DIY Electric Car Forums banner
1 - 20 of 29 Posts

·
Registered
Joined
·
5 Posts
Discussion Starter · #1 ·
Hi Guys,

I have looked high and low and not seen evidence of LiFePO4 batteries catching fire?

I have seen & read of BMS systems that have caught fire or shorts etc. but do any of you know of any cases of LiFeP04 cells catching & sustaining a fire like the Boeing incident, or like the Sony computer Li batteries? Can anyone point me to these incidents if in fact they have happened?

Seems "li" technology takes a hit in circumstances like the Boeing incident and that would be unfair to other Li chemistry batteries if they do not suffer the same types of catastrophic events.

Thanks,

CMS
 

·
Registered
Joined
·
4,005 Posts
I think Jack R. burned some A123s on a test bench. He blamed it on a BMS but didn't really explain it.

There was also the BYD taxi fire, but not much won't burn if you try to cut it in half with a tree after a 100+ mph crash.

And the 16 Karmas from TS Sandy plus shady details in other Karma fires.

Oh, and some guy likes to burn wrenches with the new CALB CAs.

I don't read much into ANY specific example. There's cell chemistry, packaging, BMS, enclosure, charge/discharge profile, HVAC, and natural disasters all to consider.
 

·
Registered
Joined
·
263 Posts
I think Jack R. burned some A123s on a test bench. He blamed it on a BMS but didn't really explain it.

There was also the BYD taxi fire, but not much won't burn if you try to cut it in half with a tree after a 100+ mph crash.

And the 16 Karmas from TS Sandy plus shady details in other Karma fires.

Oh, and some guy likes to burn wrenches with the new CALB CAs.

I don't read much into ANY specific example. There's cell chemistry, packaging, BMS, enclosure, charge/discharge profile, HVAC, and natural disasters all to consider.
Jack burned some A123 cells by putting them on a power supply to charge them and then got involved with something else, forgot about them, and they eventually went up in smoke. They looked very much like Jiffy-Pop. I've never heard of a non-bms lipo4 pack that's been burned up while attached to a working charger. If anyone has experience with this would be good to know.

I don't have any knowledge of the boeing fires but I'd bet they had some sort of bms attached to them, almost all "sophisticated" systems do.
 

·
Registered
Joined
·
447 Posts
I have looked high and low and not seen evidence of LiFePO4 batteries catching fire.
Fire? Not exactly. More like explosions and or lots of white smoke. But not actually flames.

I have seen it happening, and I have seen the results after the fact. In all cases, it was due to abuse of the battery. Some examples.

  • Shorted the 2 battery output cables during manufacture (seen it twice)
  • Two shorts to chassis, completing the circuit (seen it once)
  • Loose battery connection, arching, nearby combustible materials catch on fire, heat pops the LiFePO4 cells (seen it once)
  • Outboard charger left on overnight, no BMS control (seen it once)
  • Boat captain connecting over-discharged and damaged Li-ion batteries in parallel to fully charged ones (seen it once)
  • On purpose, for testing
like the Boeing incident
The Boeing Dreamliner incident was not LiFePO4; it was LiCoO2 (standard Li-ion).

I have ... read of BMS systems that have caught fire.
Not once. Every report I have read about that was generated by a party which may have had some interest in saying so. I am not saying that it's impossible, I am just saying that I have no reliable knowledge of any such case.

But you say that you have seen it yourself: by all means, please elaborate (with verifiable details, if possible), as I really need to know about it.
 

·
Registered
Joined
·
5 Posts
Discussion Starter · #5 ·
Fire? Not exactly. More like explosions and or lots of white smoke. But not actually flames.

I have seen it happening, and I have seen the results after the fact. In all cases, it was due to abuse of the battery. Some examples.

  • Shorted the 2 battery output cables during manufacture (seen it twice)
  • Two shorts to chassis, completing the circuit (seen it once)
  • Loose battery connection, arching, nearby combustible materials catch on fire, heat pops the LiFePO4 cells (seen it once)
  • Outboard charger left on overnight, no BMS control (seen it once)
  • Boat captain connecting over-discharged and damaged Li-ion batteries in parallel to fully charged ones (seen it once)
  • On purpose, for testing


The Boeing Dreamliner incident was not LiFePO4; it was LiCoO2 (standard Li-ion).



