ROAR of the Pelican (CR123 Explosion during use, firsthand account)

[batterystation]

This spring in this particular light is so heavy and tight that I don't think that is possible, but that is understood and I will try causing that to happen too.

[Cmoore]

I believe the Pelican is suspect. There are thousands of CR123 lights using two cells in series in use, perhaps millions. Many are low cost products and some of questionable quality. People are using all sorts of CR123s and mixing brands as well as used capacities. Yet, we seldom hear of events such as the one addressed here. All technical conjecture aside, just the numbers alone suggest to me that this model of Pelican flashlight just may the root problem.

[KevinL]

I've been busy and under the weather for the past few days and I just read this thread.. I'm pretty horrified by the secondary effects of the lithium 123s - sure sounds like a vicious chemical brew to deal with. I had no idea we had HF lurking so near us in our prized cells. However, as I read the first few posts, I couldn't help but think, were the glass fragments contaminated with whatever was in the battery - I didn't know exactly WHAT was inside the battery, but I would certainly have told my doctor where the frags came from and highlighted the potential of contamination. Not finding fault, and I too am grateful for the observant CPF members who pointed out the potential to LM.

LM, get well soon. That is a really nasty situation and I'm glad things are improving for you.

I'm also very grateful that it doesn't have anything to do with the mod I created or I too would be losing sleep.

[bwaites]

"I believe the Pelican is suspect[I][U]."

Guys, please read the WHOLE thread and each post in it's entirety. People keep saying the same thing!

Multiple people have brought up the fact that this has happened with different lights and different brands of cells.

The last "explosion" was Surefire cells, and Kevin from BatteryStation has talked with multiple sources and researched it and says, "I have done some searching and found that this is really not unique to Pelican. This is an industry wide problem that has existed since flashlights were invented but has seemingly been compounded by newer battery technologies which have much higher energy densities and load currents".

He also says, "At 2AM I spoke with the factory and they said the same thing we all have so far. Not real sure what caused it but would like the cells for examination. They have had a few incidents in Europe with other CR123A batteries that are similar to this but all seem to be "freak accidents."

The PM6 MAY have a spring issue, the cells MAY have been mechanically damaged. BUT we have no more evidence of either of those than we do that an unseen lightning bolt descended from the sky, struck the light and caused all of this.

Right now, we have exactly NO science about why this issue occurred, but we have lots of conjecture.

I hoped to try to replicate the accident last night by damaging a cell and recreating as many of the circumstances as possible, but an old friend came to town unexepectedly and we spent hours reminiscing instead. After reading about Steve's illness, I think I want to do this with some more containment of the contaminants, and I may see if I can talk my way into the blast chamber at one of our local propellant factories. (They make the explosive propellant for airbags.)

For now though, all we can REALLY say is, BE CAREFUL!

Bill

[Cmoore]

"I believe the Pelican is suspect[I][U]."

Guys, please read the WHOLE thread and each post in it's entirety. People keep saying the same thing!

Multiple people have brought up the fact that this has happened with different lights and different brands of cells.

Bill

What are the specifics of these other events such as the model of flashlight? I've not heard of them. The only ones I read about here are with this model of Pelican light!

[Diesel_Bomber]

Something occured to me reading StoneDog's post, but I don't have a PM6 to check it on. SD states that he'd unscrewed the tailcap enough to lock-out the switch. Would this allow the batteries to rattle loosely under vibration and possibly cause damage? Think of a shake-light, and replace the magnet with these lithium batteries.

LM, glad I could help with the MSDS. Hope you get well soon.


Cheers.

P.S. I, too, will not be letting kids handle my lithium powered lights while I'm LPing.

[bwaites]

It has been reported with Surefires and Streamlights used in Iraq, and those reports eventually made it back to CPF. My brother in law, while in Baghdad, heard about them in the military grapevine. Unfortunately, we don't have the level of documentation from those that we do here, but I suspect that they must have happened, or why talk about it in the first place?

But as Kevin notes..."I have done some searching and found that this is really not unique to Pelican. This is an industry wide problem that has existed since flashlights were invented but has seemingly been compounded by newer battery technologies which have much higher energy densities and load currents".

I think when it does happen, it is a freak thing, probably with multiple contributory issues, but once again, that is NOTHING BUT SPECULATION, this time on my part!

It certainly isn't EASY to make happen, as Kevin and I have so far found out!

Bill

[Paul6ppca]

LM ,
Hope your doing better today. Please keep us updated.

