Is my Headway BMS broken?

[orangutan]

Hello,

I've received 16s Headway 10Ah cells (+1 extra cell) with BMS and 5A charger from Victoria few weeks ago. I've already made 6 rides with them and noticed, that after charging the batteries they become unbalanced (from 3.38V to 4.01V). BMS board looks ok, don't see any burned elements.

Can anybody tell me how (Headway) BMS balances cells? Does it balance cells all the time of charging or only in the end? How different might be voltage of cells after charging? Any suggestions what should I do? What can I check/test?

Thank you in advance.

PS. LVC seems to be working.

 

[dnmun]

something is wrong if it charges any cell up to 4V. have you measured the actual capacity of the pack with a cycle analyst or wattsup? they usually settle to 3.35V but they should not charge up over 3.9V.

 

[Russell]

It's very possible the cells were not in a balanced state when you received the battery. If that's the case then the voltage on the cells with more initial charge will go very high while the other cells still need charge. If any one cell gets too high, usually about 3.90V like Dnmun stated, the BMS will react by opening up the charging circuit. When it opens the circuit the charger indicator LED will turn green and you will figure the battery is charged but that's not really the case.

The best thing to do with a grossly unbalanced pack is to charge the low cells up manually to the same point with a single-cell charger. If this is not possible then remove the charger once it turns green and check all of the individual cell voltages. If the BMS shunts are working properly they should drain the high cells to the BMS designed balance point (mine was 3.60V). Therefore give the BMS some time to do its thing then check the cell voltages again. If the high cells have come down while the low cells have stayed essentially the same the BMS is working properly in that regard.

If the high cells are down to say 3.60-3.70V after 30 minutes then reconnect the charger once agin. All of the cells will get some charge but the high cells may once again trip the BMS so you'll have to remove the charger and wait 30 minutes for the shunts to drain them again. Repeat this cycle as many times as it takes until the individual cell voltages are all very close (+/-0.02V).

-R

 

[Icewrench]

It sure sounds like you have a bad bms. 3.7 is the top end for my headway cells.
Could you post a pic of the bms so we know which one it is?
I have used a light bulb to draw excess volts from a single cell.
Just a bulb from a car tail light will work.
But get the extra juice out of the high cells.

 

[Russell]

How high any individual cell voltage gets is a function of the charge voltage, how well the cells are initially balanced and the maximum cell voltage detection circuit of the BMS.

For example say you have a 16 cell LiFePO4 pack with cells containing all different initial state of charges, you have a 60V charger (avg 3.75V/cell) and a 3.90V upper cut-out circuit. In this case it would be easy for a single cell to hit the upper limit of 3.90V while the others are all over the place. On the other hand if your pack is already well balanced and you have a 58.4V charger (avg 3.65V/cell) plus your BMS balances to 3.60V then it's easy to see that no one cell may ever go over 3.70V.

It's also important to note that it matters when you check the individual cell voltages. If you check the voltages at the end of the charge cycle but with the charger still connected the average cell voltage will be totally dependant on the charger voltage. Once the charger is removed the BMS shunts will bleed down the cells to the designed in balance point. My 48V charger was set to 59.3V and my BMS balanced to 3.60V so if I checked the cell voltages with the charger connected they would all be in the 3.67-3.73V range. 15 minutes after I disconnected the charger and the BMS shunts balanced the pack the cells were all 3.60V +/- 0.01V. If someone has a Ping charger set to 61V all of their cells will be hovering around 3.8V until they remove the charger.




-R

 

[dak664]

Hello,
Can anybody tell me how (Headway) BMS balances cells? Does it balance cells all the time of charging or only in the end? How different might be voltage of cells after charging? Any suggestions what should I do? What can I check/test?

The FET shunts activate at at 3.65 volts, switching in 20 ohm resistors. This allows a ~200 miiliamp bypass, not enough if the charger is putting more than that, which can happen if the cells are way out of balance. In that case the highest cell will eventually hit 3.9 volts and BMS will disconnect. After that the high cells should bleed to 3.65 volts over a few minutes, then you can restart charging and and rinse, lather, repeat until the cells are balanced. Or limit charging to 100-200 milliamps using a series resistor, around 10 ohms, and watch the BMS temperature - it will need to dissipate ~0.5 watts per fully charged cell. Several amp-hours of imbalance could take 1-2 days to correct.