Not once. Every report I have read about that was generated by a party which may have had some interest in saying so. I am not saying that it's impossible, I am just saying that I have no reliable knowledge of any such case.

But you say that you have seen it yourself: by all means, please elaborate (with verifiable details, if possible), as I really need to know about it.
Davide,

Thanks for chiming in.. I have not "seen" any BMS cause a fire. I have read "on the net" where people claim the fire was cause by a home made BMS or the shunt got too hot etc. etc...

So these batteries can have issues, but it seems, from your examples, you really need to do something to move it along to that level... When you say "explosions" what precipitated those events?
 

·
Registered
Joined
·
447 Posts
So these batteries can have issues
Not as many as LiCoO2. LiFePO4 can handle much higher temperatures before going ballistic (literally) than LiCoO2. That's why LiFePO4 cells are considered safer.

When you say "explosions" what precipitated those events?
The list I have above. For example, the first item. Numbnuts engineer runs power cable from battery through uninsulated 1" braids. Braids touch each other. Loud pop. White smoke. Crackling.
Battery is on a wheeled cart, next to a sliding door to the outside.
In 5 seconds (literally) people open the doors, push the cart out.
Call the fire department.
Battery starts popping loudly.
Small cylindrical cells blow their top (so much for the safety vent being a good solution) like projectiles, way in the air, and onto the roofs of the nearby buildings (the Quizno's in this picture).

 

·
Registered
Joined
·
3,141 Posts
Have there been any incidents where the BMS was working properly and the cell still failed catastrophically? And what would it take to contain such an explosion and fire and prevent other cells from joining in the fireworks?

I'm considering a battery pack consisting of probably 100 Li-Ion or LiFePO4 cylindrical cells (18650) in series for 320V or so, and I would like to be prepared for a destructive battery event, especially since I may use the cheap "ultra-fire" type cells if they test close to specs. I am considering enclosing each string (perhaps 10-16 in series), in a clear polycarbonate tube, which is pretty strong but only good to 200F. So I may also wrap them in a high-temperature fabric such as Nomex or Fiberglass:

http://www.mcmaster.com/#fabric/=l4jocn (Vermiculite-coated fiberglass, 1500F, 0.020 thick, about $2/sqft)
http://www.mcmaster.com/#catalog/119/3555/=l4k4zc (Lexan tube, 3/4ID, 7/8 OD, $2.71/ft)

Seeing the example where the cells blew their tops, I would mostly need to contain them in that direction, along the axis of the stack. I think I could provide end plates for the pack assembly made of steel or aluminum and perhaps some strong neoprene foam to exert some pressure on the stack to maintain connection and allow some expansion while absorbing some of the energy. The entire pack would be about 30" long for 10 18650 cells, and about 6" x 2" for a 5x2 matrix of 7/8" diameter tubes. This would be, "potentially", a 3.2Ah 370V pack or 1.2 kWh, and would cost about $240 for the Li-Ion cells. The enclosure and the BMS will probably cost twice that. I'm willing to use Li-Ion if I can be sure that the BMS and the enclosure will make it safe. I am not as concerned with reliability as it will be used in a lawn/garden tractor.
 

·
Registered
Joined
·
838 Posts
Jack (EVTV) and his partner in crime that does the extreme battery testing is a good start. It is nice to know that shorting out one cell won't cause it to spontaneously combust, but what if you have 45 of them wired together...

Also, what if they are in an tightly enclosed container on a hot day, what temperature range will cause problems?

If it works fine at 60 C, what is the failure temperature and what will happen? I would like to know what the limits are to know where I can't exceed.

I think that LiFEPO4 batteries are a good technology, but there is a bunch of engineering tests that are missing from the public record. The battery manufacturers or suppliers aren't doing that great of a job coming out with guidelines for how to use the batteries, what not to do, and how to wire them up.
 

