Jump to content
eddiemoy

KS-16S caused a fire...

Recommended Posts

There are a bunch of videos on YouTube of the "Elide fire ball" in action.  The problem with LiIon fires, of course, is that while the runaway reaction is still going on (and propagating through the pack), it is almost impossible to put out, but one of these fire ball things might at least suppress the fire and slow down its progression to the surrounding area.  I've ordered one, as well as a paired set of smoke detectors that communicate with each other wirelessly.  It's actually against code to install a smoke detector in a garage (due to nuisance alarms), but I'm going to do it anyway and put its paired sibling somewhere upstairs.  That way when the one in the garage goes off, I should get an alarm in the living area.  So my strategy will consist of:

  • Free-standing as far from shelves/walls as possible, in a garage with a cement floor
  • One sprinkler head nearby (although not directly overhead)
  • A "fire ball" suspended above
  • A smoke detector nearby, with a wirelessly paired sibling in the living area to warn of an "event"
  • A normal CO2 fire extinguisher is already hanging by the garage entrance from the living area

I think that's about all I can do reasonably, and some might say this is already overkill.

Share this post


Link to post
Share on other sites
Posted (edited)
1 hour ago, svenomous said:

There are a bunch of videos on YouTube of the "Elide fire ball" in action.  The problem with LiIon fires, of course, is that while the runaway reaction is still going on (and propagating through the pack), it is almost impossible to put out, but one of these fire ball things might at least suppress the fire and slow down its progression to the surrounding area.  I've ordered one, as well as a paired set of smoke detectors that communicate with each other wirelessly.  It's actually against code to install a smoke detector in a garage (due to nuisance alarms), but I'm going to do it anyway and put its paired sibling somewhere upstairs.  That way when the one in the garage goes off, I should get an alarm in the living area.  So my strategy will consist of:

  • Free-standing as far from shelves/walls as possible, in a garage with a cement floor
  • One sprinkler head nearby (although not directly overhead)
  • A "fire ball" suspended above
  • A smoke detector nearby, with a wirelessly paired sibling in the living area to warn of an "event"
  • A normal CO2 fire extinguisher is already hanging by the garage entrance from the living area

I think that's about all I can do reasonably, and some might say this is already overkill.

Bold highlight 1: It's important to know that the best way to halt thermal run away, is to quence the heat immediately with almost anything cold like a liquid or cooling fire extinguisher, although even a hot cup of coffee,(as a temporary measure until colder liquids  can be sourced) is colder than thermal runaway temp. Shorting out the batteries with a liquid is absolutely not your primary concern.  Thermal runaway must be stopped at all cost.  Here's an official Federal Aviation Administration (US) information and training video for in-flight personnel.  Obviously if your EUC has gone BANG and there are bits of burning battery all over the place (sounds like this was the case in question), you just have some decisions to make.  One personal warning: try to maintain as much distance as possible if attempting to quench one of these fires.  If it goes BANG again when you are close you risk injury or death.

Bold highlight 2: There's no need to put a smoke detector in your garage when a perfectly acceptable heat detector is available.  In the UK the current building code prohibits smoke detectors in Kitchens for the reason you state, but mandates a heat detector instead.

https://www.amazon.co.uk/SEBSON-Wireless-replaceable-included-interlinked/dp/B07G3DH6WQ/ref=sr_1_1_sspa?keywords=heat+detector&qid=1557778455&s=gateway&sr=8-1-spons&psc=1

 

Edited by Smoother
  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites

Interesting.  Heat alarms are not common in the U.S., and on the U.S. Amazon site I can't find the one you linked to, @Smoother.  Instead I find just a couple of hardwired options (not battery-driven).  Will do some more research to see if I can get two of these Sebson brand ones (to interlink with each other) through another retailer in the U.S.