 

[dnmun]

i don't think the shunt switching transistors are mosfets. they are regular bipolar transistors where the base is turned off below a certain voltage and when the voltage across the cell climbs to a certain level then the bipolar transistor turns on and the current is diverted around the cell and through the shunt resistor.

russell was right about how the presence of a 3.9V signal on any individual cell will cause the charge to be terminated when the charging FET is turned off to stop the flow of current into the battery.

i don't know if you just dropped off your own thread, but if you follow al's idea of draining the charge off the highest cell, that will allow the other cells to charge up more.

but there is no way that the high cells should be reaching over 4V. that may be a problem with the accuracy of your voltmeter too.

 

[orangutan]

Thank you very much for answers and suggestions!

I have one cell charger. After a ride I charged all the batteries with 58.4V charger and then charged four cells with lower voltage (about 3.38V) with single cell charger up to 3.60V, but after some time their voltage dropped to about 3.40V. What is interesting, one of these four cells consumed more than 3Ah, while other three about 0.100-0.150Ah.

After that I decided to try to charge with 58.4V charger. It worked about 1 minute with fan consuming about 70W (when batteries are discharged it consumes about 320W), then green LED switched on, fan stopped and electricity consuption dropped to about 17W becoming less.

I decided to measure temperature on BMS with IR thermometer. On resistors it became up to 60C. On FETs only about 21C (19C in room). The highest voltage cell this time became 3.85V, after an hour dropped to 3.72V. Lowest cells became about 3.45V, after an hour - about 3.40V. Difference in voltage became smaller, but still unbalanced.

 

[dogman dan]

My experience with ping packs has been that they don't settle down so great till about cycle 10. New cells just seem to tend to charge differently from each other more than ones with a few more cycles on them. Hopefully you just have a worse than usual case of this.

 

[dnmun]

3.35V is where the headway cells settle. but you need to find out why the BMS does not shut off when it charges any cell up to over 4V. 3.9V is the max and once the cells have balanced then they should all be very close in voltage at the end and you won't have undercharged cells because of one cell being too high and shutting down the charge.

 

[Russell]

I have one cell charger. After a ride I charged all the batteries with 58.4V charger and then charged four cells with lower voltage (about 3.38V) with single cell charger up to 3.60V, but after some time their voltage dropped to about 3.40V. What is interesting, one of these four cells consumed more than 3Ah, while other three about 0.100-0.150Ah.

That explains a lot, as I figured your cells were not even close to being balanced so when you bulk charged the voltage for the cells with more initial charge went way up.

After that I decided to try to charge with 58.4V charger. It worked about 1 minute with fan consuming about 70W (when batteries are discharged it consumes about 320W), then green LED switched on, fan stopped and electricity consuption dropped to about 17W becoming less.

I would suggest if posible you increase the charger voltage to 59-59.5V. Now that you got the cells state of charge closer with the single-cell charger there shouldn't be the tendency for one cell to go way up. A higher charging voltage of 3.70V/cell (59.2V) will help get the low cells up bit faster where the BMS can then bleed them all down to the same point. The balance point on my BMS is 3.60V, yours might be different. I would not go over 60V as that can cause cells to trip the high cell level circuit on the BMS too often.

I decided to measure temperature on BMS with IR thermometer. On resistors it became up to 60C. On FETs only about 21C (19C in room). The highest voltage cell this time became 3.85V, after an hour dropped to 3.72V. Lowest cells became about 3.45V, after an hour - about 3.40V. Difference in voltage became smaller, but still unbalanced.

It's normal for the shunt resistors to get very warm when they are bleeding the voltage off the high cells to get them down to the balnce point voltage. Once they have accomplished this they will feel cold. I use this as a quick "touch" indicator that the balancing is completed.

Once the BMS has completed bleeding down the high cells you should check the individual cell voltages. You should find many at the identical voltage, these would be the cells which went over the balance point and were bled down. The cells which still read low never got above the balance point so the pack is not yet fully balanced.

Most chargers turn green when the current reaches a preset value, often 0.2-0.4A, however most will not shut off and will continue to supply current until the pack voltage reaches the charging voltage and the current approaches zero. Therefore leave the charger on the pack for a good hour after the LED turns green. This will help bring the low cells up, albeit slowly. After an hour remove the charger, wait 15-30 minutes and check the individual cell voltages, they should be closer together.

-R

 

[orangutan]

Thank you all for userful information!

After last ride I charged the batteries and all cells except two were close to 3.72V, while two about 3.45V. I tried to charge them with one cell charger and they consumed only 0.050Ah and charging stopped at 3.60V (I think my one cell charger is limited to 3.60V for LiFe). That means, that these cells were almost fully charged and voltage from 3.45V to 3.60V grows very fast. Maybe after few more cycles all cells will become balanced

 

[orangutan]

After few more cycles problem repeated. As I already said, I had one cell, which was undercharged 3Ah after charging battery pack. This cell was the last one in the pack (if 1st is with black balancing wire). I charged it with single cell charger and replaced with another. Now this cell is fully charged, while cell which I replaced with after several cycles became 2Ah underchaged. That means, that something is wrong with BMS balancing. BMS stops charging last cell too early. Maybe, that cell resistor turns on too early. Next time I will measure its temperature and compare with other resistors after 80% charge time. If it will be hot, while others not, that would mean, that it turns on too early.