·
Registered
Joined
·
447 Posts
shorting out one cell won't cause it to spontaneously combust, but what if you have 45 of them wired together.
Exactly the same result.
Think about it this way:

  • A single cell sources x Amp through a short circuit
  • Two cells in parallel source 2 * x Amp through a short circuit, but each cell sees 1/2 of it, so, x Amp
  • 45 cells in parallel source 45 * x Amp through a short circuit, but each cell sees 1/45 of it, so, x Amp
Sure, the point of short circuit sees 45 * x Amp; but the voltage across it is 0 V, so the power is also 0 W. All the heat is in the cells themselves, and it's evenly distributed.


The bigger the cell, the more short circuit current it can source, but its volume is also bigger, so it can handle more current. It makes little difference whether it's a cell that is 45 times as big, or 45 small cells in parallel.


(Yes, on a second order, there are thermal differences, but a short circuit discharges a cell very quickly - 20 seconds to 4 minutes, almost faster than cooling effects.)
 

·
Registered
Joined
·
4,005 Posts
Jack (EVTV) and his partner in crime that does the extreme battery testing is a good start. It is nice to know that shorting out one cell won't cause it to spontaneously combust, but what if you have 45 of them wired together...
When did Jack hard short a cell? I'd like to watch that one. Someone should compile a list of high voltage/current events for extreme viewing pleasure.


Exactly the same result.
Not necessarily. A single cell that's shorted is making lots of heat, but in a full box of cells that are shorted will be subject to much more heat. The one in the middle will get 5x as much.

As for the series string voltage, it's not 0W being dissipated and it's not all internal, otherwise no wires would ever melt. Also, there is a big difference between a shorted condition and the act of shorting. It begins as a high voltage/current event and is likely accompanied by plasma or other destruction.
 

·
Registered
Joined
·
447 Posts
You are correct in all you say.
But that was not the question I was answering. The question that was asked was about shorting a cell. A short is a short is a short, and has 0 Ohm, and therefore 0 W.
In reality, what you describe is more likely that a 0 Ohm short. So, yes, I very much appreciate all you say there.
 

·
Registered
Joined
·
937 Posts
Power = i^2 * R = i^2 * 0 = 0 W
I think that I is irrelevant at zero ohm. Or olso infinitive.
On the other hand we know the exact value of U.
Which leaves us with the more useful P=U^2/R. With R->0

Luckily we mortals never have to work with zero ohm, so lets take a more realistic 0.001 ohm. P=3.2^2/0.001 = 28kW. Neglegting sag.
 

·
Registered
Joined
·
3,141 Posts
It's true that a short is never 0 ohms. There is always some resistance (except at cryogenic temperatures) and inductance (which will limit how quickly the current can change). As the temperature of the conductors increases, the resistance will also increase, lowering the current and increasing power. It is rare for an externally applied short to be very low resistance, since the point of application will get very hot and possibly arc and become intermittent. The battery has internal impedance which can be estimated by the amount of external resistance required for its terminal voltage to be half of normal, so a 100 Ah battery might be rated for as high as 50C or 5000 amps, at which point its voltage would be about 1.8V so its internal resistance would be 360 uOhms, and the power dissipation would be 9000 watts in the battery and the load. The maximum power with a true short might be 10kA at 3.6V or 36 kW. The energy of the battery is about 360 Wh, so at 36 kW it would be depleted in 0.01 hours or 36 seconds. Plenty of time to do lots of damage!
 

·
Registered
Joined
·
4,064 Posts
a 100 Ah battery might be rated for as high as 50C or 5000 amps, at which point its voltage would be about 1.8V so its internal resistance would be 360 uOhms, and the power dissipation would be 9000 watts in the battery and the load. The maximum power with a true short might be 10kA at 3.6V or 36 kW. The energy of the battery is about 360 Wh, so at 36 kW it would be depleted in 0.01 hours or 36 seconds. Plenty of time to do lots of damage!
That power is 9000W for the battery, 9000W for the load.... for those reading.

Battery power dissipation due to internal resistance:
5000A and 0.00028ohm would be P = (5000^2)*0.00036 or 9000W

Power to the load:
1.8V * 5000A = 9000W
 

·
Registered
Joined
·
3,141 Posts
This is at the point of maximum power transfer. At higher or lower external resistance, you can get higher voltage or higher current, but the power will be less. This is a principle of impedance matching. ;)
 
1 - 20 of 29 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top