Share this post


Link to post
Share on other sites
Posted (edited)
1 hour ago, svenomous said:

Interesting.  Heat alarms are not common in the U.S., and on the U.S. Amazon site I can't find the one you linked to, @Smoother.  Instead I find just a couple of hardwired options (not battery-driven).  Will do some more research to see if I can get two of these Sebson brand ones (to interlink with each other) through another retailer in the U.S.

Oh, that's weird.  edit.  Hard wired's not so bad, no batteries to change.  Curious why no battery only units.  All yours look very expensive.  Ours are cheap by comparison.  That's a first.

 

Second Edit: Best practice with wired detectors in a domestic installation is to wire them to an existing essential lighting circuit.  Why "essential" because this is a circuit that people will not switch off and leave off if a few annoyance alarms are tripped.  Oh they may flip the breaker in the moment, but because of the essential nature of that lighting circuit, it will have to be switch on again.  Put smoke/heat alarms on a dedicated circuit and there is a danger of them being switched off and then forgotten.

Edit 3: The above is/was best practice in the UK when I was an electrician.  Always follow your local electrical code, for safety and staying the right ride of an insurance claim.

Edited by Smoother

Share this post


Link to post
Share on other sites

I've never heard of a heat detector. Interesting concept. 

Share this post


Link to post
Share on other sites
Posted (edited)
47 minutes ago, NylahTay said:

I've never heard of a heat detector. Interesting concept. 

The rational is; in smoke prone rooms like kitchens (burnt toast, etc) nuisance alarms prompt people to silence the alarms by taking the batteries out, or switching of the (dedicated) breaker.  Often they forget to put the batteries back or switch the circuit on, so the protection is gone and no body knows or remembers that they are not protected.   Since a real kitchen fire (not burnt toast, etc) emits a significant amount of heat (more than that emitted by cooking) a heat detector does the same job, but without the risk of being switched off due to nuisance alarms.

Edited by Smoother
  • Upvote 1

Share this post


Link to post
Share on other sites
On 5/12/2019 at 6:58 AM, nickysneids said:

As far as testing this goes, I recommend waiting until we get more information on all of the variables (as much as possible) that led into this incident. You can test lots of different things but it would make the most sense to try and recreate a similar scenario to the one in this incident.

My first issue is how to store the wheel after abusing it, assuming it has not gone up in smoke. Dealing with the consequences is a second issue, so I'm planning to approach this as methodically as possible.

  • Upvote 1

Share this post


Link to post
Share on other sites
9 hours ago, winterwheel said:

My first issue is how to store the wheel after abusing it, assuming it has not gone up in smoke. Dealing with the consequences is a second issue, so I'm planning to approach this as methodically as possible.

Buy a metal box, like the ones people store tools in, and put them in there. Or go down the local metal recycle place and see if they have something appropriate. I once wanted some steel targets to shoot ( I like the clang that means I didn't miss) so I went to a steel supplier and had them cut me various plates out of 1/2" steel.  I welded some legs on et voila, my very own steel shooting gallery. Inexpensive too. bang, clang, bang, clang,  music to my ears.:D

  • Like 2

Share this post


Link to post
Share on other sites
11 hours ago, Smoother said:

Buy a metal box, like the ones people store tools in, and put them in there. Or go down the local metal recycle place and see if they have something appropriate. I once wanted some steel targets to shoot ( I like the clang that means I didn't miss) so I went to a steel supplier and had them cut me various plates out of 1/2" steel.  I welded some legs on et voila, my very own steel shooting gallery. Inexpensive too. bang, clang, bang, clang,  music to my ears.:D

Keep in mind that steel is actually one of the least heat resistent material around. When I was in school our structure professor show us a photo of an unprotected steel building after a fire; looks like a giant pile of spagetti. if you're going with a steel box, line it with gypsum board (inside) for actual fire rating. or get that 3m wrap I linked before, its expensive but since its 2 hour rated it is equivelent of 2" worth of boards.

