Endless issues with BMS on 12s LiFepo4 Pack

[alexjthayer]

Hi all, I have been working on a battery system that is mounted on my Boosted board (Electric skateboard) to extend the range for about a year now, but just constantly have issues with the BMS. I am using 4 3S1P 4.2Ah LifePo4 batteries from HobbyKig in series (https://hobbyking.com/en_us/zippy-flightmax-4200mah-3s1p-30c-lifepo4-pack.html) and a 43.2v 5a charger I had custom built from somewhere in China. I have gone through half a dozen BMS's due to them either randomly dying or destroying my batteries... Just recently I purchased this BMS (http://www.batterysupports.com/36v-60a-12s-12x-32v-lifepo4-battery-bms-lfp-pcm-smt-pcb-system-p-251.html) in hopes that it might actually work correctly, but now I'm sitting here with three or four puffed cells that are charged ranging from 3.3v to 4.4v... Another $120 of batteries trashed. Before I connected the BMS all of the cells were balanced charged to 3.6v, and the cells have only gone through about 5 cycles before getting unbalanced. I also checked all of the sense wires before connecting the pack with a multimeter and everything was in order. I'm at a lost for what to do, and can't stand to destroy anymore of these pricey batteries.

 

[kdog]

I've used batterysupports Bms's quite a few times in the past with no issues but I'm an 18650man so that could be the difference. A bms can only do so much balancing- somewhere around 20ma so if your batteries are really un balanced / different capacity, no normal bms will be able to cope.
Have you done capacity and IR checks on all your previous packs?
The qc on Rc batteries is apparently not great so maybe you got some duds an the bms couldn't keep them balanced then they died.
What is you charge/discharge regime?
I'd start with again with packs that are known equal (or close) capacity, ir and V and that bms. Then see how you go. If something goes wrong I'd be looking at habits, wiring etc.
Have you got a balance charger? Rc batteries are not known for their plug and play ease of use and require extra attention.
Having said that lifepo should be more forgiving than lipo.
Start testin I reckon, get some more info.

 

[dogman dan]

To start with, I've never heard of those HK lifepo4 packs being any good.

My best guess is you pull too much amps for the size of battery, and perhaps discharge too low, and or, pull too much amps at that point where the pack is below 80% discharged.

What is the lvc of your bms?

I smell a rat, and its not the bms. But I'm just guessing, since I don't know anything about the board, your weight, etc. Give us the low down on that stuff, before we blame the bms.

What makes me think its not the bms, is that it happens over and over. Bear in mind, its well known in the RC world that running this type battery at its max specs results in a puffed pack pretty quick. So if you have a 20c lipo cell, run it at 5c max is the way to go. But since its the lifepo4, I'd say 2c should be your max. That's 8.4 amps max.

 

[cwah]

+1 with dogman.

Boosted board pull about 2000W. Thats why they are still using A123 lifepo4

There is no way cheap HK lifepo4 can handle the load. They probably can handle 200-400w max

 

[alexjthayer]

I've used batterysupports Bms's quite a few times in the past with no issues but I'm an 18650man so that could be the difference. A bms can only do so much balancing- somewhere around 20ma so if your batteries are really un balanced / different capacity, no normal bms will be able to cope.
Have you done capacity and IR checks on all your previous packs?
The qc on Rc batteries is apparently not great so maybe you got some duds an the bms couldn't keep them balanced then they died.
What is you charge/discharge regime?
I'd start with again with packs that are known equal (or close) capacity, ir and V and that bms. Then see how you go. If something goes wrong I'd be looking at habits, wiring etc.
Have you got a balance charger? Rc batteries are not known for their plug and play ease of use and require extra attention.
Having said that lifepo should be more forgiving than lipo.
Start testin I reckon, get some more info.

Alright well it's good to hear some success stories, I know they were balanced within 0.02v when the charger and bms went on, and although I don't have a proper capacity tester, but I have figured out with my cheap rc charger that now that the batteries puffed anyway, the 3rd cell in each pack has significantly less capacity - maybe around 1000mha, I'm wondering if this could be a manufacturing defect? This appears to be a thing in the two non puffed batteries too so I'm not sure what's up there.