As a relative nubee, 6 months, I was wondering if this problem with poison gas/explosion is more concentrated with rechargables,and is there danger (less danger)with using single cell lights? I understand the 2 cell mis match concept,but are there other danges in single cell lights that might cause this catastophic results? Is it mostly during charging that presents a danger for single cell use?

[tebore]

LM ,
Hope your doing better today. Please keep us updated.

As a relative nubee, 6 months, I was wondering if this problem with poison gas/explosion is more concentrated with rechargables,and is there danger (less danger)with using single cell lights? I understand the 2 cell mis match concept,but are there other danges in single cell lights that might cause this catastophic results? Is it mostly during charging that presents a danger for single cell use?

Seems like it's more likely to happen with 2 cells.

Back in the early days of RCing when people started to use Li-Ion cells or Li chemistry based cells this also happened when you overcharge or discharge cells. From what I've seen the new Li-ion cells have protection to prevent overcharge and discharge of cells. Numbers didn't play a roll. So if you aren't using protected cells the number of cells won't matter.

[Arkayne]

W O W . I've always thought the quality of a product is only as good as the manufacturers customer service. It's good to see that Battery Station is willing to step up and replace the damaged goods! Props to you BS!

I don't think I'll store my ammo with a flashlight anymore.

[enLIGHTenment]

BTW: isn't this why they use the platnum catalyst pellet?

The catalyst is there to deal with hydrogen production. It won't do anything in the event of electolyte ignition.

I'm talking about the spring ending up next to the + button and shorting between the inside of the torch casing ( Which is directly connected to the - of the cells ) and the + of the cell.

If a cell is shorted externally, the PTC will trip off before anything goes severely wrong. To ignite the electrolytes, a cell would have to short internally or be shorted externally and have a failed/defective PTC.

[Lunarmodule]

LM ,
Hope your doing better today. Please keep us updated.

As a relative nubee, 6 months, I was wondering if this problem with poison gas/explosion is more concentrated with rechargables,and is there danger (less danger)with using single cell lights? I understand the 2 cell mis match concept,but are there other danges in single cell lights that might cause this catastophic results? Is it mostly during charging that presents a danger for single cell use?

Many moons ago I was employed as a research chemist and spent the majority of my time working with relatively benign compounds. Petroleum based solvents were the nastiest of the compounds I dealt with, and on one occasion I had an exposure incident when a storage tank catstrophically leaked in the plant where I worked. I shrugged off the stringent protection guidelines and instead went into the thick of the spill to take readings and calculate risks. I failed to consider my own welfare in that headstrong move and paid for it with a nasty respiratory tract burn. That was over 20 years ago and nothing has recalled that incident until now. I cannot stress this enough. SilverFox and others h. ave posted about the stuff from the combusted batteries being "real bad" for you in plain English. They are quite right. Moreso than I imagined. To relay my real world experience with this: even a SMALL amount of the gray gas expelled from the batteries during combustion inhaled can do significant DELAYED ONSET damage. I read about Icebreak and Bill Waites trying to replicate the failure, and I applaud their courageous and determined spirit, however I cannot stress highly enough that one should overexaggerate the danger of the fumes and wet char residue, which I handled with bare hands (no choice). This is not mamby-pamby MSDS Chicken Little the Sky is Falling scary sounding rhetoric... an MSDS would have you believe a car wash solution is so dangerous to handle you might want to invest in a full blown Hazmat suit in order to wash your car. They typically refer to catastrophic spills of material or somesuch. In the case of these batteries, I was shocked to discover almost all the MSDS was devoted to minimal discussion of how to deal with burned/burning batteries. There was a clear trend of disclaimer-ism: "in normal use the compounds contained within the metal can of the battery cannot come in contact with the user and thus are not subject to special handling precautions. There is a BRIEF mention of some of the potential decomposition byproducts of combustion, but the annoying phrase repeats saying this would only be applicable for abuse or disaster scenarios. That could stand revision. And to Bill Waites, PLEASE be super careful. You mentioned "fire is OK, an explosion is not....." in an earlier post. Well, what is happening or rather what did happen to me is the rapid venting was an out of control reaction producing large amounts of hydrogen. I recall that moment in high school chemistry of tossing a chunk of cryogenic sodium metal into a bath of water, it oxidized violently and liberated prodigious H2 (hydrogen) which eventually ignited from the intense heat of the vigorous reaction. Lots of bubbling, and a giant KaBOOM! as a climax. Strikingly similar to what occurred with the cells. The rapid venting (and I cannot emphasize this enough) is the only early warning sign but by itself is the true menace and real demon in this process. The hydrofluoric acid in this gaseous emission is horrifically toxic, I radically underestimated its impact on me, tricked by the delayed onset nature of the symptoms of its poisioning. I've turned the corner and improving, but I feel worse than when it first happened. I didnt sleep at all last night, threw up twice, and sweat so much I showered three times. I've got to get more rest, but the leg burns, I feel nauseated, and I'm still looking like a measles poster child. Nothing to do but wait it out. I can pass along to others that if you are exposed to a scenario of gray haze/smoke jetting from the battery, get yourself to a source of fresh air as fast as possible, and keep anyone else as far away as possible. Under no circumstances should you breathe ANY of the fumes if it can be avoided. In retrospect I should have gone outside instead of typing for 20 minutes in the same room with the evil haze. AVOID THIS STUFF AT ALL COSTS. There is nothing but a respirator, or better bottled air to cope with these fumes, they are more dangerous than I thought possible. BIll, I urge you to be extremely cautious if you insist on trying to induce this failure. If you manage a high pressure cell venting, the explosion is almost certain, as the cell vent exists to stop a runaway reaction. But the greatest danger is the chemical nastiness you can breathe. That needs to be emphasized in the MSDS sheets. And made common knowledge here. More than typical common sense suggests, take extra precautions to avoid any exposure to the fumes of burning batteries.

