Battery will not discharge evenly

LWGT

1 mW
Joined
Jul 14, 2017
Messages
15
I recently did my first ebike build. I built my own battery with books from ebikeschool.com. I built a 20s 12p. It charged to a full 84 volts. I tested the bms sensor wires many times. They started at 4.2 volts and incremented 4.2 volts all the way to 84 volts. I thought that my battery was good to go until I used it. It was right before winter last year and I had time for one test drive before I parked it for the winter. It ran really good. I had a hard time keeping the front tire on the ground with the sabvoton 72150 and the cromotor combination. This is where things go south. For some reason the motor quit running when it hit 79.8 volts. I had to stop testing due to snow so I still have too many volts in the battery for storing. So I hooked up my charger but never plugged it into the wall to slowly drain the battery. I kept my eye on the voltage and when it got down to 69 volts I took the charger off to stop the drain and parked it for 3 months. Then, 3 months later, I recharged the battery. It charged to 78 volts my charger's green light turned on and it wouldn't charge anymore telling me that it was a full charge. Then I rechecked the battery. On the ground end of my battery all 12 batteries read 0 volts. So now my battery is really only running on a 19s 12p. I had to ride it to discharge the batteries a little and so I could check it again. It quit running again. This time at 70 volts. I have the cutoff set at 55 volts.The ground row was 0 and the row of 12 next to the ground end was only reading 2.38 and all the other rows were reading a fairly full charge. I seem to be killing a full row batteries at one time. I assumed the battery would drain even but I'm new to all this and am not sure. Is there anyone that could help me out with this? I have a ton of money into this battery and it's killing me.
Also...Can I revive the row of 12 that are dead? Thay were brand new batteries. (Sanyo GA18650)
Any and all help would be greatly appreciated.
 
You can probably manually charge just that one group and get it up to the same voltage as the rest, but you'll need to be careful and monitor them very closely while doing it. It sounds like your BMS was still connected through the winter and draining just that one cell group to power itself, it's a good idea to disconnect it while storing the battery long term.

You could use a 5V USB charger to do this, but you need to be VERY careful as 5V will eventually overcharge the cells if you're not paying close attention. I use an iMax B6AC RC charger for things like this as you can set the charge current and end voltage and not worry about doing anything bad to the cells. They're pretty cheap, I think I paid under $40 for mine.

Alternatively if you have the patience you can try just leaving the battery on the charger for a while (possibly several days) and let the BMS slowly balance the pack. Your charger light will turn green because the BMS cuts off charging while it (slowly) drains the other 19 cell groups down to allow it to continue charging the whole pack. Eventually the low group should be brought up to match the rest but starting from such a low voltage it will take a long time. You should see the pack voltage slowly creep up over time, but like I said above at this level of imbalance it will probably take many days.
 
Sounds like the BMS drained the cells unevenly during storage. Once a cell gets below about 2.5v, it is in a danger zone for charging. Low cells need to be charged at a very low C rate until they get above 2.5v or so. Too much current on a low cell can damage the cells.

Yours sounds like it's right near the limit. 12p is a pretty big cell group so you would be safe at 1A or so until the voltage gets up to normal.

Manually charging the low cell group is the best when the difference is large. The 5V USB adapter works, or there are many single cell chargers out there for cheap.
 
The 12 cells at 0V need to be replaced. The 2nd row at 2.38V can be gently brought back. BMS are notorious for using only one parallel group for their power, which is probably what killed the first row.
 
molybdenum said:
The 12 cells at 0V need to be replaced.

I wouldn't get rid of those cells yet. If they're charged properly they'll probably work just fine.

I'm still using some 18650s that were discharged to 0V accidentally. They still work great. They may have lost a slight bit of capacity from that but otherwise they work fine. The real danger comes from discharging below 0V. There's a threshold around -12% charge where lithium cells start to take serious damage that could be a safety issue.
 