Share this post


Link to post
Share on other sites
2 hours ago, Hsiang said:

Keep in mind that steel is actually one of the least heat resistent material around. When I was in school our structure professor show us a photo of an unprotected steel building after a fire; looks like a giant pile of spagetti. if you're going with a steel box, line it with gypsum board (inside) for actual fire rating. or get that 3m wrap I linked before, its expensive but since its 2 hour rated it is equivelent of 2" worth of boards.

That may be true for a flimsy sheet metal industrial building subjected to a sustained fire, but tool strong boxes are thick steel, have no heavy roof pressing down on them, and an EUC fire does not have enough material to burn for too long.  Also you are trying to contain the explosions that happen when the batteries explode, so that the fire does not spread.  A metal box is still perfect for this.

  • Like 2

Share this post


Link to post
Share on other sites
Posted (edited)
3 hours ago, Smoother said:

That may be true for a flimsy sheet metal industrial building subjected to a sustained fire, but tool strong boxes are thick steel, have no heavy roof pressing down on them, and an EUC fire does not have enough material to burn for too long.  Also you are trying to contain the explosions that happen when the batteries explode, so that the fire does not spread.  A metal box is still perfect for this.

I have a huge industrial ‘job site’  metal box at work that is just collecting dust. It would be perfect for storing my wheels. I also have the automatic fire extinguisher coming that will be mounted above the box and will be tied into the central alarm system. My wife wanted the wheels out of the house so now they all reside back in my pole barn. I now need to protect that structure from fire so I should feel comfortable with this new setup once it’s complete.

Edited by Rehab1
  • Like 2
  • Upvote 1

Share this post


Link to post
Share on other sites
6 hours ago, Rehab1 said:

I have a huge industrial ‘job site’  metal box at work that is just collecting dust. It would be perfect for storing my wheels. I also have the automatic fire extinguisher coming that will be mounted above the box and will be tied into the central alarm system. My wife wanted the wheels out of the house so now they all reside back in my pole barn. I now need to protect that structure from fire so I should feel comfortable with this new setup once it’s complete.

By metal you mean sturdy steel or aluminium - aluminium has quite a "low" melting point and could not be enough.

Also putting all wheels together is not optimal - one feeds the possible fire with even more liion cells from the other wheels.

Normal fire extinguisers are useless against liion fires :( you should check the type and the effectiveness.

I know what a barn is, but have no real picture what a pole barn is. Google translated it to "Stangenkinder":D ("childs of the rack/pole childs"). Is it an open structure?

Maybe something made out of bricks/conrete would be better? Different compartments for each wheel separated by one brick/a concrete wall. Depending on the structure and how long and "strong" one wheel can burn maybe a chimney to keep the flames off the roof and (most of) the smoke outside the barn.

This structure could be reused as big barbecue station inbetween. :P

  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites
1 hour ago, Chriull said:

By metal you mean sturdy steel or aluminium - aluminium has quite a "low" melting point and could not be enough.

The box is heavy gauge steel. 

 

1 hour ago, Chriull said:

Also putting all wheels together is not optimal - one feeds the possible fire with even more liion cells from the other wheels.

Yes I agree . There are fireproof panels available  I may use as a separation.  I am not terribly worried as I don’t charge my wheels unattended. 

 

1 hour ago, Chriull said:

Normal fire extinguisers are useless against liion fires :( you should check the type and the effectiveness.

In the event something did happen the dry chemical extinguisher along with the box’s heavy lid would help contain the fire until assistance arrived. Our alarm system automatically calls the police and fire departments. 

1 hour ago, Chriull said:

I know what a barn is, but have no real picture what a pole barn is. Google translated it to "Stangenkinder":D ("childs of the rack/pole childs"). Is it an open structure?

40x50 foot closed structure. I’ll take a photo. 

