huge lipo fire , not cool

I thank the OP for posting.

I have been discharging and charging 20S 13.5Ah on a daily basis, both at work and at home. For me it is good to read threads like this. Performing the same actions twice a day does make me "sloppy" with the safety procedures. I consider this thread a wake up call, and I will revisit my actions to see where I may have become sloppy, and where I can make improvements.

Here is where I got sloppy:

  • Not measuring voltage prior to every charge, to prevent charging a broken cell or overcharging his neighbours.
  • Not disconnecting the charger (iCharger 1010B) overnight, trusting the charger to stop charging.
  • Charging in a wooden work house. It is outside my house, though.

Just to be clear. I will keep using Lipo, and I will try to invent an easier way to comply with the necessary safety requirements.
 
hjns said:
  • Not measuring voltage prior to every charge, to prevent charging a broken cell or overcharging his neighbours.
Click to expand...

Just as a bit of clarification, point one on your list above doesn't tell you much, I'm afraid. The key to charging safety is measuring the cell voltage during charging, so that you can spot the cell that reaches full charge voltage first. This is the cell that will initiate venting action and a possible fire unless some action is taken to prevent its voltage rising above the critical point.

I strongly suspect that there is a lot of misunderstanding around about the relationship between cell voltage and state-of-charge for lithium chemistry cells, as there is frequent mention on here of packs being "in balance" based on random cell voltage measurements made under unknown state-of-charge conditions. Having all cells in a pack within a few tens of mV at some partial state-of-charge is an extremely poor indicator of cell balance within a pack, and may well give false reassurance that the cells are closer balanced in state-of-charge terms than they really are.
 
Arlo1 said:
You use a very safe charger in a safe place like a fire pit both with individual cell monitors!
Click to expand...
If the monitors don't have a way to shut off the charger, then unless you are standing right there to do it when they alarm, it doesn't really matter if you have them or not. ;)

That's where the Methods' setup comes in handy, becuase it can do that for you if you have the HVC cutoff module. -- you just have to remember that his setup is a *last line of defense*, and isn't intended to be depended on to cut off charge or discharge every day. It's just there to stop the process if the regular everyday devices fail to stop it, so that you don't get a fire.


The only "very safe" charger for this kind of stuff is one that itself monitors each individual cell group during charge, and shuts off charge if any of them reaches the cell-level HVC, *and* shuts off charge if any of them is significantly different from any of the others, such as if there is no voltage reading from it at all, or it's say, half a volt different, or maybe a volt.

RC chargers can usually do this, but I wouldn't trust most of them to do it, because they don't seem to actually check that cells are connected to teh balance wires at all! The Accucel6 I have here will charge whether or not balance wires are hooked up, which likely means that if they were hooked up but one of them was broken, it would simply never alert me or stop if the cell it should have been monitoring went overvoltage...becuase it would not know it had happened.


Even if the charger is capapble of this, RC chargers have so many modes and settings that it's probably easy for someone to put it into a different mode if they aren't familiar with how to use them. It sure took me a while to figure out the first one I got (from SoSauty), and I read the manual first before using it, and had read a lot about them in threads by others for months beforehand!
 
Hi AW,

Good points.

In my case, I assume that charging with an iCharger 1010B will indeed cut-off charging at HVC (4.19V cell level) while checking individual cell groups. When I select "balanced charging" it will NOT charge if not all cell groups are connected to the balancing port. However, I can ALSO select "charge", and then it will just charge at whatever voltage one selected until a maximum voltage selected without regarding individual cells.

Furthermore, I also assume that if I check individual cell voltage PRIOR to charging, in case I see a cell at <3.5V, I can at least prevent some obvious bad things from happening. However, I agree that in case all cells are balanced after discharge, it does not necessarily mean charging is safe. This is where I may trust my iCharger too much....