Im feeling wiped out again, a lot of typing. Going to lay down for a while. Thanks again for everyone's support.

Last thing: I received email from Pelican, and their rep is on ball with a speedy and aggressive response to this incident. Only person with a lot less sleep than me is Kevin of Batterystation, he deserves commendation for his lightning quick reaction to this situation. Kevin, please rest easy, at the end of the day you have done all a single person can to eradicate this probability, and that has to suffice.

We need a nifty "dont eat yellow snow" type of slogan for avoiding these gray gas jets that come from the violent end of CR123s.

[lebox97]

not sure if this helps the cause or not, but found some interesting info/criteria for testing of R123's... (and some general handling precautions/reminders)

they seem to have a pretty thorough (10 step) test procedure for trying to get a cell to self-destruct!
(crushing, piercing, free fall, impact, vibration, temperature shock, heat, short-circuit, forced-discharge, overcharge)
here (bottom half of the page)

"kids, don't try this at home"!!!


get better soon LM!

[RebelRAM]

I have been following this thread and this occurrance is definitely eye opening. But at the same time it looks like there are a lot of variables involved here. It will be interesting to see if anyone here can duplicate the failure in a controlled environment.

Now that I know it could happen in any CR123 light, it does not worry me. I think it's probably a freak thing and odds of it occuring are very low. Think about how many 123 batteries just the CPF members go through. I only have 2 Surefires and I manage to go through atleast a dozen or two a year.

Even knowing all of this, I have thought many times about buying a Pelican M6
This event does not change my mind about that. I would still feel very comfortable buying one. Pelican has a great product. If there are changes that need to be made to it, I'm sure Pelican will research and address them. First responders have depended on Pelican for many years. I have several family members who use Pelicans at work.

I am very impressed by the response from Kevin at Battery Station. I have not seen customer service dedication like that in a long time. I deal with a lot of vendors and sales reps buying computer hardware for where I work. I have only ever run into one sales rep that was as dedicated as Kevin is to solving problems. We need more people like that working in customer service for all areas of products!

LM take care man! I wish you a speedy recovery and don't let this minor setback discourage your flashlight hobby. Actually come to think of it, this is the CPF, so you should acquire at least 2 more lights of any type just because!

You guys are a wealth of information here. I have learned more in the past two days about 123s than I ever had since I bought my Surefires a couple years ago.

[schill]

Back in the early days of RCing when people started to use Li-Ion cells or Li chemistry based cells this also happened when you overcharge or discharge cells. From what I've seen the new Li-ion cells have protection to prevent overcharge and discharge of cells. Numbers didn't play a roll. So if you aren't using protected cells the number of cells won't matter.

The current "hot" batteries in RC airplanes are Lithium Polymer (LiPo) batteries, not lithium ion anymore. The LiPos typically don't have the built-in protection that lithium polymer and other batteries have. The chargers and speed controllers need to have cutoffs to prevent fire and explosions - it is easy to damage the batteries by over-discharging them.

There are numerous incidents of people losing cars and homes due to battery-based accidents on forums like www.rcgroups.com. People have come up with all sorts of ways to protect things when charging (inside cinder blocks, special ceramic pots designed for charging (battery bunkers), and fire-proof bags, for example).