Thanks for the replies so far. I just received my 12 replacement batteries. I'm going to try the 5v usb charger advice first...carefully. Hopefully I can bring that row back because I only used them once. My biggest concern is why that end row of cells are getting drained so much quicker than the others. Now that the last row are all zero my system is now trying to over drain the row of batteries next to it. I actually ordered another BMS and they both do the same thing (84volt 20s 150 amp Li-ion/LIPo Supower Battery). The replacement BMS was exactly the same.
I would hate to go through all the work of replacing the 12 batteries in my pack only to have them drained again. I know I went past that point of being drained because of the charger being hooked up and not shutting off. But why is it super draining whichever last cell group that is alive. When my motor kicked out the last time the row next to the ground end row that read 0 volts read 2.38 volt and all the other rows were much higher (high 3's). That row came back but I know that going under 2.5 is not good for the batteries. Something has to be wrong.
As well, for the size of my battery and the quality of cells I assumed I was going to get minimum 60 miles on one charge and I am nowhere close to achieving that. The way it stops running as well. Charged to 84 volts then it stops running at 4.2 volts lower than that 79.8 volts. Even the second time it did the same thing just with a little lower voltage. It seems to want to shut off when the last living row of cells are dangerously low. Just the last row on the ground end. All the rows should read the same voltage all the time and the rows on my ground end are nowhere close to being the same as the rest if they are taken immediate after riding. The amp hours are a fraction of what they should be.
With everyone's help I'm really hoping I can get this figured out.
 
The cells are supposed to drain evenly. The most common problem is a bad BMS that has the balancing shunt stuck on. If you charge the cells, then remove the charge current and watch the voltage for a while. You can see a steady decline if it's being drained. I would not rule out two bad BMS units. Depending on the design, they can be easy to blow when connecting them to the pack. The sequence used to connect the wires is important. I always connect the pack negative first, then plug in the balance wires and try to angle the plug so the negative end makes contact first. Connect pack positive last.
 
I have been constantly hooking and unhooking the whole setup including the BMS and not always the same way. My BMS is external.
Can I balance charge the pack with the BMS then hook the battery directly to the controller and bypass the BMS during discharge to see if it draws evenly and stops super draining my last row of cells? Safety is always a concern with these batteries.
Also...can anyone recommend a solid BMS for a 72 volt 120 amp battery?

Thanks
 
Hi all. I'm having trouble with my battery and BMS. I have a 20s 12p with Sanyo GA18650 cells. When my battery discharges it highly drains my last row of cells on the ground end of my battery. If I take volt readings right after a ride the last row of cells will read under 2.5 volts while the other 19 rows will all read high 3's. It also cuts off power way too early. I will go down 4 or 5 volts from a full charge and it will cut out. I have already changed out my BMS, I never changed brands (84v 20s 150a Li-ion/LiPo BMS from SuPower Battery), and it exhibited the same behavior.
I keep referring to the "last row of cells". It is actually the first row (1s). Charged it's reading should be 4.2. The cell beside would be 8.2 and so on.
PS. I have posted a much longer version of my issue in the Ebike Technical section (battery discharge unevenly). I thought it was more appropriate here under Battery Technology

Anyone with some input on this subject would be greatly appreciated
Recommendations for a solid BMS for 20s 12p (72volts 150amps) battery would also be appreciated.
 
It sounds like cell group No.1 has reduced capacity compared to the other groups. This could be due to all the cells having aged prematurely and each suffered somewhat, or you could have one or more very ill cells in the group dragging the rest down. The BMS will cut off the pack during discharge when any cell group drops below a certain voltage (usually ~2.5V) regardless of how much charge is remaining in the remaining groups.

This is assuming No.1 group is charging to 4.2V as it should be. If it's lower than the pack may just be badly unbalanced and group No.1 is never getting a chance to fully charge.
 
They are all new batteries. Only used once. The first group charged to full at first. I killed the first group because I left the charger hooked up for a drain on the full battery for storage purposes. I didn't realize then it was over draining my first group so I kept it going and I think I totally killed all 12 batteries in that group. Now that the first group is dead my system wants to kill the second group in the same manner. This could be a very expensive problem to fix. I am going to try and revive the first group with a hard charge from a 5volt usb charger for more testing. I have 12 new ones but don't want to add them until I figure out what is going on. On my other post it was mentioned once that the BMS was to blame.
Will this battery discharge evenly without the BMS...for troubleshooting purposes???
 
LWGT said:
I have been constantly hooking and unhooking the whole setup including the BMS and not always the same way. My BMS is external.
Can I balance charge the pack with the BMS then hook the battery directly to the controller and bypass the BMS during discharge to see if it draws evenly and stops super draining my last row of cells? Safety is always a concern with these batteries.
Also...can anyone recommend a solid BMS for a 72 volt 120 amp battery?

Thanks

Without the BMS, you won't have a way to tell when a cell goes dangerously low. If one cell group is weaker than the others, you might not see it in the overall pack voltage. If you over discharge a cell and reverse it, it will be destroyed.

