) chip here: The writing will be TINY but a good phone may be able to take that picture. The writing may be half a mm or less.

SOT25.JPG

Could you tell me what is written on top of them? Perchance please? Just out of curiosity, but it may be relevant to what we study here. Tell me if there is anything that looks like a " D" or a "W" in particular.


They look similar to this: What is written on them? Or take a good pic, magnified, I know its tough.... But we can figure this out. For sure. If you could, I thank you in advance.


https://www.google.com/search?q=SOT25&client=firefox-b-1-d&sxsrf=AOaemvK03p78VWELjPqeLHi9zKs-ahvlow:1639679668642&source=lnms&tbm=isch&sa=X&ved=2ahUKEwjH7cbh-uj0AhVyUt8KHeV5Dv0Q_AUoAXoECAIQAw&biw=1280&bih=909&dpr=1

It looks like it does have a W on it. Maybe WLGS(??) I can take a better picture tonight but this is from blowing up the picture on my phone and changing the properties on it:

 

[DogDipstick]

I ask because there is a microchip designed for single cell BMS called the DWO1 chip... Same look. The chip is notorious for failing when the chip is used in many string BMS setups.. and causes a cascade failure. I will link a explanation from an EE that described it here on Endless sphere a few years ago. I will find that thread if you wish.

They were in the hoverboard packs,and failed regularly.
This is a typical DWO1 datasheet... Only meant for one cell governing. There are many chips that fit that footprint, so If we cannot find it, do not worry to much, i may just be overly cautions and curious which chip they use.
https://cdn.sparkfun.com/assets/learn_tutorials/2/5/1/DW01-P_DataSheet_V10.pdf

Thanks for humoring me.

 

[Martijn970]

I ask because there is a microchip designed for single cell BMS called the DWO1 chip... Same look. The chip is notorious for failing when the chip is used in many string BMS setups.. and causes a cascade failure. I will link a explanation from an EE that described it here on Endless sphere a few years ago. I will find that thread if you wish.

They were in the hoverboard packs,and failed regularly.
This is a typical DWO1 datasheet... Only meant for one cell governing. There are many chips that fit that footprint, so If we cannot find it, do not worry to much, i may just be overly cautions and curious which chip they use.
https://cdn.sparkfun.com/assets/learn_tutorials/2/5/1/DW01-P_DataSheet_V10.pdf

Thanks for humoring me.

Here is some more data:
Took it off the charger, and measured:
1....4.17
2....4.17
3....4.18
4....4.09
5....4.16
6....3.86
7....4.18
8....4.18

Then went on a short ride where the battery system shut down several times under load.

Then measured again, 39.8v total
1...4.10
2....4.10
3.....4.09
4.....3.92
5.....3.99
6.....3.75
7....4.11
8....4.11

The black and red connectors (with 2 red and 2 black wired) get 16.4v and 16.2v between each black and red wire. These are the only other wires that I can see going to the battery. The wires on the other side is the main battery power leads, capacity light indicator and switches.
Going to put it on the charger again I guess, not sure what else to do??!! I am thinking this pack is too old to do much with.

Here are some pictures:

 

[Martijn970]

did you ride it around the block yet? you're basically killing the good/overcharged cells the longer you leave them overcharged. pretty soon you won't have a pack to salvage. first things first, unless it's just a learning experience for you.

When I took it off the charger after work, the values had gone down to acceptable range. But I still went on a short ride to see. Data is posted! Do you have any next steps or know where the leads are for group 9 and 10 (or whatever number they are in the sequence..)

 

[E-HP]

did you ride it around the block yet? you're basically killing the good/overcharged cells the longer you leave them overcharged. pretty soon you won't have a pack to salvage. first things first, unless it's just a learning experience for you.

When I took it off the charger after work, the values had gone down to acceptable range. But I still went on a short ride to see. Data is posted! Do you have any next steps or know where the leads are for group 9 and 10 (or whatever number they are in the sequence..)

How are you measuring?

 

[Martijn970]

did you ride it around the block yet? you're basically killing the good/overcharged cells the longer you leave them overcharged. pretty soon you won't have a pack to salvage. first things first, unless it's just a learning experience for you.

