Lithium Fire at EVTV and Boeing 787 Dreamliner

[mdbuilder]

It'll burn

 

[kiev]

Boeing required Yuasa to perform the nail puncture test as part of battery qualification --i don't recall if there was a video shown during the NTSB investigation of the 787 incidents, but there were some photos in one of the reports.

Securaplane in Arizona built the Battery Charging Unit for Boeing. In 2006 their lab facility burned to the ground during the first time that they tested the BCU with a real battery. It appears that qualification and acceptance testing from then on used a battery simulator instead of real batteries.

Boeing repeatedly stated that the only way these batteries could catch on fire was due to overcharging, and then presented low-resolution graphs from the Flight Data Recorder to prove that the batteries were not overcharged => therefore it must have been something else.

Looking at the CT scans from both investigations you can see puffed out cell cases in the "undamaged" battery packs. Puffy cells are an indication of overcharging in Li batteries.

Looking at the Secura patent for battery charging you will notice two things of interest. The method was based upon a Ni-Cd battery charging profile, and the procedure assumes a trickle-charging phase of indefinite end following the fast charge phase.

Looking at the FDR data from the Boston event you can see that the battery charger was indeed holding a constant 32 volts and trickle-charging the APU battery at 1-2 amps for about 17 minutes after the APU start until the fire started.

Looking at the charging process for Lithium batteries from GS Yuasa and every other Li-battery manufacturer in the world and you will find that there is no trickle-charging. Instead you will find a specific charging procedure that consists of a constant current (at ~1C) charging phase up until the cell voltage hits a specified cut-off value, then followed by a constant voltage phase in which the cell is held at this specified voltage until the current drops to C/20, at which point the cell is fully charged, charging is complete, turn off and disconnect the charger.

Looking at the data sheet for the LVP-65 cells from GS Yuasa used in the APU and Main battery packs you will find that they have a nominal capacity rating (1C) of 65 Amp-hrs, nominal voltage of 3.73, maximum charging rate of 1C (=65) amps, and the cell characteristic curves use 4.025 volts as the CC/CV specified charging cut-off voltage. The C/20 cut-off charging current is 65/20 = 3.25 amps. There is no procedure called out nor should there be any additional charging below 3.25 amps after reaching the 4.025 volt level.

Attempting to trickle-charge a Lithium battery, especially one that is already fully charged, is a sure-fire way to cause thermal runaway...

Now if you want to look for battery debris, check out the battery box construction near cell #4 that vented and burned a hole thru the case--right above the area of the fire hole is a pop rivet. A hole had to be drilled for the rivet to be inserted. It would be easy for a small piece of drill swarf or rivet material to fall down between the cell and the case, and create a short to ground.

p.s. An to think nearly 200,000 battery cells in mievs around the world are charged every day and night and not one fire due to overcharging! So who know more about batteries Boeing or Yuasa?

 

[PV1]

Sounds like Securaplane had faulty equipment.

Securaplane in Arizona built the Battery Charging Unit for Boeing. In 2006 their lab facility burned to the ground during the first time that they tested the BCU with a real battery. It appears that qualification and acceptance testing from then on used a battery simulator instead of real batteries.

Sounds like that cures the symptom, not the problem. What good is a charger if it destroys actual batteries? Granted, 2006 was still pretty early for large format lithium ion, but even then it was known not to trickle charge them.

There were issues with Yuasa, too. Some early Outlander PHEV and a couple of JDM i-MiEVs suffered battery problems due to contaminated cells around the same time.

 

[Benjamin Nead]

Ah, yes . . . Securaplane. I'd forgotten all about them. This place was located in Marana, on the northwest side of the larger Tucson metro area. When we found out that the 787 fire incident traced itself back to stuff made in our own backyard, local battery power and EV enthusiasts sheepishly hung our heads low for a while.

Poorly implemented charging and balancing/management electronics can make even the best of batteries light off like Roman candles. There's a lot more to lithium battery system than just the cells themselves and this sort of thing might be behind what happened to all those cheap hoverboard toys that caught on fire this past year.

 

[JoeS]

...There were issues with Yuasa, too. Some early Outlander PHEV and a couple of JDM i-MiEVs suffered battery problems due to contaminated cells around the same time.

That was actually in early 2013, with a process issue identified and quickly solved. Once again, nothing to do with the Dreamliner.

http://www.autoblog.com/2013/04/25/mitsubishi-outlander-plug-in-battery-fire-cause/

 

[PV1]

There's a lot more to lithium battery system than just the cells themselves and this sort of thing might be behind what happened to all those cheap hoverboard toys that caught on fire this past year.

