Possible BMS Issue

[warkittens26]

So I recently purchased a BMS (see attached picture of the BMS and wiring diagram) and wired it up to my battery. Across the battery I get 41ish Volts. I am 99.999% positive I have wired this up correctly. I have taken readings from the pin connector for each bank and it matches with what I expect each bank to be at from 0 to 41ish volts. I have B- going to the negative of the first bank.

My issue is this: When I measure the voltage at Ch- (Charging negative) I see 41 volts. When measuring P- (Discharge negative) I also see 41volts. When I connect my battery to my bike and turn on the LCD display...the voltage on the P-/P+ drops down to 14volts and my LCD turns off, and then the voltmeter shows the voltage rising slowly until it reaches 41 volts again. As soon as I turn on the LCD display again, the voltage drops again to 14 volts.

When I bypass the BMS and connect to my bike directly I see that the voltage is 41Volts. When I crank the throttle the hub motor revs up and my voltage drops a bit as it under load so I know my battery is working fine.

One suggestion I found was that it was possible my BMS locked out. The suggestion was to connect the charger to P- and charge it for a minute or so to "Unlock" the BMS if it had locked up. No joy, nothing happened and voltage drops down to 14 volts even when the load is just a tiny LED light.

Do I have bad BMS? Is there a way to test this?

Michael

 

[amberwolf]

The BMS probably is reacting as if a cell is at LVC when the load is applied, and so it cuts off it's output.

When you apply the load without the BMS, what are the voltages of each cell?

What are the voltages of each cell without the load applied?

Your cells might not actually be dropping below their chemistries' LVC, but they might be dropping below whatever this BMS's cell-level LVC is set at.

Did the supplier give you any info on what that cell-level LVC might be? Or what the BMS's cell-level or pack-level HVC might be? Or at least which chemistry it is for?

 

[warkittens26]

Forgot to add some detail regarding the battery I made:
10s/9p = 41volts @ 20Ah
Used 90 18650 Lithium Ion Batteries made by Panasonic

The BMS probably is reacting as if a cell is at LVC when the load is applied, and so it cuts off it's output.
I have had this battery pack soldered up now for 4 weeks. The voltage has remained at 41volts.

When you apply the load without the BMS, what are the voltages of each cell?
With Load, No BMS, Just probing each bank of 9 cells:
B1=4.11v, B2=8.17v, B3=12.25v, B4=16.37v, B5=20.50v, B6=24.63v, B7=28.76v, B8=32.88v, B9=37.01, B10=40.95v
Voltage at P- = 14v


What are the voltages of each cell without the load applied?
With No Load, No BMS, again just probing each bank of 9 cells:
B1=4.11v, B2=8.17v, B3=12.24v, B4=16.37v, B5=20.50v, B6=24.63v, B7=28.76v, B8=32.88v, B9=37.05, B10=41.06v
As my pack is all soldered together I am unable to get individual cell readings.
Voltage at P- = 41.03v

Your cells might not actually be dropping below their chemistries' LVC, but they might be dropping below whatever this BMS's cell-level LVC is set at.
Did the supplier give you any info on what that cell-level LVC might be? Or what the BMS's cell-level or pack-level HVC might be? Or at least which chemistry it is for?
This is the information I got from the listing when I bought it:
Applicable: 36V 10 cells lithium battery protection board
Batteries Type: lithium cobalt oxide / manganese lithium / ternary materials
Single overcharge protection voltage 4.25V ± 0.05V (4.20-4.35V/0.05V per upgrade)
Single overcharge recovery voltage: 4.10-4.00V
Single over-discharge protection voltage: 2.50V ± 0.1V (2.50-3.0V/0.05V per upgrade)
Single over-discharge recovery voltage: 2.80V ± 0.1V
Protection Current consumption: ≤300UA
Short circuit protection current: 40A ± 3A
Short circuit protection time: 500MS
Temperature protection: 55/65/75 degrees
Discharge current: 30A
The maximum instantaneous current: 40A
Single balanced voltage: 4.19V ± 0.02V
Single balanced current: ≤ 55MA
Charge current: ≤10A

Take care,

Michael

 

[flangefrog]

The BMS specs look alright. Your cells don't seem very well balanced so you might want to check the voltage of each bank (such as B10) under load. Edit: I missed that you already did this.

