A reminder of lipo do's and don'ts - Fire aftermath pics

[Hyena]

I'm almost reluctant to post these photos as I think lipo is the greatest battery chemistry around and that anyone too scared to use it are a bunch of sissies and deserve bad backs or poor handling from lugging around heavier chemistries
However this does serve as a reminder of the do's and don't of lipo. This is a long post with alot of probably excess details, but think of it as case study

The back story is, I built a kit for a guy last year with an 18S lipo pack. I set up my batteries with pretty foolproof anderson connector blocks for parallel bulk charging in banks of 6S (with manual balancing when occasionally required) At worst if a cell does go bad it's potentially going to overcharge only one parallel group. And even then with the voltage set conservatively to 4.16v/cell things have to go along way south before the others end up dangerously overcharged.

So around 6 months later I got an email from the guy with the following:

Left it on charge the other night after work and was woken by panicking neighbours reporting a loud popping sound.
Opened the garage to find flames up to the roof! Luckily we extinguished it in time to cause only minimal damage to the house and a layer of black
soot over my brother's car is all that it sustained... not so lucky for my toasted bike.

Fortunately the fire was fairly well contained butI felt pretty bad for the guy and wondered how things went so wrong so quickly.
My first thought was that something must have gone wrong with the charger and a failed component ended up passing though a higher voltage to the battery - maybe even 240v AC some how.

Then I got some more info about the events leading up to it and quickly realised the problem. Just like in the early days of poor quality RC lipo packs, many of the fires were caused by using the wrong chargers. And this is what happened here, in a big way.

The first issue, which I doubt is actually related to the fire:

In my rush to attach my half finished enclosure to the frame at midnight so that I can ride to work at 6am, a single wire found its way wedged between a pack and the frame. This stripped the wire and as I had my key switch temporarily just hanging, when I hit a bump the bolts bumped against the frame causing it to arc. Bright flash and loss of power, and nice weld spot on the frame and lovely smell After pulling up and inspecting, disconnecting and reconnecting the battery... taping up with electrical tape... I continued as I still needed to get to work. Was working ok but not as much power. charged up at work ok and was ok on the way home. Checked it out and discovered the above mentioned stripped wire, mended, balanced battery, sorted enclosure and mounted switch. Had no more troubles with it using it daily (about 2 weeks before fire).

But what really caused the problem is what came next:

I bought a 600W charger, same brand as the one you supplied. It arrived with just the double red and black connector on it where I needed the six pole plug. After trying a few places (including Jaycar) no one had ever seen these types of connectors. I realised there was a double connector on the bike that connects the controller to the isolator switch. So I figured I'd give it a go charging that way (disconnecting controller and replacing with charger). It seemed to work ok except the charger did not automatically cut out when charging was complete. When I finally pulled it off and powered up the bike the battery was reading 81.6V! Ooops...At least it took off like a rocket for a bit! I didn't use that charger again and had no noticeable problems. (I think about 3 weeks before fire)

ACK - wrong charger used, WAY over voltage.
Each cell would have exceeded 4.5v. Ouch. Lucky it didn't blow up there and then. This would have damaged all the cells, greatly reducing their life. I imagine they were quite puffy after this but he probably didn't notice inside the enclosure he'd built. He continued to ride it for another 3 weeks during which time I imagine the cells rapidly deteriorated from being heavily overcharged.

I got home from work riding pretty flat out and battery warning was beeping and flashing at me. (which he ignored apparently, driving the cells probably well below 3v) I wanted to go out and get some food so I decided to give the fast charger another go and hooked it up again. Charging with that annoying beeping still going. After about 30 - 40mins I went out and checked it and the beeping was still going. (these cells obviously failed to recover above 3v after high current charging so they're officially dead by this stage) I hooked up the battery and it still showed 66V (multiple cells must have been well under voltage and presumably others were taking some of the charge) . I hooked up the charger and decided to raid the cupboard and fridge for scraps and went to bed. Next thing I know the neighbours are knocking on the door and well you know the rest.(I'm guessing the cells that DID take a charge would have ended up over 5v/cell based on the charge voltage last time when the cells were still good)

So, in summary the problems that lead to this failure.