So, here again is the human factor. Apart from the fact that even iChargers can break down.
 
hjns said:
I assume that charging with an iCharger 1010B will indeed cut-off charging at HVC (4.19V cell level) while checking individual cell groups. When I select "balanced charging" it will NOT charge if not all cell groups are connected to the balancing port.
Click to expand...
Don't assume--test it. Disconnect a balance wire while it's charging and see what happens. ;)

I say this because I damaged a NiMH pack (could have had a fire but I was lucky) when a thermistor had a bad connection on a pack. The charger did always check to see if it was connected at start of charge, but after that it didn't care--it'd happly charge right past the point of way too hot to touch instead of stopping way before that point, actually causing the celsl to vent gasified electroliyte (or water rom it, turned to steam)! So one night it's bad connection became no connection after charging started, and overheat and overcharge it did. :(


Also, deliberately unbalance a pack by single-cell charging one cell more than the others, and then balance charge the whole pack, and watch it to see when (or if) it detects the cell going over the limit, and if it shuts off charge or not. Also see if it actively cuts off the pack from the charger, or if it then begins to *drain* the pack instead.

I have a bulk charger that when it shuts off it leaves the outputs in some conditon that begin to drain a couple hundred mA from the pack after charge, which if left overnight could kill a small pack (but mine is 20Ah, so is safe for a few hours). If the charger was a balance-checking type, and shut fof due to cell-HVC or some other condition, long before the pack was anywhere near full, that drainage could kill the cells that weren't full enough. :(


Assumptions are often the cause of fires and other unfortunate situatiions. :(
 
what I saw the photo it has a aluminium casing for the cells pack, it could cause a short circuit due to the cold weather now, water slip in silently 1? Bulk charging , what I think is very important to know every cells say 15 of them are equally same level. Take NO assuming that all the cells are the same ? Who knows Now the cold super weather has drop one cells volts to a very low ? and the other cells are a lot higher. you said is apparently they should be balance ..

while charging inside battery aluminium casing, will have cause some heat to melt ICE - will it be possible ?
I like dogman method of charging, that out all the cells and arrange them properly standing order , check every cells are Okay ? spare no chance to lipo fires, because when it starts it is not easy after to clean and solved the problem very fast. i do exactly like dogman method, take out all the cells and charge them a Safer place.

My guess is ... one cell must be the black sheep at 3.6 or 3.7v and some other cells must be likely 3.9 and some is 4+, and owner ebike must have thinking all is still same level, balance cells.

thanks for putting up this thread for everyone to see, is always HOW important to do the right charging methods. i will not go for any LESS one line method of charging and practice. Im a Noob, Just join lipo club..
 
I had 17Ah of 60V Lipo on my older bike and used to bulk charge it to just under 4.1v and never discharged into the cliff i.e I left 20% in the battery. Balance charged it every week and checked the cells using a cellog 8 before each ride.

But I used to wince and cringe when charging in the home, yes it is cheap and yes it is powerful but you only get about 3 to 400 cycles out of the battery. The amount of duff cells that came with my order added a lot of cost and hassle.

So I went decent Lifepo4 still able to chuck out over 40A continuous and has over 2,000 cycle but the best bit is that I can just put it on the charger and not worry about it bursting into flames. I don't understand why people risk using RC packs.

The worst biut of this is that a certain ebike seller online with lots of bullshit power and speed...range claims sells his bikes with RC Lipo as the power source!!! No way could I sleep at night knowing that I might have sold a bike with RC Lipo to someone who does not understand the upkeep of those batteries.
 
Arlo1 said:
circuit said:
So... No BMS in this fire as well?
How many fires it will take for people to learn?
Click to expand...
99% of BMS's are junk they will MURDER your batteries. So you must be careful with them as well.
Click to expand...
Maybe more like 90%... But overall it's, sadly, true.

Arlo1 said:
Methods boards are the best second level of redundancy but the first should be the person standing there watching or.... You use a very safe charger in a safe place like a fire pit both with individual cell monitors!
Click to expand...
No, can't agree to that. If you leave ANY level of protection to user, it's a recipe for failure. Even if user is very technical and knows what he is doing, he will get sloppy one day. And no, most of you do not know what you are doing if you are using terms like "voltage" and "balance" in one sentence.
Proper charging must be done with proper BMS and nothing else. Ant there must be all secondary and third, fourth layers of protection, such as fuses on each cell (in parallel packs), fuses on interconnections, fire-retardant materials, etc.