1 hour ago, Chriull said:

Maybe something made out of bricks/conrete would be better? Different compartments for each wheel separated by one brick/a concrete wall. Depending on the structure and how long and "strong" one wheel can burn maybe a chimney to keep the flames off the roof and (most of) the smoke outside the barn.

That would be ideal but my wife would put her foot down on building such a structure. 

  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites
13 hours ago, Smoother said:

That may be true for a flimsy sheet metal industrial building subjected to a sustained fire, but tool strong boxes are thick steel, have no heavy roof pressing down on them, and an EUC fire does not have enough material to burn for too long.  Also you are trying to contain the explosions that happen when the batteries explode, so that the fire does not spread.  A metal box is still perfect for this.

No, this applies to all steel structure, and the capacity to resist heating wholly rest in the overall mass of the steel. Here's a link with some additional info:

https://www.steelconstruction.info/Fire_and_steel_construction

So for smaller structural steel sections which will still be substantially heavier than the construction of a tool box you can have failure in as little as  12 minutes (at about 800 degree). Steel would be better at containing an explosion assuming it is weak and the box vented, and I assume the risk is mostly sustain buring of the battery packs. In which case, I don't think you can count on the steel tool box to provide any protection beyond a few minutes of low temperture burns.

Share this post


Link to post
Share on other sites

My $0.02;

Cent# 1: I agree with the metal box thing. That was my first thought on all of this.

Cent#2:  If the assumption is that after some degree of wear and tear, the cells become worn through or damaged, causing the fire...is there any way to inspect for this? I have seen the housings the cells are in, but has anyone taken those apart or know how the cells are packaged?

  • Like 1

Share this post


Link to post
Share on other sites
Posted (edited)
40 minutes ago, Circuitmage said:

My $0.02;

Cent# 1: I agree with the metal box thing. That was my first thought on all of this.

Cent#2:  If the assumption is that after some degree of wear and tear, the cells become worn through or damaged, causing the fire...is there any way to inspect for this? I have seen the housings the cells are in, but has anyone taken those apart or know how the cells are packaged?

They're usually shrink-wrapped along with the BMS, but the "configuration" inside can be different from one wheel to the next. Here's an old picture of unwrapped Firewheel F260 pack:

acYuSfB.jpg

The FW pack configuration was an unusual one, I don't think any other wheel has similar... there are two packs (the other one with the blue wraps can be seen underneath the unwrapped one) with a single BMS just in one of the packs, and all the balance & charge/discharge -wirings for the other pack running between. Each pack is actually 8S2P, which are then connected in series to make a 16S2P for charging and discharging. Still, this is similar to what you'd likely find inside most packs. There's a a cardboard (some might use plastic) insulator between the two rows of cells, and each two cells right next to each other are in parallel. I didn't tear it further, but there's a factory wrap around the cells themselves, as the entire outer casing is the negative terminal, not just the bottom. In the below picture you see a couple of positive-terminal insulators to prevent the terminal from coming into contact with the edges (which are part of the negative terminal):

Battery1865022.png?resize=701,309

If there's a tear or a dent in the pack itself, or in the cells (of course you won't see them through the outer wrap), I'd be wary of it. A hard knock on the pack (or the pack moving inside the battery compartment and hitting something) could cause such, but from what I've seen, the packs usually sit pretty snuggly in their compartments and won't have space to move around. Ninebot One (C/E) had a design fault where plastic "edges" (probably to make the part stronger) on the battery compartment cover could cause damage over time to the wrapping, possibly even the cells.

The big issues are either cell puncture (probably immediate fire), the internal electrodes pushing against each other (internal short circuit, the cell will at least heat up considerably, if not catch fire/vent) if the cell has a dent, or the welded tabs on the positive terminal making contact with the outer edge (negative terminal, external short circuit).

pack1(1).jpg

18650 internal structure

18650cellShort.png?w=600&ssl=1

Welded tab has punctured the wrapping and made a short circuit, in this case it was when the guy was dismantling the pack, apparently only made a huge spark and burnt the wrapping, but if in a pack it would stay there, then there'd be trouble.

