Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Batteries, Chargers, and Battery Management Systems.
casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 08 2012 7:40pm

Hello.

I bought a "48V 15Ah Li-Ion Alloy Shell EBike Battery Pack" from BMSBattery on March 2012 as I blogged here. One month before I also bought the 10Ah version and I am being using both but much more the 15Ah one.

The 15Ah battery is dead. When I were using it before the dead, I verified that it lasted much less than usual and after being a bit discharged it just turned off and didn't worked any more. With the charger connected, it charges a bit and quick stops the charger. The motor works if charger is connected to battery.

Any ideas of to were should I look when open the battery? -- I have some electronics knowledge and tools, for example, I own an oscilloscope.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

User avatar
amberwolf   100 GW

100 GW
Posts: 25667
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by amberwolf » Nov 08 2012 7:51pm

Bad cell or bad connection or bad balance tap connection. See all the other battery repair threads; the questions and answers are basicaly the same. ;)

User avatar
dogman dan   100 GW

100 GW
Posts: 34455
Joined: May 17 2008 12:53pm
Location: Las Cruces New Mexico USA

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dogman dan » Nov 09 2012 6:50am

Yeah, first thing is to confirm that the plug between the charger and the battery is ok. Bent wire strands break and then the wire is barely contacting the pin in the plug. The green light on the charger means it's done, or it's disconnected / unplugged. A flaky plug might charge for 5 min, then break the connection.

If that's not the problem, then dig up a battery problem thread, and follow Dnmuns instructions for a good diagnosis of which cells are either bad, or merely undercharged.

Once you get into repairs or rehabilitation of a multiple cell battery pack, a good method for single cell charging can be worth it's weight in gold. My personal favorite is an RC charger. Even the cheapest 50w ones will do a very good single cell charge. They charge lipo, lifepo4, nimh, nicad, lead. Very handy for an Ebike enthusiast.

Sometimes a battery just gets way out of balance, and the bms tries it's best but will take a week or so on the charger to catch up. A single cell charge of the low cell can be much quicker. But just using a 4v power supply is risky.

If you take your batteries off the charger the minuite the charger turns green, you aren't getting your batteries balanced. Start leaving them on the charger overnight if you haven't been doing that.
Last edited by dogman dan on Nov 09 2012 2:52pm, edited 1 time in total.

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dnmun » Nov 09 2012 8:29am

you should open the battery and take a picture and also put the battery on the charger and measure the individual cell voltages while it is charging.

measure voltage of each cell and list them from #1-12. measure the voltage on the BMS where the sense wires connect to the BMS. measure while it is on the charger and the charger light is green as you said it goes to right away.

casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 10 2012 7:03am

Thank you all. I will do it and report in the hope to help others.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dnmun » Nov 10 2012 9:18am

no, it is to help you figure out how to repair your pack.

no hope is needed, just systematically identify what is wrong in order to repair it.

first thing is just the measurement of the cell voltages while it is charging so we can see how the BMS is performing and if there is a single cell that is causing the charge to be terminated early.

it would help to have a picture and the manufacturer of your battery too.

casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 10 2012 11:45am

dnmun wrote:first thing is just the measurement of the cell voltages while it is charging so we can see how the BMS is performing and if there is a single cell that is causing the charge to be terminated early.

it would help to have a picture and the manufacturer of your battery too.
I hope to have time to test/repair the battery tomorrow. The battery is "48V 15Ah Li-Ion Alloy Shell EBike Battery Pack" from BMSBattery: http://www.bmsbattery.com/48v/403-48v-1 ... -pack.html

Image
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dnmun » Nov 10 2012 11:59am

no rush, but you can fix it, i am sure.

you will be able to get replacement cells from BMS battery or greenbike kit once you open it up and measure.

casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 11 2012 12:22pm

The battery measures 0 volts at his output. And the measures on batteries connection (BMS):

B2 - B1 = 3.63v
B3 - B2 = 3.78v
B4 - B5 = 3.69v
B5 - B6 = 4.12v (the same value for the rest)

Looking at this values, what can be happen?