By the way, people. Who the hell invented the term "bulk charging"? It is not correct in so many ways.
There must be "normal charging" - all cells in series charging from one simple CCCV charger and WITH PROPER BMS, that can actually do anything about cell voltages and temperature, not just helplessly beep and wait for user's actions.

All other stuff you do here should be called "blind charging" (no BMS), "messy charging" (many parallel connections, playing around with wires or voltmeters each time), or any other type of charging, but not "bulk charging". English is not my primary language, but I see when term is badly out of place.
 
+1 + 1.. i agree, bms I don't really trust that too,. :wink:

there is another thing I even like more, Bulk charging is a nicer way to say, the other way not to say is Blind charging. that is why my greentime charger comes a blow and blew, im okay with it, this is better then getting my whole New house get fire fire. thanks thanks so much to greentime for sending me a spoil charger lose me only usd43 is better then Big fire at my house, that is I buy a proper charger comes with a Lipo safe bag to charge, this lipo fire bag cost only $1.80 .
 
circuit said:
And no, most of you do not know what you are doing if you are using terms like "voltage" and "balance" in one sentence.
Click to expand...

I couldn't agree more. I've lost count of the number of times I've pointed out on here that measuring cell voltage randomly tells you nothing about the balance of the pack.

Still it seems that people are reassured by plugging in a meter of some kind, reading the cell voltages and believing that the pack is balanced if all the voltages are close. This is a dangerous assumption if you use a charging system that doesn't monitor individual cell voltages and shut down when the first cell reaches full charge voltage. I'm near-certain that would have been the cause of this fire now, based on what's been said so far.


circuit said:
Proper charging must be done with proper BMS and nothing else. Ant there must be all secondary and third, fourth layers of protection, such as fuses on each cell (in parallel packs), fuses on interconnections, fire-retardant materials, etc.
Click to expand...

I sort of agree. I think you can shift the charging BMS function to the charger, rather than something permanently wired to the battery pack though, as the key safety feature has to be being able to control the charge current based on individual cell voltages in the pack.

Fuses are certainly useful, but can introduce some additional risks, so a fuse scheme needs to be carefully designed if it is to be effective and not add risk.

circuit said:
By the way, people. Who the hell invented the term "bulk charging"? It is not correct in so many ways.
There must be "normal charging" - all cells in series charging from one simple CCCV charger and WITH PROPER BMS, that can actually do anything about cell voltages and temperature, not just helplessly beep and wait for user's actions.

All other stuff you do here should be called "blind charging" (no BMS), "messy charging" (many parallel connections, playing around with wires or voltmeters each time), or any other type of charging, but not "bulk charging". English is not my primary language, but I see when term is badly out of place.
Click to expand...

I agree, it's a poor term that has slipped into common use. I'm amazed that so many people get away with charging using dumb chargers that don't measure cell voltages. Given the large number of people that do this, all I can conclude is that the cells we use are more tolerant than we might think, else we'd see far more fires.....................
 
Since we started to talk about current RC Li battery chargers with cell monitoring, I just wanted to highlight the A123 recommendation as it might apply here:
Pack must have dual, redundant over voltage protection, with at least protection by hardware and one via software
Click to expand...
In Implementation this would mean having a micro control a LTC6802-2 (software control) AND have like a MAXIM11081 (hardware control) watch the same battery during both charge and discharge. Either system being able to vote by hardware to terminate charge/discharge. Note that this is the architecture being taken by the industry on large format Li usage in vehicles. Also note that our hobby chargers only use software control.

Now software control is better than no control, but the industry is saying redundant control is necessary to adequately control the hazard of over charge and under discharge of large format Li Batteries.

