Post-mortem report on significant Li-ion incident

aznsx

Flashlight Enthusiast
Joined
Apr 24, 2015
Messages
1,631
Location
Phoenix, AZ USA
I came across this very detailed / exhaustive report that was recently released following the analysis of a rather significant Li-ion battery system event which occurred at a power utility storage installation here in AZ last year. I didn't read the entire report verbatim, but scanned through it and found some sections of information which I found very interesting. While perhaps 'too much information' for some, some might like to scan through it for sections which might be of interest.....or not:). Suffice it to say that when this stuff goes wrong, it can go wrong in a big way (which in itself is of course is not 'news' to most of us). The report also includes some references to / information regarding several other significant 'incidents' which made the news in the past.

https://www.aps.com/-/media/APS/APS...x?la=en&hash=50335FB5098D9858BFD276C40FA54FCE

- Richard
 

turbodog

Flashaholic
Joined
Jun 23, 2003
Messages
6,425
Location
central time
Thorough report. Did the firemen get hurt? If it said this I missed it.

Nothing new in the report. Dendrites, internal short, thermal runaway, cascading to adjacent cells, flammable gas buildup, explosion.

Someone familiar with enough cross-discipline info could have prevented this during design phase. Failures all across the board... including the fire crew that opened the door.
 

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
It's incidents like this that have stopped me from DIY'ing large powerbanks or powerwalls - more or less none of the builds I've seen seem to account for these contingencies outside of attempting to contain a fire ala siting them separate from a residence or using a metal container for the cells.
 

Katherine Alicia

Enlightened
Joined
May 15, 2020
Messages
836
Location
Central UK.
I`v ofen thought about this if I were to design another large Li Ion based power bank again, and I think I`d probably have them spring loaded as one of the connections and soldered using Wood`s metal to the bus bar for the other rail, so if any one cell ever overheated (above 70c) it would automatically be ejected from the bank by the spring because the wood`s metal holding it in place would have melted.
There are Other alloys that can be used as well for a variety of different temperatures.
but that`s as far as I got in my design, I`m sure smarter folks than me could make it work though.
 

KITROBASKIN

Flashlight Enthusiast
Joined
Mar 28, 2013
Messages
5,390
Location
New Mexico, USA
From the report:

At approximately 16:54 on April 19, 2019, a battery cell in Module 2 of Rack 15 of the BESS experienced asudden drop in voltage during a charge cycle, equivalent to a single cell voltage drop (4.06 to 3.82 V asshown in Figure 14). Module 2 was in the second module position, numbered upward from the base of therack.Moments after the voltage drop occurred, cell 7 in module 2 of rack 15 went into thermal runaway. Thisevent generated off-gassing and smoke that activated the VESDA smoke detection system and led to thedischarge of the Novec 1230 agent. The initiating cell's thermal runaway cascaded into thermal runaway ofneighboring cells within module 2 and subsequently into the batteries contained within neighboring modules.The off-gassing of battery cells from the cascading thermal runaway created a flammable atmosphere withinthe BESS.Approximately 3 hours after the initial voltage drop and discharge of the fire suppressant, firefightersopened the side container door, and approximately 2-3 minutes after the door was opened, an explosionoccurred. The doors on the side and rear of the BESS and other debris were ejected by the explosion asshown in the figures below. At approximately 16:54 on April 19, 2019, a battery cell in Module 2 of Rack 15 of the BESS experienced asudden drop in voltage during a charge cycle, equivalent to a single cell voltage drop (4.06 to 3.82 V asshown in Figure 14). Module 2 was in the second module position, numbered upward from the base of therack.Moments after the voltage drop occurred, cell 7 in module 2 of rack 15 went into thermal runaway. Thisevent generated off-gassing and smoke that activated the VESDA smoke detection system and led to thedischarge of the Novec 1230 agent. The initiating cell's thermal runaway cascaded into thermal runaway ofneighboring cells within module 2 and subsequently into the batteries contained within neighboring modules.The off-gassing of battery cells from the cascading thermal runaway created a flammable atmosphere withinthe BESS.Approximately 3 hours after the initial voltage drop and discharge of the fire suppressant, firefightersopened the side container door, and approximately 2-3 minutes after the door was opened, an explosionoccurred. The doors on the side and rear of the BESS and other debris were ejected by the explosion asshown in the figures below.
 

turbodog

Flashaholic
Joined
Jun 23, 2003
Messages
6,425
Location
central time
It's incidents like this that have stopped me from DIY'ing large powerbanks or powerwalls - more or less none of the builds I've seen seem to account for these contingencies outside of attempting to contain a fire ala siting them separate from a residence or using a metal container for the cells.

You could put the pack in a standard jobsite toolbox, clamp the lid thoroughly, and put some vents in to allow gases to escape.

But yeah... still would not sleep well. I am paranoid with the lipo packs for r/c that I have.
 

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
You could put the pack in a standard jobsite toolbox, clamp the lid thoroughly, and put some vents in to allow gases to escape.

But yeah... still would not sleep well. I am paranoid with the lipo packs for r/c that I have.

For a DIY powerwall, I've considered just burying the thing. If the cells can sink their heat to the earth and the enclosure has some means of venting to atmosphere, seems like any destructive fires could be safely contained under 12-18" of soil so long as it's otherwise well-sealed.
 
Last edited:

Eicca

Newly Enlightened
Joined
Aug 3, 2019
Messages
108
I'd say I have a healthy fear of lithium ion batteries. Ain't no way I'm buying an electric car until they figure out an alternative, like those supposed holy grail quantum glass cells. I barely tolerate laptop-size cells as it is.

Don't even get me started on internet order E-bicycles. People are buying time bombs and storing them either inside inhabited homes or next to cars and flammable garage items, never mind how awful they are to deal with as a bike mechanic.
 

mr45acp2u

Newly Enlightened
Joined
Oct 14, 2007
Messages
34
Yes, numerous firemen were hurt. They are alive because of the turnout gear they were wearing and the fact that the fence around the facility was chain link. It allowed the force of the explosion to dissipate over a greater area, vs a bock wall. And they were blown under the fence, vs being pinned against a block wall. Most of them have recovered from their injuries, one is still recovering and not back to full duty. I'm not with the fire department, but work closely with them.
 

Entryteam

Newly Enlightened
Joined
Sep 5, 2007
Messages
2
My father was a firefighter for 39 years. Glad all these fellas' turnouts saved them.

Good to hear.
 

turbodog

Flashaholic
Joined
Jun 23, 2003
Messages
6,425
Location
central time
I`v ofen thought about this if I were to design another large Li Ion based power bank again, and I think I`d probably have them spring loaded as one of the connections and soldered using Wood`s metal to the bus bar for the other rail, so if any one cell ever overheated (above 70c) it would automatically be ejected from the bank by the spring because the wood`s metal holding it in place would have melted.
There are Other alloys that can be used as well for a variety of different temperatures.
but that`s as far as I got in my design, I`m sure smarter folks than me could make it work though.

FYI, for those not in the loop, "wood's metal" is an alloy of mainly bismuth... melts at ~150F.

However, to note, some of these packs swell up before they blow... thereby jamming in the rack/container/etc.
 
Top