What I would do is try testing the BMS to see if it is blown. One easy test is to disconnect the BMS and use an ohmmeter to measure across each cell circuit. The resistance across each cell circuit should be fairly high, like around 1M ohm. It would be normal to have a slightly lower reading on cells 1-3 as most use the cell 3 tap to power the control circuit. A stuck on shunt will measure much lower, like 100 ohms or whatever the shunt value is.

Another test is to charge the cells with BMS attached and see if any of the balancing resistors are getting warm.

Yet another test is to charge the pack and immediately disconnect the BMS and watch the cell voltages. They should stay steady.

Can you post a picture of the BMS? There are many types.

As far as a solid 20s BMS that can handle 120A, this is a moving target. A quick search on eBay shows this one:
https://www.ebay.com/itm/20S-72V-84...939138?hash=item4b3dd55842:g:HL4AAOSwqlFasjCY

I can't say if this one is any good or not. And it's only 100A, but it may be fairly easy to increase the trip point. There are some fancier Bluetooth programmable ones out there but they tend to be very expensive.
 
I have been using my newest BMS and it seems to be discharging the remaining 19 cell groups evenly and the last discharge went all the way down to 65 volts and still wasn't cutting out. I am going to go ahead and replace the 12 batteries in cell group one and then retest. Here is the BMS I selected.
 

Attachments

  • BMS.zip
    1.5 MB · Views: 28
My battery and completed bike.
 

Attachments

  • ebike batteryy.zip
    643.4 KB · Views: 32
  • ebikee.zip
    556.4 KB · Views: 30
On the old ones, you could try measuring ohms between adjacent balance wires and see if there is an obvious short.

Good you at least got it working. The BMS looks pretty beefy and seems to have the typical circuit arrangement.
 
Update...Repaired the battery. It only charged to 80.4 volts. It cut out and didn't want to run when it got down to 76.4 volts. When it got that low my V3 Cycle Analyst said low volts, battery flashing empty and then the screen shut off. I disconnected main power and then reconnected it. It powered up and my tank said 3/4 full. It ran again until I spiked the watts and then it repeated the same low volt scenario. I repeated this 3 or 4 times then I bypassed the BMS for the remainder of the discharge and it ran fine all the way down to the cutoff voltage 55 volts. I checked the battery and my first group of cells, the ones I replaced, are a little lower than the rest of them by about 0.25 volts. I am charging it again now, Maybe after a couple charges it will balance but I don't think the BMS is working properly.
 
Check state of charge after it comes off the charger full charged. Like this from the neg end as is # 1. Write it down on paper a must-see necessity.Plus you can put a set of 5s sense wires for a life of ez checking.
1. 3.98v
2. 4.12v
3. 4.20v



20. Xxx volt
Battery full charge voltage. Xxx volt
Charger voltage. Xxx volt.
Need to know.
1
 
999zip999 said:
Check state of charge after it comes off the charger full charged. Like this from the neg end as is # 1. Write it down on paper a must-see necessity.Plus you can put a set of 5s sense wires for a life of ez checking.
1. 3.98v
2. 4.12v
3. 4.20v



20. Xxx volt
Battery full charge voltage. Xxx volt
Charger voltage. Xxx volt.
Need to know.
1
[All this information you're asking for was put in my original, #1, post. ]
 
The BMS is designed to cut discharge off when any cell group drops below a certain voltage, and cut charge off when any cell goes above a certain voltage. If that one cell group was .25V lower, it will trip prematurely regardless of whole pack voltage. That's the BMS working properly.

If you want the BMS to balance out the cell groups, you need to leave it plugged into the charger even after the charger indicates that it's finished. The BMS will bleed off the other cell groups, and when they are down to a certain voltage it will resume charging the whole pack until a cell reaches the upper limit again. Then it bleeds a bit, charges a bit, etc. Eventually the cells will be balanced.

But you must leave the charger connected for any of that to work.
 
Thanks for the input.
I'm discharging the high cells in my pack now.
Hopefully now my BMS will work properly and keep it balanced now that the cells are closer in voltage. I think the second BMS is working because it stops charging when my highest cell group reaches 2.21 volts, but I have left the charger on after a full charge for 12 hours and the volts will not move from 80.4 volts. When I hammer the throttle the motor kicks out, but if I bypass the BMS on discharge nothing kicks out.
 
Thanks all. I drained my high cells and my charge is now up to 83.1 volts. The first four cells are now just 0.02 volts high. After I drained the high cells they were all the same which means the cells that were high went high again. When my charger is fully charged and green it will not turn back on for the balance step of the process. So, I'll try manually unplugging after 30 mins. and then plug back in to see if I can get this fine balance process going.
 
Back
Top