The problems are clearly statistical. Some of us have had no problems what so ever. Sort of like the gal smoking and pumping gas. Some have caught the tail of the dragon as it sweeps by. Kudos to Amberwolf and Jeremy for starting to archive these incidents. Real data puts the observable statistics into perspective. For me, the extreme consequence of a undetected and uncontrolled fire in our homes warrants a continued diligence in this area.
 
I sure have tried to blow up a lot of different kinds of battery cells by overcharging them.

The most easy ones to get to blow up by overcharging are rc lipo cells.

Never been able to get a laptop cell to explode by overcharging.

Got a bootleg A123 cell to explode by overcharging.

I say if you can get a cell to explode by overcharging it then move on and find a better cell for your ebike build.

It is not that easy to get quality li ion to burn by overcharging.
 
amberwolf said:
Arlo1 said:
You use a very safe charger in a safe place like a fire pit both with individual cell monitors!
Click to expand...
If the monitors don't have a way to shut off the charger, then unless you are standing right there to do it when they alarm, it doesn't really matter if you have them or not. ;)

That's where the Methods' setup comes in handy, becuase it can do that for you if you have the HVC cutoff module. -- you just have to remember that his setup is a *last line of defense*, and isn't intended to be depended on to cut off charge or discharge every day. It's just there to stop the process if the regular everyday devices fail to stop it, so that you don't get a fire.


The only "very safe" charger for this kind of stuff is one that itself monitors each individual cell group during charge, and shuts off charge if any of them reaches the cell-level HVC, *and* shuts off charge if any of them is significantly different from any of the others, such as if there is no voltage reading from it at all, or it's say, half a volt different, or maybe a volt.

RC chargers can usually do this, but I wouldn't trust most of them to do it, because they don't seem to actually check that cells are connected to teh balance wires at all! The Accucel6 I have here will charge whether or not balance wires are hooked up, which likely means that if they were hooked up but one of them was broken, it would simply never alert me or stop if the cell it should have been monitoring went overvoltage...becuase it would not know it had happened.


Even if the charger is capapble of this, RC chargers have so many modes and settings that it's probably easy for someone to put it into a different mode if they aren't familiar with how to use them. It sure took me a while to figure out the first one I got (from SoSauty), and I read the manual first before using it, and had read a lot about them in threads by others for months beforehand!
Click to expand...
Only problem with Methods boards is they are a second level of defense you should not rely on them to shut the charger off, I think... (im going to check now) they cut at 4.29v so that's higher then I want to charge.

What I do is. I plug in all of methods boards hooked up to circuit breakers to stop the charger all this stays on the bike. Then I plug in cell monitors from HK then I hook up the bulk charger. I then make sure its in a safe place for a fire. The batteries are in a 3/16 thick aluminum box. then I stay near it while I work on something usually this is at my shop so I can fix motorcycles or work on my projects while its charging. I take note of the condition of the cells a few times thought the process so if something happens I can hope to spot it early. I here the beeping cell monitor then I stop charging. If I notice the pac is out of whack at this time I will balance using my balance charger I got from DX always staying near it to make sure all is well. Also having methods boards on you're system will save the batteries during LVC by pulling back the throttle. Which should help you to prevent your cells from getting out of whack in the first place.
 
Usually when there is a lipo fire, it is pretty easy to determine what the issue was. In this case I am not sure we have determined the causality.

We know :

Age - 2 years ?
Cycles - around 400 - this is probably high for lipo.
Batteries were 20C - 15/20 C batteries seem to have a higher catastrophic failure rate than 30C+ batteries. Maybe they are just more commonly used?

In this case we have not received a couple key pieces of info:

* Was the lipo abused (overcharged / overdischarged) during it's lifetime?
* what kind of charger was used?
Nominal Voltage?
Charge Current?
Auto shutoff?

Thanks,

-Warren.
 
Personally I still like my DC-DC converter charge system. The units themselves are high reliability devices, with over-voltage and over current protection, plus over-temperature protection (I bought them surplus for a song, but the retail price was around $50 per converter) . They are probably an order of magnitude more reliable than a typical commercial LiPo charger. The big downside is that they give no indication of the pack being fully charged, I have to measure the voltage of each cell group to see whether they've reached termination voltage.