Check out the LiPo Fires videos on www.utahflyers.org. These are videos of experiments used to see what these batteries are capable of.

[Lunarmodule]

Phone rings. Again. The toxicologist from Queen's Hospital asking questions. They want me to go back in. At least for more blood. I hate that part the most. I just now read StoneDog's post and feel terrible. Its not fun, but what I've got is minor and nothing compared to what could have happened if I fell asleep reading with that light on. Jon, I dont believe you did anything negligent and there is no WAY I could have forseen this, much less you, so dont worry. Be grateful as I am that there is only minor damage to me and surroundings. The scary part is the chemical toxicity of what I interpreted to be a very slight exposure.

I want to clarify a point brought up earlier. The front cell exploded first, well, vented then the hydrogen BANG which made me drop the light, continued uninterrupted mean hissing/jetting fumes then a second (rear cell) explosion. The batteries were stuck inside the light, fused together in a single charred WET lump. The polarity was correct, positive end toward the head of the light, to clear up one poster's misunderstanding of reverse charging. The reverse charging discussed is a phenomenon where the polarity of the battery is correct, cells aligned and pointed the right way, but significant differences in cell potential cause current to flow "backwards" and a stronger battery tries to charge a weaker one, discharging at a much higher than normal rate doing so and supplying the lion's share of the load at the same time.

Also, the batteries stayed together and inside the light because there was insufficient force to blast the tailcap loose, there was no place for them to go, and they swelled up and occupied every square centimeter of the tubes interior. It took some aggressive coaxing with a Philips screwdriver to get the batteries out, thats when I took the pics. The rubber button switch blew apart in the first explosion and sent parts flying. I knew it was the first cell (head end) to explode first because I saw a much smaller flame jet from the first bang. The pattern was eerily similar, like a copy-paste thing. One cell vents violently, hydrogen ignites *BANG*, second cell vent opens for roughly same amount of time (10-12 sec) with similar *Bang*. The second bang had a much larger flame come from the tailcap, plus I heard the bang was much louder, it seemed it came from just the other side of the tailcap instead of the more muffled one from the head end of the light. The screw in lamp assembly held a perfect seal. NO combustion by products escaped that seal. IN fact, the lamp was not only intact but looked almost unused, but as you can see in the pics everything south of the lamp assembly seal was armageddon. Also, forget about the spring jumping sideways or anything like that. Turbo DOg brings up some EXCELLENT points. I have 2 Surefire M6es and a slew of 2- and 3- cell lights and I have never had a problem. Statistically the M6 should be many orders of magnitude more likely to report similar incidents if it was a simple issue of a statistical cell mismatch. Not so. The light and batteries are going on a long trip, parts to China, for a detailed analysis. I have asked as a prime consideration to be throroughly informed of their findings and any design changes considered. Rather than look at who is the bad guy, focus attention on the fact that to my suprise and delight, both Pelican and battery station have come out right away with a determined attitude to spare no effort to make the whole scenario a thing of the past. I applaud that, and its a heck of a lot better than pointing to that "evil" light or battery. Bill Waites is spot on in his reasoning about the situation. I just dont want him to attempt any more simulations. Im dying for more answers but am very glad to give this to the fellows with the tag closed cup calorimeters and spectrochromatographs.

[StoneDog]

I still can't believe this happened. This light sat by my bedside as a "bump in the night" light for a long time basically unused. It was going to host a McModule, I just never got around to building another one. The batteries were basically new (as I said, less than 2 minutes over the course of a year) and I shipped them only as a courtesy. In retrospect Steve could've done without that courtesy...

I was away at a conference when Steve first posted this and, since I don't normally read the General forum anyway, had no idea until this morning.

Obviously I have refunded Steve's purchase in full and will be decomissioning my SF G2 and any other high-current multi-CR123 lights I have. I gave my father a PM6 modded with a V2T LuxV McModule two years ago but I believe I'll ask him to either pitch it or store it without the batteries.

Again, a heartfelt and sincere get well soon from me and my family. Kudos to both BatteryStation (of whom I am a regular customer) and Pelican for stepping up.

Jon

[batterystation]

I am glad Steve seems to be feeling better. I have a host of things going to try and duplicate this problem with various brands of batteries. We had a late meeting here and are going to put the L-AA project on hold while we devote more time and resources to this CR123A thing. I also still need Steve's address for his shipment.

Thanks for the testing ideas. We are going to do everything possible to blow something up. Another one passed the clamped in vise test. Crushed the button level with top of the can.

We will be comparing notes with Pelican and supplying them with any needed info regarding this situation.

