New BMS cuts out

[dcuste]

I replaced my old fried BMS and then took the bike out for a 5 mile ride and the battery (13s5p all Samsung 35E cells) was working great. Plenty of power (up to around 1600 watts max) and voltage was good. I turned off the display and let it sit for about 15 minutes. When I tried to turn it back on, the display only lit for a fraction of a second and then went dead. The battery also would no longer take a charge. After taking the battery apart and checking that all individual cell groups had good voltage (4.0 to 4.1), and the output on the BMS (C-) was the same as the battery input (B-) (53v) with no load. As soon as any load was added the BMS output drops to zero.

I then pulled off the BMS sensing wires connector and re-seated it. The battery once again will take a charge and also supply load power. Although I have not put the battery back together yet for a more though test, I believe the battery is now working properly again. I am now wondering if once the BMS shuts down for any reason, that momentarily disconnecting the sensing wire connector is the way to reset the BMS. What do you think? Is there a problem with the BMS, or with a weak battery group (even though they all have normal voltage) or the connector for the sensing wires, or one of the sensing wires intermittent open or maybe the battery is just out of balance? Any suggestions on a test that will reveal the issue?

 

[e-beach]

Have you tried to recreate the problem a second time? It could be a number of things. Unplugging the BMS and plugging it back in could have reset the BMS. It might have sensed a low voltage and needed to be reset. Check all your connections and make sure that are all good. The little connectors on a bms can sometimes get loose.

Also, did the BMS get hot? It might have a thermal cutoff circuit.
Are you 100% sure your cell groups are all good? Sometimes cell groups can bounce-back when not under load.

Do you have a link of the type of BMS you own so we can look at the specs?

I would first make sure all your connections are good and then see if the problem reoccurs.

 

[dcuste]

Thanks for the quick reply. I have not tried it out again for a second time yet. Connections have all been triple checked and look good. The BMS did not get hot. I am not 100% sure the cell groups are good. I did put the battery under a small load of maybe 50 watts and didn't see any drops. I would think that if I had a weak cell group, the BMS would have cut off when I was pulling 1600 watts. The battery worked great until I shut the bike off. I bought the BMS off ebay without any manufacture data. Here is a description...

Specifications

Size:120*60*8mm
Continuous Operating Current:45A
Voltage:48V/52V
String:14
Balanced Charge: YES
Short Circuit Protection: YES
Wiring: charge and discharge the same interface

The ad also states..
"You can use it as 14S for a 52V battery, or 13S for a 48V battery by leaving the last balancing wire disconnected."

So I left the last wire disconnected as instructed. Now that I think about it, wouldn't the BMS not sensing that 14th cell group voltage make it cut off? How does that work for a 14s BMS to work as a 13s?

 

[e-beach]

Your BMS probably only reads single cell (cell group) voltages when balancing, not the cumulative total of all the cells. I might be wrong but I think that is how it works.

As for your current concern. Repeat the ride and see if your bms repeats the problem. Pay attention to your cell groups and notice if one goes to a lower voltage then the others.

Here a another link to check out. It will give you a start on what to do with that 14th wire.

https://endless-sphere.com/forums/viewtopic.php?t=95346

 

[amberwolf]

Your BMS probably only reads single cell (cell group) voltages when balancing, not the cumulative total of all the cells. I might be wrong but I think that is how it works.

That's how just about all of them work, from what I've seen. Some of the Smart Bluetooth BMS units may do things differently, but I haven't seen anything definitive, so would assume it's still teh same for those.

When I tried to turn it back on, the display only lit for a fraction of a second and then went dead. The battery also would no longer take a charge. After taking the battery apart and checking that all individual cell groups had good voltage (4.0 to 4.1), and the output on the BMS (C-) was the same as the battery input (B-) (53v) with no load. As soon as any load was added the BMS output drops to zero.

This indicates the BMS itself detected a problem and shutdown both charge and discharge ports to protect the cells. The three things most of them check for are overcharge HVC, overdischarge LVC, both of which are cell-level, and overcurrent, which is pack level.

A poor connection on the balance wires could cause LVC to be triggered by having low or no voltage at a sense line. Normal LVC would only prevent discharge, but if it's a good BMS it'll prevent charge too if the voltage on any cell is below a safe recharge voltage (and a disconnected line should show 0v, which is definitely below that).

A question: you indicate the output is on the C-, which is usually for charge port. Does it have a separate discharge P- port? Or is that the same electrical connection as the C- port? If it is electrically separate, you need to use the P- for discharge, or else the pack cannot protect against overdischarge, *and* the charge fets on the C- port are being overpowered by the discharge current that is much higher than the charge current they are typically intended for, which can overheat and damage them.

If it is electrically the same, then you would be connecting both charge and discharge wires to the same point (or to the P- and C- pads, as they are electrically connected in this case).


Second question: Did you measure the cell voltages at the BMS itself, on the solder pads for the balance wire connector? Or at the cells themselves? If the former, then all conections are good, and there is another problem. IF the latter, the connetion could've been a problem, and reseating it could have fixed it (if ti doesn't there could be a spread contact in the connector, making an intermittent connection).