The system does give me a fair degree of peace of mind, though, as it is extremely unlikely that any of the converters would exceed the set voltage (4.14V in my case), just because of all the built-in safety features and the inherent high reliability of the design. There is another safety feature too, in that each cell is being charged by an electrically isolated individual charger, which slightly reduces the chances of a cascade type failure.
 
etriker said:
I say if you can get a cell to explode by overcharging it then move on and find a better cell for your ebike build.
Click to expand...


This means, no NiCd, No Lead Acid, no NiMH, and no Lithium batteries of any type.

Also, as an FYI, all rechargeable lithium batteries we use on ebikes are lithium ion batteries (LiFePO4, LiCoO2, LiMnO4, LiNiMnCoO2, all modern types using a polymer saturated with electrolyte). Some are rolled up and stuffed in a can, some are sealed in a pouch.

The part that is sketching me out is the QC of RC LiPo seems to be getting worse than it started. While the formulas/recipes themselves gain better safety, the QC of the product seems to be spiraling down, and it wasn't quite high enough to begin with.

In this case, he had 2 years of good service from the pack, and THEN it burned while charging. It seems unlikely that QC issues were to blame in this case, unless it was some strange defect that wouldn't show any symptoms until 2 years in. A single fatigued foil joint that opens and takes one of the packs out of parallel with the others can cause a fire when bulk-charging. As can a number of other issues.

Designing a battery to be reliable and safe is an massive undertaking. Many hundreds of thousands of dollars need to be spent in vibration testing, thermal shock/cycle testing, moisture/humidity/salt-spray testing, UV degradation testing, short-circuit testing, overcharge protection testing, crash/impact/shock safety testing, and then a few hundred grand on BMS development to work for your specific application.

We generally buy the cheapest $/Wh, find the absolute least effort path to get it to stick to the bike (duct-tape, velcro straps, a cargo rack or a bend of sheet metal or two), and go run the piss out of it and generally ignore any monitoring or safety checks on it after the first few months. I'm not saying I'm any better, I have done and still do the same thing, but I don't disillusion myself into thinking it wont possibly burn as a result someday.
 
Still relatively new to this but I'm going to put my say in, if only to learn. Internal resistance. Is there not a device that can decide for the simple folk that a cell has reached it's usable lifespan and needs to be taken out? As simple as a LCD readout of individual packs (or cells) of voltage readings under load and normal use. LVC's are nice, but to see what's actually happening at the individual cell level would be a better option (you'd need some amount of electrical knowledge to understand and use it though).

The bulk charger. This is what is scaring my newb mind. How many preventative measures are in place in a bulk charger to keep things from running away and into dangerously hot territory?

It's simply a disaster waiting to happen. You ride like a pro and undetected by you, you've punctured or damaged the battery in some manner. You place it on your high performance 1-2 hour charger for your next run and next thing you know, SIZZLE!
 
liveforphysics said:
etriker said:
I say if you can get a cell to explode by overcharging it then move on and find a better cell for your ebike build.
Click to expand...


This means, no NiCd, No Lead Acid, no NiMH, and no Lithium batteries of any type.

Also, as an FYI, all rechargeable lithium batteries we use on ebikes are lithium ion batteries (LiFePO4, LiCoO2, LiMnO4, LiNiMnCoO2, all modern types using a polymer saturated with electrolyte). Some are rolled up and stuffed in a can, some are sealed in a pouch.

The part that is sketching me out is the QC of RC LiPo seems to be getting worse than it started. While the formulas/recipes themselves gain better safety, the QC of the product seems to be spiraling down, and it wasn't quite high enough to begin with.

In this case, he had 2 years of good service from the pack, and THEN it burned while charging. It seems unlikely that QC issues were to blame in this case, unless it was some strange defect that wouldn't show any symptoms until 2 years in. A single fatigued foil joint that opens and takes one of the packs out of parallel with the others can cause a fire when bulk-charging. As can a number of other issues.