Here are the pictures I took:
48V_15Ah_battery-01.jpg
48V_15Ah_battery-02.jpg
48V_15Ah_battery-03.jpg
48V_15Ah_battery-04.jpg
48V_15Ah_battery-05.jpg
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

SamTexas   100 MW

100 MW
Posts: 2714
Joined: Dec 29 2010 5:34am
Location: Houston, Texas

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by SamTexas » Nov 11 2012 1:30pm

You're lucky! Your pack can be easily revived.

Your BMS was drawing power from the first 3 cells. BMS = Battery Murdering System. Anyway, you caught it early enough, so just recharge the first 3 cells.

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dnmun » Nov 11 2012 1:47pm

how can you tell it takes the circuit current from the first three cells?

can you measure the voltage on the gate of the mosfets? put the black probe on the wire that runs to the battery where it comes off the pcb next to those four shunt resistors and put the red probe of the voltmeter on the leftmost leg of the mosfets.

casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 11 2012 6:20pm

After let the battery on charger some hours (charger always with fan off and with green LED), the cells seem to be more charged however battery still "dead".

While before B5 - B6 = 4.12v (the same value for the rest) now B5 - B6 = 4.27v (the same value for the rest). Also the others that were uncharged have now more ~0.1V. What this means?
dnmun wrote:can you measure the voltage on the gate of the mosfets? put the black probe on the wire that runs to the battery where it comes off the pcb next to those four shunt resistors and put the red probe of the voltmeter on the leftmost leg of the mosfets.
0V on all leftmost leg of the mosfets. That means they are "off" and hence the battery is dead?
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 11 2012 6:28pm

dogman wrote:The green light on the charger means it's done, or it's disconnected / unplugged.

If you take your batteries off the charger the minuite the charger turns green, you aren't getting your batteries balanced. Start leaving them on the charger overnight if you haven't been doing that.
I thought the green LED means "it's done/charged", hence I disconnected after let the all night the charger connected.

Maybe you are right about not letting the charger balance the cells... who knows? (the charger didn't come with a manual).
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

SamTexas   100 MW

100 MW
Posts: 2714
Joined: Dec 29 2010 5:34am
Location: Houston, Texas

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by SamTexas » Nov 11 2012 8:04pm

One more try.
SamTexas wrote:You're lucky! Your pack can be easily revived.

Your BMS was drawing power from the first 3 cells. BMS = Battery Murdering System. Anyway, you caught it early enough, so just recharge the first 3 cells.

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dnmun » Nov 11 2012 9:28pm

your pack has some cells that are already fully charged so the BMS stops the charge while those cells drain some charge off and then the charging mosfet will turn back on and then it will charge until the high ones go to the HVC and then shut off and keep cycling until the pack is fully charged and balanced. can you put an ammeter in the charging circuit while you are charging? use the 10A scale and move the probe lead to the 10A plug. then watch the current.

you can do like he said also and put a 12.5V current source on those three cells to charge them up but you don't wanna overcharge them but i think those are the konion type limn2o4 chemistry. so get them all to the 4.1V level and then see if it will work. i have a little 12V power supply i use to charge three lifepo4 at a time. i can adjust the voltage. up to 17.4V actually.

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dnmun » Nov 11 2012 9:36pm

if you don't wanna charge up the three low cells, 1,2,3 then you can drain the high cells. there are 10 at 4.12V so 41V would charge up a 36V SLA pack if you have one laying around. drain the high cells down and see when the charger turns on and keep discharging the high cells until they are the same voltage as the others and then bulk charge them up with the charger if the mosfet will turn on then. and you can drain charge off with light bulbs or space heaters or big power resistors if you happen to have some laying around you can jumper together.

casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 12 2012 6:03am

dnmun wrote:your pack has some cells that are already fully charged so the BMS stops the charge while those cells drain some charge off and then the charging mosfet will turn back on and then it will charge until the high ones go to the HVC and then shut off and keep cycling until the pack is fully charged and balanced.
So the BMS is doing it's work, right?