I only have the need to run the whole system below 50% about once a week, and then I run until a low voltage alarm hits 2.7v and notifies me rather loudly. I wish I could just rely on the bms for this but I think it's lvc is lower that sounds healthy for the battery. Then I just plug in the whole system and pretty much leave it until I need it again.

That's sounding like what my plan will be, I do have a cheap hobby charger that I will balance before hand with, but I really want the whole system to be plug and play with one port.

Thanks for the help!

 

[alexjthayer]

To start with, I've never heard of those HK lifepo4 packs being any good.

My best guess is you pull too much amps for the size of battery, and perhaps discharge too low, and or, pull too much amps at that point where the pack is below 80% discharged.

What is the lvc of your bms?

I smell a rat, and its not the bms. But I'm just guessing, since I don't know anything about the board, your weight, etc. Give us the low down on that stuff, before we blame the bms.

What makes me think its not the bms, is that it happens over and over. Bear in mind, its well known in the RC world that running this type battery at its max specs results in a puffed pack pretty quick. So if you have a 20c lipo cell, run it at 5c max is the way to go. But since its the lifepo4, I'd say 2c should be your max. That's 8.4 amps max.

I know they're not the highest quality, but they have an excellent form factor for my application and there's another thread around here of someone using them for the exact same thing. I've wondered about that quite a bit, the batteries advertise a 30c discharge, but I know that's just false. This pack is running in parallel with the internal battery pack, so the biggest load it should ever experience is about 30amps, factoring in all of my lighting and extras. Maybe it would experience 50amps if I ever decided to run the board with the internal battery dead.

My lvc plan so far has beeen to wait for my voltage alarms to rather loudly alert me that a cell has fallen below 2.7v... Which is just a number I found online somewhere and might me a bad idea? My bms's lvc is 2.1v, and I cannot grasp why they would set it to that, so I never let the bms do the low voltage protection.

The board is just a Boosted board duel+ with about a 2000w draw, and the idea is that this system shares the load with the internal battery and bms. I weight about 160lbs and admit that I commonly push the board to the extream.

The reason why I am relatively sure that is not the batteries (Or atleast the battery model, perhaps my current set is defective) is that the first time I ever attempted this two years ago, I didn't use a bms and just relied on a 4x giant hobby charger to charge everything up, and voltage alarms to let me know when to stop riding. This setup worked better than anything with a bms ever has for several months, but the charging setup was a huge pain, as well as having to turn everything on and off in a specific order and unplug wires to turn off the board. My current setup uses a couple relays to short the far left and far right pins of a MOSFET on the bms to act as an E-Switch, and this is controlled by the built in bms turning on or off. Someone at supower told me that it was safe to short the MOSFET in this way to use as a switch.

Thanks for the help!

 

[alexjthayer]

+1 with dogman.

Boosted board pull about 2000W. Thats why they are still using A123 lifepo4

There is no way cheap HK lifepo4 can handle the load. They probably can handle 200-400w max

Keep in mind that I am using 4 3S packs in series, hobby king advertises a 30c discharge rate that obviously isn't true, but in reality the load is shared with the internal pack, and as long as that isn't dead, I will never experience a current draw above 30amps. I have run with this setup without a bms for a couple months using a hobby charger with no problems, so I'm pretty sure the issue falls on the bms.

 

[Alan B]

I have a couple of 4S HK LiFePO4 4200 Zippy batteries, and with no abuse, using balancing chargers, and no BMS every one puffed. They continue to work ok, but they all puffed for no good reason that we understand, on their own. I know others with the same result, and this is in Ham Radio use with light loads.

 

[alexjthayer]

I have a couple of 4S HK LiFePO4 4200 Zippy batteries, and with no abuse, using balancing chargers, and no BMS every one puffed. They continue to work ok, but they all puffed for no good reason that we understand, on their own. I know others with the same result, and this is in Ham Radio use with light loads.

Well maybe I didn't do anything wrong after all, aside from buying crap batteries. Do you know if you experienced any capacity loss in the puffed cells? Some of mine have lost about 1000mha of capacity after just a few days of being puffed.