Other than that, based on what I know, voltage changes are a good indicator of cell health, if one cell or cells in parallel (two or more cells directly in parallel will always have the same voltage) have lower voltage than the the rest, it's an indication of wear and tear (they discharge faster, possibly also recharge faster). Slight changes (something like around ten millivolts or a little more, 10mV = 0.01V) are probably just normal, that's what the balancing's for, but if the voltage of some cells drop much further from the others (or raise above them), there's something going on. Either the cell(s) has/have "aged" faster than the rest, or there's some other internal damage, like dendrite build up.

Edited by esaj
  • Like 3

Share this post


Link to post
Share on other sites
12 hours ago, Circuitmage said:

If the assumption is that after some degree of wear and tear, the cells become worn through or damaged, causing the fire...is there any way to inspect for this? I have seen the housings the cells are in, but has anyone taken those apart or know how the cells are packaged?

There was a post recently of a ?MSX? rider who discovered dented cells and mechanicly abused contacts after a crash by visual inspection.

The other "danger" of dentrites causing internal shorts with aged cells should imo go hand in hand with lowered cell voltages.(1)

We have no statistical data which of these two possibilities have real life relevance for the seldomly happening burning euc's.

And from the burned euc there is not enough left to do forensics - at least i never saw one.

So it the fires could start by different reasons. Could be, as @esaj wrote the cell wrapping getting worn over time causing some short (there is afaik no adfitional layer between the ) cells, could be a badly positioned motor/battery cable/connector getting too hot and melting some insulations - since we don't know the exact reason(s) there is plenty of room for wild speculations.

(1) lowered cell voltages can only be measured by opening the pack. As needed for the visual inspection. But opening and rewrapping could create a "real"/new problems, if this for example then allows humidity getting in/gathering inside, what was stated as reason for the inmotion battery problems....

Overall pack voltage measurement gives only restricted/delayed insight. The weakest cell is discharged first and also the first to be charged again. In the first step there is just some slight imbalance between the batteries - the sum is still 84V. The next step is, that the weak battery gets charged to the upper threshold around 4.25-4.27V and the charger input gets cut off somewhere whilst the CV phase. The other cells had time enough to charge up, so the sum stays at 84V. This stage can be discovered by the charger shutting off at a higher resisual current than before (and a bit less overall charging time/energy put in)

The next stage is, that the weak cell reaches the upper voltage threshold fast enough, that the other cells don't get charged enough and the overall pack voltage gets below 84V.

As the weak cell gets by high and low voltages more stressed than the others the difference will increase and this weak cell will reach the "dangerous" voltages below 2V (~1.5V?). The cell can still "survive" this some time and get a full charge till the upper voltage threshold, but after enough deteoriation it will stay at this ~1.5V. This stage is easy to discover - the pack can only be charged not be charged to 100% anymore (even the app with their "hysteresis" will not show 100% anymore).

The first stages are hard to detect - depending on the charger/wheels internal measurement and used multimeter accuracy. But with ongoing measurements to compare, and if the timely drift of this (in)accuracies is not to big one should be able to get it quite early.

At least the "real" dangerous/increased risk state once one cell stays at or below 1.5V has (almost) to be actively ignored - this 19*4.27V+1.5V ~ 4.13V per cell should be quite around the threshold the app shows less then 100%. Ups - just double checked - the apps show 100% down to 4.11V per cell...:(

  • Upvote 1

Share this post


Link to post
Share on other sites
On 5/10/2019 at 10:07 PM, NylahTay said:

So this leads me to ask, when you are finished with an EUC, and have determined that the wheel has finally gone too much to be safely ridden, how do you properly dispose of one? I mean the batteries are arguably the most dangerous part of the disposal.

Note: I don't get to a post on this yet before I posted this...

