Prius 2001 battery fire

Here's a link to the article if you want to read it:

Is Toyota's make-good offer fair? By Ralph Vartabedian, LA Times Staff Writer

 

hello, i came across this forum when i started researching why in the world my perfectly good 2001 Toyota Prius spontaneously caught fire.
The Prius had been parked in my driveway for 48hrs. At about midday my neighbor came to the door and told me my car was full of smoke.
I went out and heard crackling coming from the rear end. I opened the trunk and the back of the seat and all the carpet was flaming and melting.
We got the fire out before it progressed too far and the firefighters said the fire started where heat had melted a hole in the far right corner of the protective metal casing for the Prius Hybrid battery pack.
I should also say I'm in Australia. I phoned the dealership to tell them about the problem and they asked if they could come take photos.
Naturally i agreed and waited to hear back from them. They did not call, so I called and asked for their assessment they told me we would have to arrange for a tow at my expense and then they would get a very busy specialist to investigate the car in a few weeks or so.
The car was not fully insured. I did not receive any letters of any type from Toyota to bring my car in for service or recall.
I have attached photos on carspace in a similar forum.
Toyota Australia told me this was the first time this has ever happened. I informed them that my research identified other similar fires in the Prius.
Does anyone know who i can talk to for help or if Toyota is doing anything to rectify this situation and help their customers who have had their Prius spontaneously combust?

 

Are these your photos?

Carspace.com - suzikn: Prius Battery Fire

The first photo is a classic example of a battery fire but it located in the battery control electronics area. The plate just below the hole appears to be the cover plate that the primary power leads, normally orange but apparently smoke and melt covered, come out.

Normally, the battery and control electronics are isolated from the case. So this is pure speculation but given the well melted, external vent on the right, I'm wondering if an insect managed to get in and complete a circuit. A centipede or millipede would be my guess although a small snake could also complete the circuit.

Speculation on my part but I suspect the short happened in the 19 modules on the control electronics side. The safety interlock is located there on the trunk side so half of the traction battery, quite enough, could have shorted on that side. Once the arc started, nothing will stop it.

You have my sympathy.

Bob Wilson

 

Of course, now that times are getting tougher, I wouldn't be surprised if some Prius owners look to get out of car payments by having a case of "spontaneous combustion" happen to their Prius

I should warn the lurkers that in the US and Canada, the local fire marshall is trained to look for evidence of accelerants, and tests can be performed on the remains to determine if accelerants were used to start the fire

There was a case of a Prius being burned, but the firebug in question was targeting a lot of vehicles, not just hybrids.

 

That is true, but the fire marshal may not be called to investigate. If the fire suppression crew doesn't suspect anything suspicious, they may not alert the fire marshal's office. Problem is, firefighters usually are not as well trained as fire investigators in determining origin and cause, so they often will make their best educated guess as to cause and leave it at that.

 

I could never hurt any of my cars


How often can we get a 40G my last one was in 05, can i request it?

 

I see even see many non hybrid cars catching fire. And no, it was not arson. Many times electricals are the culprit. Even the 12 V battery can produce enough energy for combustion, hybrid battery with much higher voltages and amperages, of course even more. Most probably fire did not start in cells themselves, but in some surrounding electronics going belly up in short circuit type of way.

 

SSC 40G is applicable to 2001-2003 Prius only. I do not believe that the intention was to perform this SSC more than once per affected vehicle.

Sorry to hear about jayma8's fire. You'll have to work with Toyota Australia to see whether financial help will be made available; however I suspect that not much assistance will be offered.

 

Bob, Jaymans fire is at:

Carspace.com - jayma8: prius_001.jpg

Discussion at:

http://townhall-talk.edmunds.com/direct/view/.f12cc22

get out of car payments by having a case of "spontaneous combustion" happen to their Prius I should warn the lurkers that in the US and Canada, the local fire marshall is trained to look for evidence of accelerants, and tests can be performed on the remains to determine if accelerants were used to start the fire

Jayman, note small holes burned into the metal. Torch maybe, but no one tossed around gasoline.