When I took it off the charger after work, the values had gone down to acceptable range. But I still went on a short ride to see. Data is posted! Do you have any next steps or know where the leads are for group 9 and 10 (or whatever number they are in the sequence..)

How are you measuring?

At the BMS connector, off the charger, battery switched off, and with a multimeter on the 20v setting.

 

[E-HP]

How are you measuring?

At the BMS connector, off the charger, battery switched off, and with a multimeter on the 20v setting.

Those are the same points where you'd apply charging voltage for those groups. You need a low current 4.2V supply and you can insert small gauge wire into the connector and charge, then monitor/measure until the group match the others. If you don't have a supply, you could use something like this. It's only 1 amp, so will take a while, but the balance leads can probably only carry a couple of amps anyway:
https://www.ebay.com/itm/124264445619?epid=16039850959&hash=item1ceebcf6b3:g:zVkAAOSw2GpfEim8

I bet if you carefully remove the connector housing, you might even have pins that can be inserted into the connector. Make sure you match the polarity.

 

[Martijn970]

How are you measuring?

At the BMS connector, off the charger, battery switched off, and with a multimeter on the 20v setting.

Those are the same points where you'd apply charging voltage for those groups. You need a low current 4.2V supply and you can insert small gauge wire into the connector and charge, then monitor/measure until the group match the others. If you don't have a supply, you could use something like this. It's only 1 amp, so will take a while, but the balance leads can probably only carry a couple of amps anyway:
https://www.ebay.com/itm/124264445619?epid=16039850959&hash=item1ceebcf6b3:g:zVkAAOSw2GpfEim8
lipo charger.jpg
I bet if you carefully remove the connector housing, you might even have pins that can be inserted into the connector. Make sure you match the polarity.

Thank you. On that ebay charger thing, are you saying I could just use some wire and stick it into the ebay charger ends, and the other ends of the wires on the balance charger leads that need to come up? Any idea what is going on with the remaining 2 cell groups? Would be nice to check them too to see if they are low/bad....

 

[E-HP]

Thank you. On that ebay charger thing, are you saying I could just use some wire and stick it into the ebay charger ends, and the other ends of the wires on the balance charger leads that need to come up? Any idea what is going on with the remaining 2 cell groups? Would be nice to check them too to see if they are low/bad....

I'm assuming that since the ebay charger is for charging 1S lipos, it has a male connector pins in that black housing. So if you are able get the pins out, they may be able to plug into the balance connector at the right locations. I have no clue what's going on with your BMS. Isn't the typical BMS.

 

[docw009]

[

Here is some more data:
Took it off the charger, and measured:
1....4.17
2....4.17
3....4.18
4....4.09
5....4.16
6....3.86
7....4.18
8....4.18

Then went on a short ride where the battery system shut down several times under load.

Then measured again, 39.8v total
1...4.10
2....4.10
3.....4.09
4.....3.92
5.....3.99
6.....3.75
7....4.11
8....4.11

Subtotal = 32,2V. Subtract from 39.8V leaves 7.6V for cell groups 9 and 10 .

There;s nothing on those eight groups that is close to the 3.0V LVC where a BMS trips. Your low voltage must be on one of the two cell groups you cannot check via the one balance connector. The eight voltages above add up to 32.17V, Total voltage is 39.8V, so the two unmeasured groups will total 7.6V. For the sake of argument, lets say one is OK, so it will be 4.1V, leaving the lowest one at 3.5V,

While 3.5V is still above 3.0V, perhaps voltage sag is pulling that group below LVC. Old batteries will do that, The puzzle is that the balance wires for 9 and 10 are missing. It seems to me that the bare spot to the side of your long white connector would have held it. Will a BMS work if two wires are missing? I dunno,

Let's assume groups 9 and 10 are somehow connected to the BMS. If you have alligator clips, put some pins into them and see if you can carefully probe the BMS solder pads directly. I'm guessing these are the last two cells, but you can figure it out. Don't short out a group with the pins.

It does appear to be a balance BMS, but it cannot balance that one group. Charging that group manually, assuming it exists, would be hard, unless you unwrap the battery.

Good luck.