Yes battery as in any electronic is a hazard to dispose of. But in most cases you can somewhat easily get the battery pack out. 

But there is another hazard too. The strong magnets in the motor. Those can cause a lot of harm too. The sheer force each magnet can produce is high. As an example I can say the magnet used in lenses for digital camera are strong but small. Those are used for the image stabilisation gyro. The motor magnets in an EUC is much bigger. 

If you get a finger tip in between Is gyro magnets it can pierce/tear you skin. And it hurts like h.... 

The magnets are easily overlooked as a hazard.... 

  • Upvote 1

Share this post


Link to post
Share on other sites

This thread gave me the motivation to take action, and here's my setup:

20190518-045008486-i-OS.jpg

The EUC is sitting on its stand on a cement floor in the garage, with a little over a foot of space in every direction (except for the tripod standing next to it).  Can't really make much more space as my garage is cramped.  Using my bike repair stand tripod, I've hung an Elide fire ball directly over the machine.  The ceiling directly above that has a wireless-interconnect smoke detector (ionization type), interconnected with another same-model unit I keep upstairs in the living area.  You can see that there's a sprinkler head nearby on the ceiling as well.  Finally, you can also see a CO2 fire extinguisher on the wall by the door to the stairs going up to the living area (that's always been there, as has the sprinkler head of course).

If the battery pack should go into thermal runaway, the idea is that the smoke detector would trigger and the alarm would sound in the garage and upstairs at the other detector; the fire ball would go off (making a very loud bang that also serves as an alert) and coat the immediate area in a fire suppressing powder to delay ignition and propagation beyond the EUC itself; and eventually the heat would pop the glass trigger on the sprinkler head and get some nice flooding going; sprinkler system activation also causes a mechanical bell on the side of the building to start ringing.  I'm lucky to live in a relatively new condominium that has a sprinkler and fire alarm system.  Another stroke of luck is that my local fire house is literally 600ft away, approximately, so response time would likely be very quick.

I could get a metal box or build gypsum-lined coffins, or whatnot, but I feel this is a reasonable amount of due diligence for a low-risk, high severity event.  Charging is unattended but I do go into the garage to disconnect the charger when the charge is complete, and don't allow overnight charging.  A heat detector was suggested, and I'd have liked to try one of these as smoke detectors can give nuisance alerts in garages due to exhaust from a car, but these are not readily available in the U.S. (except as part of a full wired system), and not commonly used here for some reason.  Also, the sprinkler head would already act as an effective heat detector.  Now we'll see whether entering or exiting the garage with my car will set off the smoke detector, which would of course make the smoke detector idea untenable.

The only thing I'm looking into changing is that since I don't need a smoke detector upstairs (the whole home is wired with building-wired interconnected smoke alarms already), I'd like to get a "smoke sounder" (Kidde's RF-SND), which interconnects to the smoke alarms I purchased and simply provides a remote alarm function.  However, I haven't been able to find this product for sale anywhere.  I've sent a support request to Kidde customer support to ask where I can purchase one.

  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites

Wow @svenomous you really took this issue to heart.  I've been thinking about the issue too, (although not for myself because I'm not giving the issue as much consideration as you have) as an alternative to the steel tool box discussed, or your nuclear fall out shelter :D...how about laying the wheel flat on its side, building a simple wooden box that covers it on  5 sides (not the floor).  Make the box the same dimension as some end of line/remains, on sale ceramic tiles, and line the box with the tiles.  Now one has an explosion proof and fire proof box that will contain the fire and most of the smoke, for a few dollars of supplies and some time.  As a bonus, if the house is robbed the robbers might not think to look under the plain wooden box on the floor, especially if some oily rags and a can of STP is stored on top of it.:D

Share this post


Link to post
Share on other sites

sounds like he beat the hell out of something with giant lithium batteries... it's not absurd that something like this could happen especially with that many miles.. they're not invincible

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...