You could also check the balance connector plug and socket for bad connections. As long as you can measure a voltage for each cell bank on the BMS's balance socket when the battery is plugged in it should be fine. The connection doesn't need to be capable of very high current except when passively balancing the battery.

If all the parallel cell banks show a voltage in the range allowed by the BMS under load you can complete the below to check for any faults in the BMS itself (or whether the BMS is for some reason disabling the output).

To rule out problems with some of the BMS connections, traces and failed open-circuit FETs (rare) you could try discharging through the charge connection.

When a FET is turned off it is not completely off, just very high impedance. Thus you might read the pack voltage on the discharge port without a load even when the FETs are turned off. You could try measuring the voltage between the Gate and Source pins of the mosfet to confirm whether it's supposed to be off or on.

If you confirm that the gate voltage is high enough (check the FET datasheet) you could measure the voltage across the Source and Drain pins under load (should be almost 0V) which will definitively tell you if the FET is faulty.

Once you have the BMS working you should charge the battery at 42V for a long time to balance all the cells.

 

[warkittens26]

If you confirm that the gate voltage is high enough (check the FET datasheet) you could measure the voltage across the Source and Drain pins under load (should be almost 0V) which will definitively tell you if the FET is faulty.
Under Load Using Diode Check mode to obtain Ohms
FET1 (DS) =34.62V Ohms=Open
FET2 (DS) =34.62V Ohms=Open
FET3 (DS) =34.62V Ohms=Open
FET4 (DS) =14.04V Ohms=0 Ohms (audible alarm) (FET4 is next to the P- pad on the board)

No Load
FET1 (DS) =40.91V Ohms=Open
FET2 (DS) =40.91V Ohms=Open
FET3 (DS) =40.91V Ohms=Open
FET4 (DS) =40.91V Ohms=Open (FET4 is next to the P- pad on the board)

Data sheets says that when the FET is ON, should be 5.5MOhms (typically)
Under Load:
FET1 (Rds) = 0Ohms
FET2 (Rds) = 0Ohms
FET3 (Rds) = 0Ohms
FET4 (Rds) = OL MOhms (Open)

No Load:
FET1 (Rds) = 67.1K Ohms
FET2 (Rds) = 67.1K Ohms
FET3 (Rds) = 67.1K Ohms
FET4 (Rds) = OL MOhms (Open)

Long story short I'm thinking FET4 is blown. Makes sense as I remembered when I first hooked up the charger I got a huge spark (Im a dumbass and didn't check the polarity in my excitement charge my new pack, this destroyed my charger brick as well)

So, I placed another order for a new BMS. Lesson learned, even if you think your polarity is right, triple check that again... sigh.

Thanks again for the all the help. It was fun to take a trip down amnesia lane and relearn Mosfet properties again.

Take care,

Michael

 

[amberwolf]

You show both V (volts) and Ohms in some of your measurements, but this can't be correct. Did you measure Volts, or Ohms? Which one was your meter set to?

Please note that you shouldn't try to measure Ohms using your meter while power is applied to the circuit, only with all power disconnected from the FETs (other than from Gate to Source). If there is power across Drain to Source, it will make your Ohms reading incorrect (or potentially damage your meter).

 

[warkittens26]

I have two Volt meters, so I had one taking Volts and the other taking Ohm measurements. Sorry, I should have made that clearer on my last post.

 

[amberwolf]

Even so, everything I said still applies, and the questions still need answering.

 

[warkittens26]

I will hook up the BMS again and do as you suggest, might be able to get numbers until this weekend. Thanks again for the tips and information.

 

[warkittens26]

So I'm more than a little frustrated. Today I hooked up a Supower BMS. Was no longer seeing the voltage dip when any kind of load was applied. With my battery still at 41v, hooked up my battery to my bike and did 36 miles. In fact it was a great ride with 2 steep hills that my battery chewed through pretty good. Got home and saw on my meter that my battery was 37.01V. I went and hooked up the female connector (neg to P-)(P+ from 10th pos bank). As this is a Supower, P- is used for both discharging and charging. Plugged in my charger (Output: 42V @ 2amps) and got nothing, bupkiss, zero. The charger has a green/red LED to indicate battery fully charged, and red/red for charging. Mine is green/red which means no charging. Hooked up my Volt meter and set it too amps and visually confirmed that there was no current flowing.