 

I'm very sorry for your loss :-(

I do hope Toyota will investigate and determine the cause if possible. If nothing else, it always helps an engineer to understand the eventual failure mechanisms of his design. All mechanical and electrical systems eventually fail. We do our best to make sure that all possible due care is taken to make them fail as controllably and infrequently as possible but you can not avoid the fact that things will eventually fail. It does not seem to me there is any evidence so far that there is any sort of systemic failure here, something just wore out, or broke loose, or as suggested above was infiltrated by foreign matter. It certainly demonstrates the respect you need to have for carrying around an arc welder in your trunk. I think we should not lose site of the fact that this potential danger pales in comparison to the gasoline bomb riding in the trunk with it. A quick google search indicates that there are 290,000 car fires a year. Unfortunately given the nature of the materials and systems at work in a vehicle this is always a possible failure mode. That this is the first failure of this nature that I am aware of (if it even is a failure and not ingress of some conductive object) I personally find reassuring. The Q/A engineers I know hate failures, but the only thing they hate more is the lack of failures. If something never breaks, you have no idea when or how it will fail. Now have some idea. Slow, smoldering, smoky fire in the rear. Scary for sure, but doesn't seem like the kind of explosive, rapid event that is likely to take a life. And as far as when, one would have to say so far very infrequently.

It will be very interesting to see what more we find out from this event.

Rob

 

The most important thing is that we know it's the batteries ... and NOT the 10 gallons of toxic liquid high explosives that burn people to death every day ... no ... lets concentrate on the handfull of battery fires. That's what blows me away.

 

Hey Bob,

It sounds as if you're seeing photo 1 as being a view of the electronics end (driver/port side) of the battery, a view that would have to be from the cabin looking aft into the trunk through what's left of the rear seatback. I'm not sure I'm seeing it that way.

I think the electronics/port end is shown in photo 3. The cooling exit duct attaches there to the top of the case. Photo 1 shows the end of the case where the top has no opening and the cooling entrance duct attaches low on the case end. I think that's the starboard end as seen from the trunk looking forward, and what's visible in the hole burned through the steel is the remains of module 38.

I would speculate that combustion gases flowed inside the case chimney-style until they were able to melt through the plastic exit duct that attaches on top, causing the preponderance of damage on the port side of the car.

I'm not sure it matters what side of the service plug/fuse the fault occurred on; either way you slice it, two faults are needed to complete any subcircuit that doesn't include the fuse. You can get a 14.4 volt short by bridging any two adjacent terminals that aren't bussed, and with the energy capacity of the modules that might be enough. The odd thing is, I think the aft terminals of 37 and 38 are bussed, so the nearest place a two-adjacent-terminal short could have happened would be 36 to 37, which doesn't look like where the hole is.

It might also be possible that the fuse was part of the fault circuit but didn't blow. Residential codes are now requiring arc-fault interrupters on some circuits for just that reason - it's possible for an arc to make enough heat for ignition while never drawing enough raw current to trip a pure overcurrent safety. The arc-fault gadgets watch for characteristic waveforms seen when arcing occurs.

It would be interesting to know the condition of the service plug and fuse after the event. I can see in photo 3 that the plug has been removed.

If a short to the case was involved, it still seems that two ground faults would be needed, since all the high voltage is isolated from the chassis by design. The second fault to occur would be the time of badness, but the first one should have been detected and a code stored. Unless it was some event like a critter that bridged in two places at once.

It's a very strange puzzle, and my sympathies to the owner too.

-Chap

 

Nope, wasn't *my* fire

I was making a general comment, not a specific comment to this incident.

As far as accelerants, things like turps can be used. Gasoline is actually a poor accelerant due to it's characteristic odor, dangerous volatility, etc

 

It blows me away too. We've already seen the YouTube video of the young woman with that delicious tight wool sweater, spark a fire at the gas station due to ... um ... friction

 

Wow! Two battery fires.