[cheapo]

it isnt the springs... it just isnt, i have had my pm6s with really tough spring for soo long, it it isnt nearly strong enough to break through batteries.... Even with r123s, which fit REALLY TIGHTLY, and you cant even twist the tailcap all the way down... it doesnt break the batteries. It might just have to do with both the bulb and faulty batts.

-David

[enLIGHTenment]

The second bang had a much larger flame come from the tailcap

What color was the flame?


@gl22man: That was pretty tasteless.

[cheapo]

The batteries are used in series. Voltage has little to nothing to do with what you think you are talking about as lithium cells maintain their voltage extremely well. The real issue is SOC or state of charge.

I'm not ready to go boycott pelican, but it seems there is something about this light.

However, even IF it is the light... we're trying to pinpoint a failure mode for a light than many CPF people own. A light that _many_ have ran countless cells though. A light that certainly cells with varying SOC have been used in, and that did _not_ destructively fail.

So we've got 2 hard searches going on: 1) batterystation and 2) the pelican dissection.

thats not my point, i am saying that it is the combination of the pm6 and a "faulty" battery that caused this. A pm6 wouldnt explode with all good batteries, and the reason it only happens with the pm6 is because other lights can handle the faulty batt.

-David

[IsaacHayes]

There isnt any such thing as a light handling faulty batts so it doesn't vent them. Now handling the venting so it doesn't blow apart is diffferent.

[TinderBox (UK)]

would activated charcoal absorb the hydrogen gas before it ignites.

could a small packet somewhere in the flashlight help.

regards.

[bwaites]

David, you keep saying that "and the reason it only happens with the pm6 is because other lights can handle the faulty batt."

I'm confused, what do you think other lights do to "handle" that faulty battery?

These are dumb lights without circuitry or anything else to handle a malfunction. Other 2 cell incandescents are EXACTLY the same, with the possible exception of the springs, which do indeed, appear to be stiffer on the PM6 than most similar lights, and you yourself don't think it has anything to do with the springs, so what is different about the PM6?

Bill

[RA40]

This is a very devistating event and Steve, you are in my thoughts.

Sad this has occured yet the awareness this has raised, has positive aspects. I applaud the desire of those involed in determining what caused this failure...this will in the end make using these tools much better for all of us.

[FlashlightOCD]

I think some people are under the wrong impression that the nasty stuff Hydrofluoric [HF] Acid is in the 123 batteries. It is not there in a normal battery, the HF is a byproduct of the chemical reactions that take place during thermal breakdown.

LM,

Sorry you have to go back, the emedicine site posted a few pages back says; "The major pharmacologic intervention is with calcium gluconate. Depending on the clinical situation, it may be used in a gel, intravenously, or intra-arterially".

Please, follow the doctors advice and get well soon.

[KDOG3]

I think some of us need to take a deep breath and a step back. I don't see Lunermodule - who is the victim here - blaming anybody. So I don't think anyone else should either.

I guess what I mean is that for the amount of incidents vs. the amount of these batteries sold/used, I don't think there is much to worry about, its more of a fluke. Not to lessen the pain and suffering LunarModule is going thru mind you, but lets' not get ahead of ourselves.

[cheapo]

David, you keep saying that "and the reason it only happens with the pm6 is because other lights can handle the faulty batt."

I'm confused, what do you think other lights do to "handle" that faulty battery?

These are dumb lights without circuitry or anything else to handle a malfunction. Other 2 cell incandescents are EXACTLY the same, with the possible exception of the springs, which do indeed, appear to be stiffer on the PM6 than most similar lights, and you yourself don't think it has anything to do with the springs, so what is different about the PM6?

Bill

some people blame it on the pm6, some on the batts. Now, it seems that only the pm6 blows up right? And since it only happens occationaly, it is probably do to a variation in batteries... in other words, a "different battery". All lights have the same probability of getting a "different batt," and since hardly any other lights have exploded, it makes on think that there is something about the pm6 that causes the "weird batt" to explode.

Just my theory....


-David

[snakebite]

heres my take on this incident.
light has unusually stiff spring.got roughed up in transit or dropped.light is turned on and gets nice and warm.the insulator around the + plate insulating it from the can and crimp after being damaged by the combination of a possible thin spot and the above events fails as the plastic softens.
the cell is now dead shorted without the benefit of the polyfuse in the contact.poof!
try shorting one in a controlled test to see if it gets hot enough to hit thermal runaway.

[cave dave]

I would expect the first "bang" actually came from the rubber button cover popping like a ballon.