I just found (yesterday) a nice guy here on my city that have experience and knowledge about electric scooters and electric bicycles. He builds his own battery packs using BMSBattery BMSs, etc. He told me that the charger will turn green LED when charge of some cells but will keep charging the others (balance) -- seems that is what is happening now (the charger and battery doesn't have a manual :-(

Here is pictures of this guy:
motor_repair-01.jpg
motor_repair-03.jpg
SamTexas wrote:You're lucky! Your pack can be easily revived.

Your BMS was drawing power from the first 3 cells. BMS = Battery Murdering System. Anyway, you caught it early enough, so just recharge the first 3 cells.
SamTexas, you mean that BMS was the problem? can you please explain your idea?
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

SamTexas   100 MW

100 MW
Posts: 2714
Joined: Dec 29 2010 5:34am
Location: Houston, Texas

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by SamTexas » Nov 12 2012 7:35am

casainho wrote:
SamTexas wrote:You're lucky! Your pack can be easily revived.

Your BMS was drawing power from the first 3 cells. BMS = Battery Murdering System. Anyway, you caught it early enough, so just recharge the first 3 cells.
SamTexas, you mean that BMS was the problem? can you please explain your idea?
I'm glad you're finally listening. Here's my take on your situation:
Your battery was left unused/untouched for quite a while. But the BMS was still drawing power from the first 3 cells of your pack (braindead design common to most BMSs). Because of that:
1) a huge imbalance was created (3.6x-3.7x for first 3 cells and 4.12 for the rest)
2) the combine voltage of the first 3 cells (~11V) has dropped too low. This is where the BMS draw power from.

One or both of the above results (most likely is the huge imbalance) is preventing the BMS to operate, so everything is shut down.

Here's what I would do:
1) unplug the BMS
2) use a cell phone charger (~5v) to individually charge each of the first 3 cells to around 4.12V (same voltage as the last 11 cells). It will take at least 5 hrs for each cell.
3) plug in the BMS

Yes, the BMS is the root cause of your problem (by design). But it should still be fully functional. Your pack, including the first 3 cells, is ok too.
Good luck and have fun.
Last edited by SamTexas on Nov 12 2012 7:50am, edited 1 time in total.

casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 12 2012 7:47am

SamTexas wrote:
casainho wrote: I'm glad you're finally listening.
I were always listening, I did read all messages and I take in account all info.

Great all the information you shared. And I think I understand that BMS circuit uses is own energy from the cells and can enter in a "deadlock". Any way, the only way to leave the lock is to charge the cells -- right now seems that the BMS is able to charge them, I will need to wait and see. The other way, the guy I meet from my city have a lot of tools including an individual charger for this situations, I will use that charger (but I also own a lab power supply, I cal always limit the current and set the voltage to charge each cell).

Thank you. I will report later.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

SamTexas   100 MW

100 MW
Posts: 2714
Joined: Dec 29 2010 5:34am
Location: Houston, Texas

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by SamTexas » Nov 12 2012 7:57am

casainho wrote: right now seems that the BMS is able to charge them, I will need to wait and see.
...
Thank you. I will report later.
No, the BMS is NOT working at this time as evidenced by your earlier post: The last 11 cells are now up to 4.27V, above the maximum 4.20V. You need to stop charging immediately before you kill the last 11 cells. You also need to discharge the last 11 cells back down to 4.20V or lower.

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dnmun » Nov 12 2012 8:58am

casainho wrote:
dnmun wrote:your pack has some cells that are already fully charged so the BMS stops the charge while those cells drain some charge off and then the charging mosfet will turn back on and then it will charge until the high ones go to the HVC and then shut off and keep cycling until the pack is fully charged and balanced.
So the BMS is doing it's work, right?