 

[Alan B]

I did not test for capacity loss, the way I was using them was for running radios and not discharging them significantly. So they may well have suffered capacity loss or even increases in internal resistance.

I would suggest changing to different batteries.

 

[alexjthayer]

So on the assumption that my batteries are flat out terrible, and puff themselves up without the BMS doing anything wrong, does anyone know how I could end up with a cell charged to 4.40v? I thought the BMS would have cut off all charging when a cell hit the hvc at 3.9v?

 

[kdog]

Am I right in reading that you discharge until the low voltage alarm sounds every time you use them?
This IS a good way to kill your batteries. Read up on lithium care- in a nutshell; don't deep discharge, don't store charged, charge to 90% if possible with a balance every now and again, and charge before use.

 

[alexjthayer]

Am I right in reading that you discharge until the low voltage alarm sounds every time you use them?
This IS a good way to kill your batteries. Read up on lithium care- in a nutshell; don't deep discharge, don't store charged, charge to 90% if possible with a balance every now and again, and charge before use.

Not exactly... I'd say maybe once or twice a week I will run it down until the alarm sounds, but the alarm is not set to the rock bottom lvc on purpose, I have it set to 2.7v so I never run the pack flat. Some of what you are quoting may be true of li-ion batteries, but not lifepo4, for instance they do not have the same issues with being stored charged.

 

[kdog]

Sorry to have to burst your bubble but all Li chemistries have the same requirements. Flatten any Li battery once or twice a week and it will have a premature death. Likewise storing charged is also an early killer-lifepo is no exception. It's just you get a longer cycle life with lifepo naturally, so you don't notice it as quickly. I bet the combo of regular deep discharging, storing charged , and poor qc is leading to premature death oftheweak cells in your pack.

 

[alexjthayer]

Sorry to have to burst your bubble but all Li chemistries have the same requirements. Flatten any Li battery once or twice a week and it will have a premature death. Likewise storing charged is also an early killer-lifepo is no exception. It's just you get a longer cycle life with lifepo naturally, so you don't notice it as quickly. I bet the combo of regular deep discharging, storing charged , and poor qc is leading to premature death oftheweak cells in your pack.

Okay perhaps I need to be more careful to ensure longevity, specifics aside, nothing I've done should cause me to have cells at 2.4v and cells at 4.4v after just a few charge cycles. I tore everything apart last night and cannot figure out what is going on, all of the sense wired meansure the correct voltages at the plug. I plug in the charger, and by the time it goes green it's pushing 4.5v on one cell, obviously not going to balance. Any suggestions for when I get new batteries? I'm afraid I'm going to plug them in and watch the same thing happen.

 

[kdog]

2.7v is a deeeeep discharge. A quick glance of the wiki states lifepo min is 2.8v. Lifepo has a very flat discharge curve... so anything below 3v is essentially empty. Normal operating range is 3- 3.6 V with 3.3 being nominal.
Try setting the Lvc to 3v, and don't ride till it cuts out. Use the upper 3/4 of battery capacity as a very rough rule.
Not sure why the bms let a cell go to 4.5- it's obviously toast as is that cell.
Dead cells with no capacity and internal shorts could easily give you the V discrepancy you are experiencing.
I would start with new cells ( you could salvage existing good ones after testing capacity, IR and self discharge),new bms and a new regime for discharge. I'd also put an external port on the pack if you are bulk charging so you can monitor the cells easily to keep an eye on things.
Hopefully that will solve it, pity about the hip pocket though!

 

[kdog]

Also the bms LVC is usually designed to stop serious damage occurring right there and then. It's usually too low in practice ( regardless of what they state on the packet!) to rely on for an everyday cut off. I'm out of practice with lifepo having gone to 18650s but most Li-ion Bms's I've tested go to about 2.5v which is very very low for Li-ion but above immediate damage level. If I was discharging like this 2-3x weekly I would expect significant deterioration to show up in months. That's why your lifepo bms Lvc is 2.1v

 

[alexjthayer]

2.7v is a deeeeep discharge. A quick glance of the wiki states lifepo min is 2.8v. Lifepo has a very flat discharge curve... so anything below 3v is essentially empty. Normal operating range is 3- 3.6 V with 3.3 being nominal.
Try setting the Lvc to 3v, and don't ride till it cuts out. Use the upper 3/4 of battery capacity as a very rough rule.
Not sure why the bms let a cell go to 4.5- it's obviously toast as is that cell.
Dead cells with no capacity and internal shorts could easily give you the V discrepancy you are experiencing.
I would start with new cells ( you could salvage existing good ones after testing capacity, IR and self discharge),new bms and a new regime for discharge. I'd also put an external port on the pack if you are bulk charging so you can monitor the cells easily to keep an eye on things.
Hopefully that will solve it, pity about the hip pocket though!

