troubleshooting 36V lipo early cutoff

[jpodtbc]

hi all.

i've done a bunch of searching but can't quite seem to find direct and helpful info.

basically my 36V lipo battery is cutting off early. my device had shorted out due to overheating so i thought it this may have damaged the BMS but i don't know how to verify. it also sounds like the suggested course of action is to balance charge to make sure there isn't a faulty cell.

however i can't seem to figure out how this is done. do i need a special balance charger? can you guys suggest one? does balance charging require charging each cell individually? after balance charging what do i do next?

many thanks!

 

[dnmun]

what do you have for a battery and what is cutting out? do you have a voltmeter and you should post up pictures of your battery and the BMS if it has one.

 

[jpodtbc]

its a 36V 20AH LiFePO4 battery with a BMS. its all sealed up so i don't have helpful pictures at the moment. i will peel it all apart when i have time.

i bought it custom sized from victpower under a year ago and it worked great until about a month ago when it would early cut off after about 15 - 20% of the previous discharge.

 

[dnmun]

if it is a factory built pack then it has a BMS and that is cutting out because the battery pack has become so unbalanced.

you will need a voltmeter and pull the BMS out so you can measure the cell voltages on the BMS pcb while charging.

 

[jpodtbc]

ok so here's the BMS:

can you tell me how to voltage test the cells? my best guess is to put the positive on each of the 3 bits in a node while putting the negative where the black wires are soldered.

once i discover which cell(s) are low, how do i charge them individually?

 

[markz]

Just dont use that BMS, for now.
If you want buy a new BMS, they are cheap.

 

[jpodtbc]

markz: your comment doesn't really help my situation at all. a simple description of how to change the BMS would be a minimum since this is clearly my first time dealing with them.

 

[jpodtbc]

.

 

[jpodtbc]

i just did a bunch of testing. there's only 12 wires coming out of the cell pack so there definitely is not one wire per cell. not sure how to test individual cell voltages.

i tested each of these lines and the voltage for each of these is basically the summed voltage depending on where in the series its at. so respectively the voltages for each of the wires is 3, 6, 9, 12, 15 ... up to the max of 39 volts or so.

i also tested the voltage of the little metal groupings of 3 on the BMS and they all read identical voltages of about 3. there's only 16 of those that give a reading.

i dunno what to do at this point besides ripping apart the entire cell pack and testing each cell. but i wouldn't be able to put it back together. i'm doing the best i can here folks - any helpful info would be greatly appreciated!

 

[dnmun]

if you can measure the cell voltage on the pcb of the BMS then that helps eliminate the contact as an open but if you can only reach the cells, measure each channel voltage using the 20V DC scale.

measure to .01V while the pack is charging and preferably just at the point where the charger turns green.

if you can find the cell that goes to 3.90V and makes the charger turn green that is the best info.

but first measure all 12 channels and post up the voltages.

for the first cell, the black probe goes to B- and the red probe goes to the lowest wire in the sense wire bundle.

measure between each cell or pin on the sense wire plug to get the cell voltages. list them here for #1 to #12.

 

[jpodtbc]

i will do this now. fyi i looked up the specs on my BMS ( pcm-l16s60-562 ) and it is supposed to have auto balancing. i also noticed that its max discharge is 60A even though i asked the manufacturer for 85A. perhaps this is the problem although i have no idea why it would have worked great for 10 months.

 

[dnmun]

they all autobalance when you reach full charge. but your battery has one cell so far outa balance that it limits the ability of the pack to be fully charged.
so we have to find the one and drain it down to match the others, and then it will balance and you can store the full charge in it again.

so measure the cell voltages and post them up so we know what your battery's condition is. measure while charging.

 

[999zip999]

Write down the cell voltages starting at the neg. end like this.
1. 3.55v
2. 3.66v
3. 3.22v

12. 3.65v

 

[jpodtbc]

i noticed something strange just now. the charger was going back and forth between charged and charging. while doing this the voltage on the first pin was fluctuating between 6.55 (when green) and 3.52 (when yellow). that seems wrong.

as such i couldn't accurately measure the voltages while charging since the readings would jump depending on charge state. here are the voltages with it NOT charging:

3.06
3.67
3.36
3.36
3.35
(the following readings could only be done with 1 decimal resolution)
3.8
3.4
3.8
3.4
3.5
3.4
3.5

 

[dnmun]

and what are the voltages when it is charging?