I just found (yesterday) a nice guy here on my city that have experience and knowledge about electric scooters and electric bicycles. He builds his own battery packs using BMSBattery BMSs, etc. He told me that the charger will turn green LED when charge of some cells but will keep charging the others (balance) -- seems that is what is happening now (the charger and battery doesn't have a manual :-(


i really have no idea how sam texas knows that the circuit current is taken off the first three cells. it would help if he could explain that since i cannot see where the circuit current comes off the third cell from looking at the pictures of the BMS.

he may be familiar with this BMS, so he knows it already but i see no evidence of that myself. i assumed it takes the circuit current from the top of the pack like the other BMSs do. the ping signalab is the only BMS i have seen that takes the circuit current from the top 4 cells, not the top three and it is the only design i have ever seen supply current from that level. maybe there is some evidence in the traces i did not see.

your local friend is right for the lifepo4 chemistry in that the charger continues to supply the balancing current even after the charging light has turned green. but if the charging mosfet is turned off because the gate is 0V then even that balancing current cannot enter the pack.

the charging mosfet is turned off i think because the high cells have reached their HVC so they have to drain down in order to allow the charging mosfet to turn back on again. 0V on the gate means the mosfet is turned off by the BMS logic in those ICs on the side over there. but they have 100 ohm shunt resistors so the balancing current is tiny, only about 40mA so it takes forever for the low cells to charge up to a fully charged state. these low cells are the reason the BMS shuts off when you use the pack since they are not charged up and when charge is removed they rapidly go to LVC and shut down the BMS again.

i think the local guy has the best chance of making this clear to you but i think your pack is still functional, just discharged and unbalanced. you can drain the high cells down with a load like i said or use a 12V charger or power supply to charge up the first three cells. when the first three cells are charged to the same level as the others then the pack should begin to balance on the balancing current but these are different from the lifepo4, these are the limn2o4 chemistry and they normally are self balancing when charging.

SamTexas   100 MW

100 MW
Posts: 2714
Joined: Dec 29 2010 5:34am
Location: Houston, Texas

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by SamTexas » Nov 12 2012 9:53am

dnnum: No I have never worked or touched this particular BMS before. I did have experience with another typical no-name chinese BMS. I was seeing the same problem the OP saw. After disassembling it, I also found that the voltage of the first three cell was much lower than the rest. So after charging up the low three cells, the BMS and the battery worked again. Then I disconnected each of the sense wire one at a time to see where the BMS got its power. As expected, it was from the 3 low cells.

What difference does it make
1) whether the BMS is drawing power from the first or last 3 (or 4) cells?
2) whether the BMS is using (3s)V or (4s)V nominal?
As far I as know, there is no standard dictating these requirements.

Isn't too much of a coincidence for the first 3 cells to have the same low voltage while the rest of cells are at the same high voltage?

BTW, this is the battery/BMS I was working with:
http://endless-sphere.com/forums/viewto ... 27#p591795

casainho   100 MW

100 MW
Posts: 2799
Joined: Feb 14 2011 2:43pm

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by casainho » Nov 12 2012 9:57am

SamTexas seems to have reason. Now the cells go at 4.3V and the others first 4 cells to 3.7V. I stopped to use the battery charger and I will now charge each individual cell.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by dnmun » Nov 12 2012 10:04am

yes, that explains it. i just could not see it on the traces. you can charge just the first three up to the 4.1-4.2V level with a regular 12V charger if you don't have a single cell charger. use a jumper wire with alligator clips on each end to connect your charger from the bottom of the #1 cell to the top of the #3 cell. you don't have to use the other chargers or use a single cell charger since those three will charge up together then.

999zip999   100 GW

100 GW
Posts: 8488
Joined: Jan 10 2010 1:40pm
Location: Dana Point So. Cal

Re: Ideas to repair a dead 48V 10Ah Li-Ion Battery Pack?

Post by 999zip999 » Nov 12 2012 11:35am

I think there are telling you to drain the high cells above 4.2v NOW. And it would help to charge the low cells to help balance the pack as not to wait for a slow balance of the bms.

Post Reply