I've got to run some errands and have some thoughts to share when I get back.

Bob Wilson

 

I've been thinking more about those pictures, and something doesn't add up to me. If as stated the location of the burn through hole is over the HV electronics, I can see that. Live power has to come up as far as the positive side contactor, so thats feasible. Whats not sitting right in my mind is that the picture looks pretty much exactly like what you'd expect if the battery HV was arcing to the case. Problem is, isn't the case electrically isolated? My understanding was that all of the HV electrical components are isolated from chasis ground. As I recall there are contactors and current sensors on both the positive and negative battery lines. Now we are into that uncomfortable territory of needing two things to fail at exactly the same time in order to cause the fault. My understanding is that if either thing had failed independently at a previous time, it would have triggered a hybrid system fault and disabled the system, which the owner would have noticed.

I would think the most likely scenario for both the positive and negative side of the battery failing is if it were in a car that was, say, on fire?

One more question. Can the HV wires actually carry enough current to sustain a plasma fire? I've no doubt the bus bars could, but how much wire is there between the bars and the location where the event apparently occured? I was under the impression that it took 100s if not around a thousand amps to really get a plasma discharge event going. To handle those kind of currents intentionally, the 144V EVs I've seen often use 0000 gauge wire (about 3/4 of an inch in diameter not counting insulation). I've always though the HV wiring in the Prius looked a bit piddly, which I had assumed was intentional. "Right sizing" wire is often done as a safety precaution. You want enough capacity for it to safely carry the design current, but under a high current event such as a short you'd ideally like the wire to burn out and break the circuit like a fuse. For that matter can the 6.5Ah batteries actually put out enough current to sustain a plasma fire? This is a real concern when you have a 100-200Ah lead battery designed to put out 1000A continuously, particularly when you stack 12 of them in serious to build up 144V. The Prius pack certainly has the voltage, but I would think the internal resistance of the small NimH cells would be naturally current limiting in a short. Burning through the metal case would take pretty substantial amount of current for a fairly sustained period of time. The one event I've heard about where a Prius battery caught fire was a PHEV experiment gone wrong. The internal battery was accidentally connected to the external battery while it was being charged overnight. The cars main 12V rail was powered up externally, and various safety systems had been disabled for experimental purposes. In PHEV conversions the external pack/charger voltage is higher than the internal pack voltage, so the charger just kept forcing charge into the internal battery all night long. After a number of hours of the battery modules started failing and exploded due to extreme internal pressure/heat caused by overcharging. This did cause a small fire, in part because all of the protective coverings had been removed and everything was exposed.

Its also hard to figure how the battery sealing service would have anything to do with an event like this. My understanding is not sealing the batteries properly allows for a bit of electrolyte to leak out over time and create a conductive path from one of the terminals to the other or more often to the case. While this path is far too resistive to allow a real short to occur, it can allow a few milliamps (thousandths of an amp) of current to flow. Because the Priuses current sensors and safety systems are so sensitive, this tiny amount of current flow is enough to trip the HV systems "ground fault" protection and cause a hybrid system fault disabling the whole the system.

I don't have enough details of the event or really the specific technical knowledge (particularly of the Gen1) to call BS, but as I say, something about it just isn't sitting right with me as far as the cause being the battery. I guess I'm skeptical in part because I've seen an HV battery out of a Prius that was under water in Huricane Katrina. Even under those circumstances there was no evidence of a plasma event.

Rob

 

Consider this: it takes 11,400V to arc across 3mm or a little over an inch in dry air. How far do you think a 250V battery will arc? About 0.05mm if at all. I seriously doubt corona was an issue here. Likely something dropped across two points and caused the arc or like someone above stated - a critter may have crawled in.

 

All it takes to make a safe battery a high termperature source is a single piece of loose metal. A left over screwdriver, a misplaced knife, or a loose piece of wire could be one of the factors involved with this.

Given the large number of items left inside patients after surgury, it might be possible something was left inside the battery cover.