 

[jpodtbc]

AGAIN: I COULD NOT GET READINGS WHILE CHARGING because the voltage kept jumping. on the first wire it was at 6.55V most of the time while the charger was green and would drop to 3.52V when the charger was yellow.

what causes this spike when green? i feel like that shouldn't be happening.

 

[dnmun]

yes, it should not be happening. there is no way the battery cell's voltage is jumping around so you have to figure out how to measure the cell voltages while it is charging and that means you have to figure out why the voltage is jumping all over.

there is no way to know what you are doing. i asked one thing and then you measured using a different technique which involved doing something else and i have no idea what that was either so it is really difficult to even figure out what you are doing to get the voltages jumping. cell voltages do not jump around.

maybe show a picture of where your probes is located and the scale on your DVM.

 

[jpodtbc]

i understand its hard for you to know what i'm doing. i'm not a complete idiot but there are things that i don't know.

the "jumping" voltage is not something i'm doing. while charging it keeps switching back and forth from charging to charged - i'm not touching anything to make this happen. if i measure the voltage on any of the lines the reading jumps about 2-3 volts above what is expected - corresponding with the changing of the charge state. obviously the charger is supposed to go green when any cell voltage goes above a certain amount so it could just be doing its job if the voltage jump is being caused by something else.

something tells me the BMS is fried. my controller has this problem where it overheats if i run the vehicle too hard. the connectors melt and the battery shorts. my guess is that when this happened last it fried the BMS.

damn electrical is such finnicky shizzle!

 

[dnmun]

i try not to guess. it is obvious there is a bad connection somewhere if the voltage jumps around since as i said and i think you would also that the cell voltage cannot jump 100% in microseconds. so there is a bad connection somewhere. since i do not know what you are contacting with the probes or what the construction of the pcb is like i cannot guess anything beyond the obvious.

 

[jpodtbc]

if you look at the above picture there's 12 white wires coming off the left of the BMS. i stuck a pin in each one and measured that with the red positive while putting the black in the battery neg terminal that goes to the controller.

how can i chase down the jumping problem? take apart the entire cell pack? there's gotta be upwards of 90 cells in there!

 

[dnmun]

the 'battery negative terminal that goes to the controller" is the P- discharge output from the BMS and you are connected to the drain of the mosfet so the voltage is jumping because the mosfet is turning on and off rapidly under some adverse condition.

if you remove that lower fiberboard shield under the BMS then that exposes the solder pads where the sense wire pins are soldered to the BMS. measure your cell voltages by sticking the voltmeter probe points into the solder for each adjacent cell. see what the voltage measurements are then. measure between adjacent pins while charging. if the voltage continues to jump then disconnect the charger and measure again. those numbers tell you everything you need to know.

 

[jpodtbc]

i measured between each adjacent point and the readings were identical: about 3.41 when charging 3.34 when charger was green. all identical.

 

[dnmun]

so now the voltage readings are stable and it is only partially charged. it appears to be about 40-50% SOC from those numbers.

so if it is not charging above that level and the cells are all balanced then the charger itself is not putting out sufficient voltage to fully charge the pack.

does your charger voltage output show that same voltage jumping that you mentioned before? your charger should have a stable voltage of about 43.2-43.6V on the output when measured with the meter.

except there is the cell#1 voltage which has to be measured between that first empty hole and the first solder bump. that is cell#1. you should have a total of 12 measurements.

 

[jpodtbc]

the charger measured at 44.4 V stable.

the jumping has to do with it switching from charged to charging state. this only happens when it is plugged into the battery. i'm not sure how to measure the charger output when it is plugged in. most of the time the charger is in the charged green state. but every 20 seconds or so it flips to yellow charging for about 1 second.

i measured the first pin this time (didn't do it last time) and it seems to be acting strange:
- when not plugged in it reads 3.61 V. all the others read about 3.35 V.
- when plugged in and the charger is green (charged) it reads 3.90 V while all the others continue to read about 3.35 V.
- when the charger flips to yellow it seems to go to about 4.18 V for a moment before going back down to 3.90 V.
- if i unplug the charger, the first pin will initially show a 3.90 V reading but over the course of about 5 minutes it will slowly drain back down to 3.61 V.

 

[dnmun]

ok, finally got an answer been looking for all this time.

the cell that is at 3.90V is causing the charging to shut off. you have to drain that cell down to the same level as the others in the pack. use a power resisor or little light bulb to drain it down.