* Used wrong charger
* Way overcharged cells with high voltage
* Ignored LVC and over discharged cells
* Ignored continued under voltage cell warning after charging, then left to charge unattended.

So basically everything you're NOT supposed to do with lipo!
1 simple email to me when looking for a second charger would have prevented all these problems.
But the other issue was ignoring the HVC and LVC buzzers.
Yes, you could argue a LVC that hard cut the throttle would have prevented over discharge and a more expensive / complex charger would have prevented over charge - but basic LVC buzzers are perfectly adequate if you actually heed their warnings, which I instruct people they must do. And with an ounce of common sense, these are the cheapest and easiest solution. If you think this leaves too much to chance remember that for nearly 100 years cars have used oil or check engine warning lights and beepers and everyone knows that if that comes on then you have to stop the engine or risk damaging it (in this case the battery)

So yeah I just thought I'd share this story, not to scare away the noobs or further propogate unsubstantiated hysteria around lipo use, but to serve as a reminder to follow the basic rules and don't be a stooge.

 

[Ykick]

Actually surprised it took that much wrong to go fireball. Cells are fairly forgiving up to a point which this end user found. Glad nobody died but a good reminder this stuff ain't for everybody...

 

[circuit]

In summary: a good BMS would have dealt with it in proper manner.

 

[miuan]

As much as having ignored all the good practice and obvious warnings the battery sent him, the owner did a good job of parking the bike away from flammable stuff. Thumbs up for that, it might have saved his house.
I've had various HK cells get anywhere from 0 to 4.5V and recover.. but even for good sane cells, the day of doom may come in random moment, so you need to mess with that cell medic every now and then to discover that before it's too late. I've also found the buzzers are unnecessary once you use a bar mounted volt meter and establish a good practice of avoiding near LVC conditions.

 

[dogman dan]

Man, he couldn't have done much more wrong if he was washing car parts in gasoline at the kitchen table while smoking.

Good example of why I wouldn't put together a bike battery, and turn it over to somebody myself. In this country, the insurance company would be suing you. Never mind that he changed everything, they'd still sue you anyway here.

 

[bigmoose]

Thanks for sharing the details of this incident. To me the insight is into how a user of the technology, who means well, will start to dither a well designed system and introduce fatal errors into it without knowing it. It also supports the golden rule of design that the best risk controls are in hardware, and that procedural controls (listen to buzzer and take corrective action) often fall short.

This incident had a very fortuitous outcome.

 

[Punx0r]

"A little learning is a dangerous thing" - Alexander Pope

 

[Ypedal]

I dont know if the guy is reading this but i dont care.. he is a royal idiot.... bottom line.. :?

This actually is encouraging in a way, those cells took way more than expected to go up in flames.. supports my single cell abuse testing results.

 

[ohzee]

Yea Lipo is not for everyone and id say that guy should be on that list.

If I ever built a bike like that again for anyone id send them to this thread or hand them that story.
Should be enough right there to get some lipo respect.

Thanks for sharing.

 

[Alan B]

Very interesting case study. An example of a little knowledge is a dangerous thing. Very difficult to protect against someone plugging the wrong charger into the wrong place. I suppose an onboard AC charger/BMS would help. I'm not sure what documentation was supplied with the system but something there might have helped, if it had been studied. Glad the outcome was such minimal damage overall.

 

[grindz145]

He's lucky thats a brick wall...

 

[mvly]

At first I was like shit he did nothing wrong and the lipo burned! Glad there were multiple reason why it burned.

 

[jtchoy]

Now that was an accident waiting to happen!

Lipos definitely need prior knowledge and respect before use. Once you have that, they are one of the best batteries to use.

 

[pff7]

I don't understand, you supplied the charger----why did he need to buy one?

 

[LiFe]

IMHO, your mistake was in fabricating a LiPo pack for a mate. You assume the responsibility and liability when shit happens.

Granted, it was likely done as a favor. And no good deed goes unpunished

 

[deVries]

Man, he couldn't have done much more wrong if he was washing car parts in gasoline at the kitchen table while smoking.