 

[patrickza]

Ok just to get this out of the way, are you charging using the ch- port?

Then as to one cell staying lower I had the same problem. With me it was the charger putting out slightly too little voltage. I upped it by 0.5v and ever since it's been perfect.

 

[orangutan]

Yes, I charge through ch- BMS port. I connected wires as discussed in this topic: http://endless-sphere.com/forums/viewtopic.php?f=14&t=11651&start=15. I use 58.4V charger, that is 3.65V for each cell, but I usually get about 3.70V on other cells.

 

[patrickza]

I use 58.4V charger, that is 3.65V for each cell, but I usually get about 3.70V on other cells.

From my experience 3.65v per cell was a little low. I had to go from 87.6 to 88.3 to get everything perfectly balanced. That could be what you need. Also my BMS cuts all charge voltage if a cell hits 3.9, and then bleeds down to the 3.66-3.67 level. If your BMS doesn't do that then there is a problem.

 

[magudaman]

I believe I am using this same BMS and have been having similar woes. The bms doesn't seem to really do a true balancing but instead just shaves anything off the top, Like it should activate much sooner and do balancing based on its fellow cells, not once it goes over a certain point. I'm also using 58.4v charger. I am concerned about my cells hitting 3.9v, that is pretty darn high for those cells! They are supposed to be charged to 3.65 peaking out at maybe 3.75v. Looks like it might be time to look for a gary fecter bms :-/

 

[dak664]

If your Headway BMS has 20 ohm bypass resistors (like mine from a year ago) anything above a 185 milliamp charging current risks overvoltaging the full cells to >3.7 volts. If you limit the current to 150ma with a series resistor the the cells will clamp at 3.65 volts and it will safely balance (but watch the BMS for overheating).

I normally charge my 12 cell 10AH pack with a 2 amp Bionx charger that cuts out at 41.8 volts, 3.5 volts per cell so the balancing doesn't activate. After ~60 cycles I have been getting the low voltage disconnect at the end of my 8AH ride. When that happens, after the next Bionx recharge I connect it to a 48 volt supply through a 10 ohm resistor and watch the current as it slowly drops from 150ma to 100ma. It takes ~15 hours to bring up the cell that is 2AH low.

 

[cell_man]

This does sound a little strange. It it were me I'd be tempted to firstly try balancing the pack as best you can by bringing up any low cells or a better way might be to drain the high cells with a resistor whilst charging the pack. A 1ohm resistor across a single cell will pull just over 3A and require a rating of over 10W.

Putting a series resistance as was suggested above is also a good idea as it will taper the current as it reaches the max voltage and allow the BMS to do it's job. I have found that some chargers will stop charging is either the charger is disconnected (by HVC) or it reaches it's max charge voltage. If this is the case, it will be difficult to get an out of balance pack in balance again using that type of charger. The king power chargers can be modded to remove this stupid function.

 

[snowranger]

This thread is super useful. I was using 58.4 volts on my headway 16s for two days with no apparent balancing current occurring. After tweaking up to 59.1 volts a few minutes ago, I am getting about 170 ma of tapered charging. Awesome.

 

[dnmun]

i think that is why ping has adjusted his charger voltages up too.

 

[snowranger]

After 5 hours, another 1 amp-hr added.

The Ping BMS seems to behave differently though. Whereas the Ping intermittently goes from 300 watt to 10 watts during balancing, the Headway seems to stay at 10 watts all the time. Thoughts?

 

[dnmun]

not watts, you mean milliamps. i measured the shunt current when fully charged at about 70mA on the v2.5 signalab

 

[snowranger]

180 ma at 59 volts = approximately 10-11 watts.

 

[orangutan]

After not using my ebike for 3 weeks I decided to have a ride... But BMS immidiately disconnected when I pushed the throttle. I checked all batteries voltages and found out, that one battery voltage is only 0.3V. I disconnected it and charged with one cell charger. It consumed 12Ah. Gave that battery to a friend to measure its capacity.
Then I decided to find out what was the reason. And I think the reason was short circuit in Headway BMS.

As you can see, balancing wires' connector has burnt. But I don't understand how short circuit could happen in connector, when it is connected.