I'm starting to think that maybe one bank, or more, is too out of balance and the BMS is shutting down the flow of current. As I am making this post at 1am in the morning I dont have the voltages written down from bank to bank. I did check them with I got home and it seemed they discharged pretty evenly between them though several were at 3.75, there were two that were at 3.65. That's a whole volt right there, so that might be my issue. Some research on the web I found that some suggestions were to circumvent the BMS entirely and charge the banks that are low until they are all about the same voltage on every bank and then hook up the BMS and let it balance charge for a few days. This seems like a good approach, but am I missing something here? Any suggestions would again be helpful.

I should mention this as well, I had a 12s BMS thats been collecting dust for several months decided to give that a try , the last two wires (B11+ and B12+) were obviously not used, trimmed the wires a bit and taped them off with black electrical tape so as to avoid a short. This has been the first BMS that has actually worked when discharging. I'm going to buy the Supower 10s BMS, but just wanted to see if I could get discharge/charge working. Not sure if this might be causing the non charging issue as maybe the BMS is seeing those as out of balance?

I will post bank to bank voltage readings in the morning.

Take care, ,

Michael White

 

[warkittens26]

As promised here are the voltage readings of each bank:
B1=3.76V
B2=3.66V
B3=3.67V
B4=3.76V
B5=3.756V
B6=3.76V
B7=3.76V
B8=3.76V
B9=3.76V
B10=3.65V

Currently the battery is at 37.22V. I can already see that I have three banks (B2, B3, B10) that are approximately a volt below the rest of the other banks. My question then is do I charge the three banks up to the levels of the other banks, and will this get the BMS to finally start charging the whole battery pack?

Michael

 

[Ykick]

Volt below?

 

[warkittens26]

This is what happens when you try to post something before coffee. My apologies. It's not a volt below. Is there a tolerance on the BMS? Somethings I read last night says that it could be as low as .02 to as much as .08 difference between banks. I cant find anything that is specific to the Supower 12s BMS.

Sorry for the confusion.

Take care,

Michael

 

[flangefrog]

This is what happens when you try to post something before coffee. My apologies. It's not a volt below. Is there a tolerance on the BMS? Somethings I read last night says that it could be as low as .02 to as much as .08 difference between banks. I cant find anything that is specific to the Supower 12s BMS.

Even if it was a volt below the BMS should charge it up and start bleeding the highest cells once they reach ~4.25V. Edit: Actually I think that's the cut off voltage, it will balance them before they reach 4.2V.

 

[warkittens26]

Then I'm completely at a loss as to why the BMS will discharge as expected, but not charge. At this point I dont know what to do to make this work.

 

[flangefrog]

When you posted the resistance tests you made on the FETs, did you first check the gate voltage as in my first post? If you found the correct gate voltage there, then the problem can only be with the power stage. If you didn't, then it means the problem is with the measurement part of the circuit.

I know you found one FET that was off when the rest were on, I think it would be best to make a diagram of the power stage by following the traces back from the P-, CH- and B- terminals. This way we can get an idea of what each FET is for and which ones should be on or off.

As far as I know this is a fully analog BMS so it should not be too hard to find out what's wrong and fix it.

Did you also try discharging through the charge port? (Unlikely to be a problem but still an easy test)

 

[warkittens26]

Oh my god!!! I figured it out. I had ordered a charger socket that my existing charger would plug into. This whole freakin time I have been using that charging socket and seeing that my battery was not charging. I went back and measured the voltage coming out of my charger as 42 volts. Then I would plug my charger into the charger socket and measured it again...nothing...it was zero. Feeling really dumb right now.

I wrapped my p- lead around my charger tip and used electrical tape to hold it in place. Took the wire from my pos rail and put that inside my charger and now my battery is charging. :lol: :lol: :lol: Yes, I have totally MacGyvered this. Going to go back and put on my 10S BMS and get a proper charging socket for my BMS that works with my charger brick.

I want to thank everyone for all of your help!

Take care,

Michael