The problem with this non-comparison is most any guy understands what happens with above scenario, but no one should hand over RC LiPo to "retail" first time ebike buyers. Either you "the buyer" already handles HK type RC LiPo as a hobby, or you're a *very* serious ebike "nut" that loves spending lots of time learning and doing this as a *serious* hobby w/excellent learning abilities and clear understanding about the risks of using RC HK LiPo. HK RC LiPo should *never* be put into the hands of a newbie without the above prerequisites, IMO.

Drunkskunk's video of a single cell catching fire is enough to explain "why" I'm certain about this.

 

[cal3thousand]

Thanks for sharing that and not closeting it. We have a need to know about these types of failures ( in this case human).

Glad that nothing really got damaged; he owes those neighbors big time.

People need to start treating LiPo (actually all batteries) like little canisters of gas. Once you give it that level of respect, then you MIGHT be ready to handle the proper use. To store that much energy in something and not respect the potential it carries is a dangerous way of living. There are certain rules that must be heeded when dealing with gasoline. Different, but just as important are the rules to handling LiPo.

If you're gonna play with fire, respect its power.

 

[cal3thousand]

Man, he couldn't have done much more wrong if he was washing car parts in gasoline at the kitchen table while smoking.

The problem with this non-comparison is most any guy understands what happens with above scenario, but no one should hand over RC LiPo to "retail" first time ebike buyers. Either you "the buyer" already handles HK type RC LiPo as a hobby, or you're a *very* serious ebike "nut" that loves spending lots of time learning and doing this as a *serious* hobby w/excellent learning abilities and clear understanding about the risks of using RC HK LiPo. HK RC LiPo should *never* be put into the hands of a newbie without the above prerequisites, IMO.

Drunkskunk's video of a single cell catching fire is enough to explain "why" I'm certain about this.

That's why my wife will never be allowed to charge the LiPo on her new bike. I will handle all the responsibility and liability.

 

[mr.electric]

A full BMS would have prevented this incident. The wrong charger and a host of other mistakes would just result in automatic shut off.
The only people qualified to use lipo hobby packs unmanaged for ebikes are those who put the time in studying the dos and dont's...basically people who are on endless sphere frequently.
I'm eager to see a full bms designed for use with hobby king.

 

[Alan B]

So the customer buys the wrong charger and plugs it into the cable going to the motor controller since the plug didn't fit the charging port.

In many cases the BMS would NOT protect against this situation. The BMS is protecting against charging through the charger port, which was not used in this case. Even if the BMS had an FET to disconnect the motor on low voltage, this path from the controller back into the batteries would be passing through the FET's inherent body diodes, and so would not be blocked even if the BMS opened the FET (which would not be part of its normal operating design).

Essentially you have a customer rewiring the setup and using the wrong charger. Pretty hard to protect against that.

 

[Jason27]

The only person that can be a liable for this is the person who wired the battery packs. I agree that if you don't have any experience you should not be wiring the battery packs. I only use my lipo short rides less than 5 miles.

 

[Chalo]

In summary: a good BMS would have dealt with it in proper manner.

If it worked.

I have had enough Chinese made chargers malfunction that I would not trust my house to a Chinese BMS.

Chalo

 

[Chalo]

People need to start treating LiPo (actually all batteries) like little canisters of gas.

You don't plug your gas tank into a maybe/maybe-not igniter when you get home. But you do with your lipo battery.

This is less a reminder of lipo battery safety guidelines than it is a reminder of why e-bikes, and EVs generally, are not ready for prime time yet.

Chalo

 

[etriker]

I really like taking apart battery packs to see what is inside.

The failed rc lipo packs I have taken apart looked like they were quickly built without much care.

Poorly built battery packs made not to last a long time.

After seeing how poorly they are built it is no wonder they fail like they do.

I think they are a waste of money for ebike use when the A123 20ah stuff is so cheap.

Here is a don't for you.

Don't waste your money on rc lipo for your ebike.

Rc lipo for ebike = fail !

 

[Ypedal]

wrong wrong and wrong.

this is a case of an ignorant dumbass user.. buzzers going off.. WARNING WARNING.. yet keeps on charging and riding.. without reaching out to the seller for feedback = fail.