Designing a battery to be reliable and safe is an massive undertaking. Many hundreds of thousands of dollars need to be spent in vibration testing, thermal shock/cycle testing, moisture/humidity/salt-spray testing, UV degradation testing, short-circuit testing, overcharge protection testing, crash/impact/shock safety testing, and then a few hundred grand on BMS development to work for your specific application.

We generally buy the cheapest $/Wh, find the absolute least effort path to get it to stick to the bike (duct-tape, velcro straps, a cargo rack or a bend of sheet metal or two), and go run the piss out of it and generally ignore any monitoring or safety checks on it after the first few months. I'm not saying I'm any better, I have done and still do the same thing, but I don't disillusion myself into thinking it wont possibly burn as a result someday.
Click to expand...

I used a 8s 6p A123 m1 cell homemade pack to get to work today designed to make it to my job and back on a single charge.

It was balanced charged with a Hyperion charger.

Used the most forgiving cells I could find and the best charger I could find to charge them.

That is my style.
 
etriker said:
I used a 8s 6p A123 m1 cell homemade pack to get to work today designed to make it to my job and back on a single charge.

It was balanced charged with a Hyperion charger.
Click to expand...


Here is an A123 pack being properly charged by a guy with many years of battery pack experience on a much higher-end charger than your hyperion catching fire and burning.

http://www.rcgroups.com/forums/showthread.php?t=772399
 
liveforphysics said:
etriker said:
I used a 8s 6p A123 m1 cell homemade pack to get to work today designed to make it to my job and back on a single charge.

It was balanced charged with a Hyperion charger.
Click to expand...


Here is an A123 pack being properly charged by a guy with many years of battery pack experience on a much higher-end charger than your hyperion catching fire and burning.

http://www.rcgroups.com/forums/showthread.php?t=772399
Click to expand...

Right ? Here is me burning one too.

http://www.youtube.com/watch?v=YVSZHdyWmIo&feature=plcp

I know they will burn. :)

I am a battery pyro too ! :)

I do my battery pyro stuff way outside in the yard though.
 
Ok not saying anything about one or the other being better BUT...

It seems like of all the chemistries people use in this forum the one being talked out as SAFER is A123. Yes they are bulky and are hard to find and expensive but there must be a reason why so many people bought packs from cellman to the point where he ran out of cells to build them.

I've heard he is doing NMC packs now! This brings another player to the game :D
 
migueralliart said:
Ok not saying anything about one or the other being better BUT...

It seems like of all the chemistries people use in this forum the one being talked out as SAFER is A123. Yes they are bulky and are hard to find and expensive but there must be a reason why so many people bought packs from cellman to the point where he ran out of cells to build them.

I've heard he is doing NMC packs now! This brings another player to the game :D
Click to expand...

I very strongly suspect that pack build quality, battery management strategy and the safety features in the charging system have a massively greater influence on battery pack fire or explosion risk than cell chemistry. I'll go further, by suggesting that a badly built or managed LiFePO4 pack (such as A123 cells) will be significantly more dangerous than a well built and managed LiCoO2 cell pack (say cheap HK LiPo).

For the majority of DIY pack builders it is their skill, care and knowledge that is going to determine battery pack safety, not the chemistry of the cells. In fact I doubt the chemistry of the cells is even really important in practice.
 
My question is, will all LiPo packs eventually end up in flame if cycled enough times? Should they be retired after 250cycles etc? Would it still be safe to ride if a 10Ah pack if it is only giving 4Ah of energy, as long as it stayed in voltage range?

AND how do i know when it's time to retire my cell_man A123 Pack, I don't want it exploding between my legs while riding. I guess i just trash it when the 11.5Ah battery starts only gives me 6Ah or so?
 
Is insurance going to fix the house? If for some reason you are going to try to fix it yourself, I'd suggest you rent an ozone generator. Worked good on my stuff that got smoke damaged.
 
You must log in or register to reply here.
Back
Top