That's exactly what happened. A few were caused by faulty cells and faulty chargers, but ultimately many of them lack a BMS completely. Plus, I don't believe the packs were balanced before assembly in the boards that caught fire.

I think it goes back to people adopting NiMH or NiCD care tactics to lithium ion, hence why so many people think lithium ion batteries have memory (when in reality, that was only NiCD batteries due to state change of the cathode/anode). NiMH packs don't require a BMS as long as the charger has the correct charging profile because they are tolerant of a little overcharge and can self-balance to some degree. Lithium ion does not.

Even with the media scare and recalls, there have been 99 reported hover board fires. 501,000 have been recalled, and who knows how many millions have been sold. There is still more of a chance of an ICE car catching on fire and not because of a collision (roughly 1 every 7 minutes).

Thanks for the Yuasa source, Joe.

 

[Phximiev]

Another lithium nightmare before some of the airline safety modifications (shocking):



https://youtu.be/pbL_rgeJAZY

 

[Phximiev]

Safer batteries in the future, more than likely:

https://chargedevs.com/newswire/nohms-raises-5-million-to-commercialize-non-flammable-liquid-based-electrolytes/

 

[Phximiev]

Anyone predicting exploding BMW i3 or i8s?

http://www.nytimes.com/2016/11/05/technology/samsung-recalls-washing-machines-galaxy-note-failure.html?smid=nytcore-ipad-share&smprod=nytcore-ipad

 

[Phximiev]

Grim, exploding batteries?

http://tech.firstpost.com/auto/tesla-model-s-crashes-and-explodes-in-giant-fireball-two-fatalities-reported-345377.html

 

[PV1]

I'd say those batteries were justified in their explosion given the damage to the car. For the tree to go far enough into the front of the car to split the main pack apart and scatter modules all over the road, they were going entirely too fast for any public road.

Ignoring it's a Tesla, it's a wreck where a car hit a tree at a high rate of speed, caught on fire, and the people in the car got killed. Hate to say it, but this happens every day. The exploding batteries were a result of physical damage, the same as almost every other EV battery fire that has occurred.

 

[Phximiev]

Lithium battery fire in Phoenix:

http://www.azcentral.com/story/news/local/phoenix-breaking/2017/05/06/north-phoenix-fire-sparked-lithium-batteries/312091001/

 

[JoeS]

Lithium battery fire in Phoenix:
http://www.azcentral.com/story/news...nix-fire-sparked-lithium-batteries/312091001/

Ah, yes, makers of the stealth electric bike.

https://www.treehugger.com/cars/gru...-bike-no-one-will-know-but-your-mechanic.html

Renamed company:

http://2014.vivax-assist.com/en/unternehmen/index.php


Interesting about the issue of dousing water onto a lithium battery fire. I knew about water and Sodium but not Lithium.

Interesting that the dismantling guide for the i-MiEV recommends totally submerging the battery pack underwater for 72 hours, emphasizing fresh water and not salt water.
http://elvsolutions.org/wp-content/uploads/2013/06/2012_I_Dismantling_guide.pdf

 

[PV1]

Wouldn't have anything to do with the record heat, would it? Article dated the same as the lithium fire.

http://www.azcentral.com/story/news/local/arizona-weather/2017/05/05/record-heat-phoenix-107-degrees/311941001/

 

[Phximiev]

Wouldn't have anything to do with the record heat, would it? Article dated the same as the lithium fire.

http://www.azcentral.com/story/news/local/arizona-weather/2017/05/05/record-heat-phoenix-107-degrees/311941001/

Could be, if the AC broke on the same date, which can and has happened here before. There will probably be some kind of news follow-up as to the cause tho.

 

[jray3]

Interesting about the issue of dousing water onto a lithium battery fire. I knew about water and Sodium but not Lithium.

I think it's either a misquote or a misunderstanding on the fireman's part. IIRC, lithium battery fires with the cobalt (and possibly magnesium) chemistries may not be put out by water alone because the cells contain their own oxygen supply, but copious amounts of water are actually recommended to cool things down below the ignition point and also to keep the fire from spreading.

Edit: Good ol' Battery University: http://batteryuniversity.com/learn/article/safety_concerns_with_li_ion
They say that only "Lithium Metal" batteries react with water, not Li-ion, LiPO or the other more common chemistries.