Also the bms LVC is usually designed to stop serious damage occurring right there and then. It's usually too low in practice ( regardless of what they state on the packet!) to rely on for an everyday cut off. I'm out of practice with lifepo having gone to 18650s but most Li-ion Bms's I've tested go to about 2.5v which is very very low for Li-ion but above immediate damage level. If I was discharging like this 2-3x weekly I would expect significant deterioration to show up in months. That's why your lifepo bms Lvc is 2.1v

Good to know, I will use 3.0v from now on for the alarms. I'm thinking you are right with the dead cell thing... I split out the 12s pack into the 4 3s packs it is composed of and put one of them onto an rc charger... Cells 1 and 2 took 2800mha more than cell 3 which was showing 4.4v a while ago and is slightly puffed, cell 3 charged to 3.6v in about a minaute, but then dropped to 3.3v within seconds of coming off the charger. I cant imagine how cell 3 would end up legitimately full while the other two are empty. Even so, it is troubling that the bms did not cut off all charging sooner. I'm defiantly going to scarp all of the batteries, but hope to keep the BMS since I just dropped $50 on it, and will be taking what you said about discharge depth vs cell life into account.

That kinda makes sense, but it sure would be useful if there was a BMS with a lvc in the 2.7v-3.0v range, unfortunately I cant find one anywhere short of dropping $300 on a programmable one.

Thanks for the help!

 

[elbotas2700]

I've used batterysupports Bms's quite a few times in the past with no issues but I'm an 18650man so that could be the difference. A bms can only do so much balancing- somewhere around 20ma so if your batteries are really un balanced / different capacity, no normal bms will be able to cope.
Have you done capacity and IR checks on all your previous packs?
The qc on Rc batteries is apparently not great so maybe you got some duds an the bms couldn't keep them balanced then they died.
What is you charge/discharge regime?
I'd start with again with packs that are known equal (or close) capacity, ir and V and that bms. Then see how you go. If something goes wrong I'd be looking at habits, wiring etc.
Have you got a balance charger? Rc batteries are not known for their plug and play ease of use and require extra attention.
Having said that lifepo should be more forgiving than lipo.
Start testin I reckon, get some more info.

Alright well it's good to hear some success stories, I know they were balanced within 0.02v when the charger and bms went on, and although I don't have a proper capacity tester, but I have figured out with my cheap rc charger that now that the batteries puffed anyway, the 3rd cell in each pack has significantly less capacity - maybe around 1000mha, I'm wondering if this could be a manufacturing defect? This appears to be a thing in the two non puffed batteries too so I'm not sure what's up there.

I only have the need to run the whole system below 50% about once a week, and then I run until a low voltage alarm hits 2.7v and notifies me rather loudly. I wish I could just rely on the bms for this but I think it's lvc is lower that sounds healthy for the battery. Then I just plug in the whole system and pretty much leave it until I need it again.

That's sounding like what my plan will be, I do have a cheap hobby charger that I will balance before hand with, but I really want the whole system to be plug and play with one port.

Thanks for the help!

Hey I had a similar problem with my Segway 42 volt bms battery I would take off & the battery voltage all off a sudden dropped to almost completely dead so what I did is add a battery in parallel through the charging port 4 now & battery goes up & down then all of a sudden it would get a boost of power the. Go to a
Most 32 volt drop then back up to 38 volts also I noticed on some battery packs through the charging port the voltage ranges allot in difference to some pack have MV which would not allow me to run in parallel until I found a battery pack with 40 volts through charging port & the thing FLYS by all the othe cyclist lol