I am now wondering if once the BMS shuts down for any reason, that momentarily disconnecting the sensing wire connector is the way to reset the BMS.

Though I have not seen this myself, there is a thread around here somewhere that says that doing this will destroy a BMS if all the connections are not remade in exactly the right order at the right time, so until that's disproven I wouldn't recommend this.

 

[dcuste]

This indicates the BMS itself detected a problem and shutdown both charge and discharge ports to protect the cells. The three things most of them check for are overcharge HVC, overdischarge LVC, both of which are cell-level, and overcurrent, which is pack level.

That makes sense to me, but I would have expected the shut down would have occurred while I was running the maximum current on the ride. The last 2 miles of my ride was at about 300 watts and when I turned it off all appeared normal. It wasn't until I tried to power it up again, did the BMS shut down. Seems strange to me.

A poor connection on the balance wires could cause LVC to be triggered by having low or no voltage at a sense line. Normal LVC would only prevent discharge, but if it's a good BMS it'll prevent charge too if the voltage on any cell is below a safe recharge voltage (and a disconnected line should show 0v, which is definitely below that).

This makes sense, but I wasn't expecting it to lock out and only reset when I pulled the balance wire connector off. This protection also makes me wonder how is it possible for a 14s BMS work with only 13 cell groups connected to it. I would think the BMS would not like having 0v on the last balance wire.

A question: you indicate the output is on the C-, which is usually for charge port. Does it have a separate discharge P- port? Or is that the same electrical connection as the C- port? If it is electrically separate, you need to use the P- for discharge, or else the pack cannot protect against overdischarge, *and* the charge fets on the C- port are being overpowered by the discharge current that is much higher than the charge current they are typically intended for, which can overheat and damage them.

If it is electrically the same, then you would be connecting both charge and discharge wires to the same point (or to the P- and C- pads, as they are electrically connected in this case).

There is no P- port. My old BMS had a P- port, but the wiring schematic and seller made it very clear that on this BMS the charge and discharge wires are both connected together on the C- port. I assume there is some extra circuitry in this BMS that prevents the problems that you reference, but don't really know if there is a down side to not having a separate P- port.

Second question: Did you measure the cell voltages at the BMS itself, on the solder pads for the balance wire connector? Or at the cells themselves? If the former, then all conections are good, and there is another problem. IF the latter, the connetion could've been a problem, and reseating it could have fixed it (if ti doesn't there could be a spread contact in the connector, making an intermittent connection).

I did measure the cell voltages at the cells. I'm not familiar with a spread contact, but I'll research that and try to eliminate it from being an possible issue.

Though I have not seen this myself, there is a thread around here somewhere that says that doing this will destroy a BMS if all the connections are not remade in exactly the right order at the right time, so until that's disproven I wouldn't recommend this.

Thanks for the warning. My reading on replacing the BMS mentioned that plugging in the connector should be the last step. I'm pretty sure there is no way around plugging and unplugging it, but I'll try to minimize it and do as evenly as possible.

 

[e-beach]

Did you try disconnecting the B- connector before unplugging the BMS balance connector?

Also, this is from the only eBay site that matches your description. Is it correct?

 

[dcuste]

Yup. That looks like my BMS diagram. I didn't even think to put a connector in the leads to the BMS. The original BMS was hard wired in and I thought that is just how it's done. I can easily cut it in.

 

[e-beach]

.... I can easily cut it in.

put a high quality connector in place and if it happens again disconnect it and see if the BMS resets.

 

[dcuste]

Thanks everybody for the help. I'm writing this update just in case anybody is curious how this saga ended. I did buy the connector suggested by e-beach, but decided to try equalizing all the cells before installing it. So I charged and then drained the whole battery very slowly over a few days while frequently measuring cell group voltages.

If there are any dead cells, my theory was that the group with the cell issues would show themselves with a voltage drop below the rest of the cell groups. Eventually all the cells groups dropped down to just above 3.00v without any groups dropping lower. I then charged the battery back up to 54.6v and left it on the charger for an additional 5 hours. I took the bike out for some pretty hard riding with periods of up to 1600 watts going to the motor while keeping an eye on the total battery voltage. The battery performed well and the BMS didn't heat up at all and never shut down. I stopped when I had the battery voltage down to right around 41v. The battery charged back up as expected.

My conclusion is that the this 14s BMS is working for a 13s battery and that the earlier problem I had was probably a result of my old BMS failure and maybe the cell groups were just in need of rebalancing. I really don't know enough on how this new BMS works to know if it is good not to have a P- terminal and if just leaving the 14th sensor wire disconnected will cause other issues, but for now I'll just use it until it brakes. I do believe that a 45 amp BMS was needed and this was the one I found that I could make fit in the battery case.

Thanks again for all the help.

 

[e-beach]

Humm sounds like when you put the new one on, it didn't have enough time to balance the cells. Sounds like all is well now. Good News! :thumb: