Lipo failure

cfrgsn1

1 mW
Joined
Jun 28, 2015
Messages
10
Location
brisbane, australia
Hi, just wanted to point out once again that you can never be to careful when using lipos. I just had a 24s 5ah lipo pack burst into flames while charging. To charge i split the pack down to 6s4p and balance charge using a rc charger. I was in the shed while it was charging outside, when i heard the charger make it's charging is finished noise, walked outside to hear the pack hissing, while the voltage on the charger read 23.84v. I quickly unplugged the charger and wheeled the bike into the middle of the yard. Turned back to get the hose and withing 10 seconds there was a roaring fire with a 1m jet of flame coming from the battery.

It took about 10 minutes for the fire to go out, even with the hose to try and limit the damage to the rest of the bike. Damage to the bike was minimal, with only the back tire melting, and the brake/gear cables being destroyed. Thankfully the controller and motor weren't damaged. I only have a vague idea what caused it, the batteries were discharged to around 22.3v per 6s prior to charging, however they have been pushed pretty hard, being a few years old, and recently changing from a 48v 20amp controller to the 96v 65a controller.

Anyway, if this had happened inside, the results would have been disastrous, I thought that i was following the lipo rules, and then this lipo pack exploded without much warning, no swelling, no overvoltage, and no physical damage to the pack. A possible thought would be that the pack temporarily over discharged while trying to draw so much current from it.
 
Sorry for your loss. Good no one got hurt.

RC-Lipo have a long history of making ebikes a bad reputation.

Time to upgrade to some new 18650 with CID protection.
 
Was the battery balanced and kept balance all through it's lifetime?

Did you do any discharge-charge cycle graphing or manual tracking of the balance before you put it on the bike and used it?
 
Yep, now to start the long process of looking and building a new battery pack, hopefully this time it won't catch fire! I balance charged the batteries every time, although some of the connectors were a little bit dicey, so there is a possibility that it didn't balance properly. I did check every cell occasionally with a multimeter and they were all the same within 0.1v ( the resolution of my meter). I had only been using the pack in its 96v configuration for 3 days, with a maximum 5 charge discharge cycles, so i assume this had damaged the batteries in some way, as I had at least 2 years running them in the 48v pack.
 
Did you check with a multimeter at full charge or almost entirely drained ( 3.5v or below ) ? because that's where the majority of disbalances show their ugly faces.

23_chargingmechanics.gif


^-- what a good pack looks like on a graphing charger.

badturnigy.gif


^-- what a bad pack looks like on a graphing charger. Notice the purple cell hits the lowest voltage first? and then rises the highest first as well, after the nominal voltage ( 3.7-3.8 ) ? take this pack down to just 3.5v average and you will see two cells dip into below 3 volt territory as well.

Any chance that you had a weak cell that you overdischarged, ie below 3.0v? this can do serious damage to a pack over time and turn it into a potential firebomb over time.

If a pack takes a really, really, really long time to balance ( hours ), this is usually an indicator that you've got a bad pack in the mix.

Yup, lipo's dangerous. It's kind of like owning a super crazy fully automatic weapon. They're awesome, but with great output power comes great responsibility to provide them meticulous care. I can't recommend them to the average person at all, but still use 'em myself - only because graphing them periodically ( once a year ) means safety to me thus far.
 
I'm pretty sure i checked them when fully charged, which might have been part of this packs downfall, and there's every possibility that one of my cells had gone bad and i just didn't pick it up. I will definitely consider a safer battery technology for the next build. However all the alternative battery types seem to lack the cheapness, weight and power of lipos, as well as the ease of building the pack. Do you think maybe using a battery medic or the like would catch these failing cells before it resulted in a fireball?
 
Graphing periodically, or having a charger that shows you cell balance in real time is a big help to increase safety.
Battery medics have hit and miss accuracy, and so do cheap chargers, which is why i don't advocate going cheap on the thing that could mean the difference between years of happy lipo use and.. well, kaboom.

Yeah, it's unfortunate that many high power options out there are either rare or incredibly expensive. There are some fantastic 18650 cells out there which have great balances of power and capacity today, edging out RC Lipo in lightness by far for lower power builds. But yeah, it's a pain in the ass to build them. It's another reason that i am still stuck in lipo land here. ( i have a battery that confidently outputs 100A and fits in my triangle bag - that's awesome. )
 
I think it might be wise for me to invest in a better charger then, i'm pretty sure my current charger is not the best quality around. Any recommendations on a good brand of charger/ something to accurately graph the discharge of each cell?
 
Bummer, that's why I could never recommend them to anyone getting into Ebikes. Plenty of safer battery chemistries out there and services that will put a pack together for you.
 
I do it OCD style, i use a Cellog to check cell voltages before every single charge cycle and check them again when charging is complete, sometimes mid-way thru the cycle.

If i come back from a ride that burns over 50% of my rated capacity i check the voltages when i step off the bike at home..

every.. single.. time.. no exception. i NEVER charge or head out for a ride without checking my voltages. it only takes a few seconds and when a bad cell happens, you know it right away.
 
I hear many thing that are hurting my conception of safe lipo maintenance.

"I was in the shed while it was charging outside"

"been pushed pretty hard, being a few years old"

"around 22.3v per 6s prior to charging"

"I thought that i was following the lipo rules, and then this lipo pack exploded without much warning"

This telling me that you don't monitor every cell while charging, that you are still using an old lipo pack and even up the Amps, and that you are letting them to charge out of your immediate presence and constant monitoring. Obviously you were not following the rules, and you were not there to notice the smell and heat that sure started long enough before, to prevent the fire.

I would say sorry for your loss, but there is none for thoses lipo's have had a long life already.
I believe it's a good thing that happened to you. Now you decide if you are ready to apply all the safety measures using RC Lipo, or get a safer chemistry.
 
+1
22.3V per 6s pack isn't too low imo, but 65A on a 5ah pack is really pushing it hard in the middle of summer. That's 13C if my math is right. What was the brand and specs on the lipo?
 
Ypedal said:
I do it OCD style, i use a Cellog to check cell voltages before every single charge cycle and check them again when charging is complete, sometimes mid-way thru the cycle.

If i come back from a ride that burns over 50% of my rated capacity i check the voltages when i step off the bike at home..

every.. single.. time.. no exception. i NEVER charge or head out for a ride without checking my voltages. it only takes a few seconds and when a bad cell happens, you know it right away.

I only use my old lipo for my lawnmower , but I still do the exact same thing. Check each cell prior to putting on the charger, during and when it comes off to make sure the cells are all happy. Just went thru this procedure before coming inside as Im about to mow my grass.

It sounds like you had a cell out of alignment and then charged it. I thought I read you mentioned you 'thought' you checked it. IMO lipo is one of those things where there is no room for error and you should know beyond a fact that you checked it because you check it every single time.

Glad the damage was minimal. GL on the new battery.
 
A couple things here;

First, I get so annoyed when people say "Never drain a lipo cell below 3 volts". Holy cow! If you are draining a Lipo that far, you are ruining it! So, I am glad the OP was keeing the voltage relatively high before a recharge.
Second, there are ceratin things that make a pack flame out;

Older cells with high internal resistance tend to go into thermal runaway.
Damaged cells are, obviously, dangerous.
Over discharging a pack is what throws cells into imbalance.

I have been running Lipo for over 10 years now in RC and bike applications. I have only ever seen one pack go out of balance more than .1 volt and that was from over discharging. You do not need to balance a Lipo pack with every cycle (though I understand why people do it). I check my packs with a hand held Battery Medic every so often and I never need to balance them. I stop charging my 12S packs at 49.9 volts and I typically recharge them when they drop to 46 volts at rest (44 volts under load at the lowest).

I have my LV cutoff set for 42 volts, but I never hit that threshold. I make sure I build every bike with far more battery than I will ever need so I do not find myself running the pack low to get home.

All of that said, anyone can have an issue. Heck, many cars catch fire in the garage. Yet, Lipo gets such a bad rep. ANY battery chemistry has spontaeous fire cases. Lipo are quite stable at this point. The early Lipo and Lithium Ion cells were VERY dangerous, indeed. That was from their very high internal resistance and poor cell matching, though.

Matt
 
It's important to state '3 volts' on a per cell level, not an average level. 3 volts per cell average of course means that some cells are way below 3v, yes.

24_dischargingmechanics.gif


Cells with high internal resistance show their true faces during charging or discharging with a graphing discharger.

Here is an example of a totally friggin' thrashed pack on discharge:

leadershobby_sag2.gif


I cover this all in my lipo tutorial in the battery section. Too bad most people won't read it :lol:

Anyway, i don't balance charge either, except once a year. When i get packs from hobbyking, i cherry pick them and never put a pack that is wonky into use. Out of a batch of 10 packs, you can expect 1 or 2 to be imperfect. Most people don't do this, and they run into problems down the line.

My favorite graphing charger is the iCharger line, second choice would be hyperions.. there are many other brands that do graphing as well, and almost all RC Chargers are compatible with the logview software that generated the graphs i've posted here.

recumpence said:
A couple things here;

First, I get so annoyed when people say "Never drain a lipo cell below 3 volts". Holy cow! If you are draining a Lipo that far, you are ruining it! So, I am glad the OP was keeing the voltage relatively high before a recharge.
Second, there are ceratin things that make a pack flame out;

Older cells with high internal resistance tend to go into thermal runaway.
Damaged cells are, obviously, dangerous.
Over discharging a pack is what throws cells into imbalance.

I have been running Lipo for over 10 years now in RC and bike applications. I have only ever seen one pack go out of balance more than .1 volt and that was from over discharging. You do not need to balance a Lipo pack with every cycle (though I understand why people do it). I check my packs with a hand held Battery Medic every so often and I never need to balance them. I stop charging my 12S packs at 49.9 volts and I typically recharge them when they drop to 46 volts at rest (44 volts under load at the lowest).

I have my LV cutoff set for 42 volts, but I never hit that threshold. I make sure I build every bike with far more battery than I will ever need so I do not find myself running the pack low to get home.

All of that said, anyone can have an issue. Heck, many cars catch fire in the garage. Yet, Lipo gets such a bad rep. ANY battery chemistry has spontaeous fire cases. Lipo are quite stable at this point. The early Lipo and Lithium Ion cells were VERY dangerous, indeed. That was from their very high internal resistance and poor cell matching, though.

Matt
 
Repeated connecting/disconnecting combined with "connectors were a little bit dicey".

Please change the threat title. We should never combine lipo with the other word, because is is counterproductive for the cause since the public has an inaccurate impression of the overall risk with lithium batteries. Instead let's borrow a quote from Elon Musk regarding the explosion when trying to land a SpaceX rocket, and always refer to these events as "rapid unscheduled disassembly" of a lipo battery pack.

In fact, with the forum software that deals with cursing, can't the forum do that automatically for every instance of "lipo fire" or "battery fire"?
 
neptronix said:
It's important to state '3 volts' on a per cell level, not an average level. 3 volts per cell average of course means that some cells are way below 3v, yes.

24_dischargingmechanics.gif


Cells with high internal resistance show their true faces during charging or discharging with a graphing discharger.

Here is an example of a totally friggin' thrashed pack on discharge:

leadershobby_sag2.gif


I cover this all in my lipo tutorial in the battery section. Too bad most people won't read it :lol:

Anyway, i don't balance charge either, except once a year. When i get packs from hobbyking, i cherry pick them and never put a pack that is wonky into use. Out of a batch of 10 packs, you can expect 1 or 2 to be imperfect. Most people don't do this, and they run into problems down the line.

My favorite graphing charger is the iCharger line, second choice would be hyperions.. there are many other brands that do graphing as well, and almost all RC Chargers are compatible with the logview software that generated the graphs i've posted here.

recumpence said:
A couple things here;

First, I get so annoyed when people say "Never drain a lipo cell below 3 volts". Holy cow! If you are draining a Lipo that far, you are ruining it! So, I am glad the OP was keeing the voltage relatively high before a recharge.
Second, there are ceratin things that make a pack flame out;

Older cells with high internal resistance tend to go into thermal runaway.
Damaged cells are, obviously, dangerous.
Over discharging a pack is what throws cells into imbalance.

I have been running Lipo for over 10 years now in RC and bike applications. I have only ever seen one pack go out of balance more than .1 volt and that was from over discharging. You do not need to balance a Lipo pack with every cycle (though I understand why people do it). I check my packs with a hand held Battery Medic every so often and I never need to balance them. I stop charging my 12S packs at 49.9 volts and I typically recharge them when they drop to 46 volts at rest (44 volts under load at the lowest).

I have my LV cutoff set for 42 volts, but I never hit that threshold. I make sure I build every bike with far more battery than I will ever need so I do not find myself running the pack low to get home.

All of that said, anyone can have an issue. Heck, many cars catch fire in the garage. Yet, Lipo gets such a bad rep. ANY battery chemistry has spontaeous fire cases. Lipo are quite stable at this point. The early Lipo and Lithium Ion cells were VERY dangerous, indeed. That was from their very high internal resistance and poor cell matching, though.

Matt

Taking any single lipo cell down to 3 volts will harm the cell. The 3 volt limit that is constantly thrown around is just a waste of time and misleading.
 
Thanks for all the helpful posts, I realize that my level of care with charging was not perfect, this was more of a post to remind people that complacency happens, and a bit of carelessness can result in a potential major disaster. This is why i charge the bike outside, where it's not going to cause much damage to anything else. I was less than 3 meters away from the bike. I think that even if i was standing over it, by the time i realized what was happening, it would have been to late to save them.

The batteries were 20c, so were defiantly not quite up to the challenge. It's the middle of winter down here in australia, and during the bike rides they did not get hot, only mildly warm, and i let them cool before charging again. I will be sure to have a way to monitor each cell for next time then, as there was no outward signs from the battery of impending failure.

On a different note, how would you dispose of the burnt battery pack? there are still quite a few intact cells, and several semi burnt ones. I am yet to move the pack from where it caught fire in case i provoke more flames from this thing.
 
If you were getting them warm in the winter, you were getting them pretty hot in the summer i bet.
Yeah, lipos don't particularly like that either. Neither do many other batteries.
 
Yucky...

A bucket of salty water to submerge them is often recommended, puncture each cell that is not already breached ( carefully obviously ) and then dunk in salt water for a few days.. then they can be thrown out safely.

I have a whole backyard bunker of packs that i need to dispose of as well.. was waiting for the lawn to grow and dry leaves from last fall to be done with before i " experiment " .. *rubs hands in evil villain gestures..
 
cfrgsn1 said:
However all the alternative battery types seem to lack the cheapness, weight and power of lipos, as well as the ease of building the pack.

Very true! However, those other chemistries also lack the extreme desire to undergo runaway thermal decomposition and subsequent meter-long jets of flame. So, you know, pick what matters to you.

If you're looking for a somewhat unsafe but high power density chemistry, LiCo 18650s are a good option. For safer chemistries, LiFePO4 is still hard to beat, though LiMn and some of the other new chemistries are reasonable.

Whatever you build should absolutely include a BMS with per-cell low voltage cutoff, per-cell high voltage cutoff, amperage limits, etc.

I personally think that hobby lipo based ebike battery packs are unreasonably hazardous and shouldn't be used by anyone at this point, but I'm just a random guy interested in not seeing a wave of apartment buildings and such burned down by ebike fires. That gets the attention of insurance agencies in a hurry, and leads to things like ebikes being banned by rental agreements in apartments. Not particularly useful, long term...

John in CR said:
Please change the threat title. We should never combine lipo with the other word, because is is counterproductive for the cause since the public has an inaccurate impression of the overall risk with lithium batteries. Instead let's borrow a quote from Elon Musk regarding the explosion when trying to land a SpaceX rocket, and always refer to these events as "rapid unscheduled disassembly" of a lipo battery pack.

In fact, with the forum software that deals with cursing, can't the forum do that automatically for every instance of "lipo fire" or "battery fire"?

Except that the vast majority of battery pack fires seem to involve lipo packs! Yes, things can go wrong with other chemistries, but lipo packs are far and away the most likely to catch fire.

BionX, as an example, uses LiMn cells. I find zero hits on Google for BionX Battery Fire that refer to an actual fire - just general "how to keep your ebike from catching fire" type stuff.

Search for Lipo Battery Fire, and you get pages upon pages and thousands of videos of such things burning violently.

Stop. Using. Lipo. Packs. They're entirely unsafe to use in electric bicycles, and it's genuinely hazardous to the long term viability of electric bicycles.
 
Syonyk said:
cfrgsn1 said:
However all the alternative battery types seem to lack the cheapness, weight and power of lipos, as well as the ease of building the pack.

Very true! However, those other chemistries also lack the extreme desire to undergo runaway thermal decomposition and subsequent meter-long jets of flame. So, you know, pick what matters to you.

If you're looking for a somewhat unsafe but high power density chemistry, LiCo 18650s are a good option. For safer chemistries, LiFePO4 is still hard to beat, though LiMn and some of the other new chemistries are reasonable.

Whatever you build should absolutely include a BMS with per-cell low voltage cutoff, per-cell high voltage cutoff, amperage limits, etc.

I personally think that hobby lipo based ebike battery packs are unreasonably hazardous and shouldn't be used by anyone at this point, but I'm just a random guy interested in not seeing a wave of apartment buildings and such burned down by ebike fires. That gets the attention of insurance agencies in a hurry, and leads to things like ebikes being banned by rental agreements in apartments. Not particularly useful, long term...

John in CR said:
Please change the threat title. We should never combine lipo with the other word, because is is counterproductive for the cause since the public has an inaccurate impression of the overall risk with lithium batteries. Instead let's borrow a quote from Elon Musk regarding the explosion when trying to land a SpaceX rocket, and always refer to these events as "rapid unscheduled disassembly" of a lipo battery pack.

In fact, with the forum software that deals with cursing, can't the forum do that automatically for every instance of "lipo fire" or "battery fire"?

Except that the vast majority of battery pack fires seem to involve lipo packs! Yes, things can go wrong with other chemistries, but lipo packs are far and away the most likely to catch fire.

BionX, as an example, uses LiMn cells. I find zero hits on Google for BionX Battery Fire that refer to an actual fire - just general "how to keep your ebike from catching fire" type stuff.

Search for Lipo Battery Fire, and you get pages upon pages and thousands of videos of such things burning violently.

Stop. Using. Lipo. Packs. They're entirely unsafe to use in electric bicycles, and it's genuinely hazardous to the long term viability of electric bicycles.

The vast majority of major fires have to do with other chemistries. Lead acid batteries have been causing fires and deaths for a century.

Stop spreading untruths. Lpf did some very interesting experiments regarding this. He actually had a difficult time getting lipo packs to blow up and found it relatively easy to blow up Headway cylindrical cells.

Matt
 
Back
Top