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

[Topper]

Quote:
Originally Posted by Geologist
I thought Topper said he was using SF cells?


"Thanks, Geologist. I had forgotten that."

Maybe not forgotten..

Gee, I thought Topper said they were not Battery Station cells.
I also thought Topper was not mentioning "what" cells he used because he thought there was more involvedthan just the cells??
But I might be misstaken.
Topper

[PlayboyJoeShmoe]

I also believe he was keeping what brand of cells "under his hat" so-to-speak.

If you have an early model PM6 or any other light with lots of spring pressure you just might want to remove some of that pressure.

Of course NOTHING we know so far indicates that I have any idea what I just said.

[Icebreak]

SNIP...I used different cells than Battery Station...SNIP

ACK.

I don't like making mistakes in a technical thread. My apologies.

- Jeff

[Art Vandelay]

Does anybody know if BatteryStation batteries have PTC protection? Not that there is anything wrong with that.

Thanks

[Icebreak]

Art. Check out post 509.

[NewBie]

Okay, I ran two tests now, one with a cell depleted to 40% and another depleted to 60%, each of these in the head end of the light, and a fresh new cell.

I used an older PM6 that has the adjustable focus type head.

The body of the flashlight, in the center, reached 110F.

I let each sit for over two hours, taping them on video, at least 1 1/2 hours after the light output quit.

Upon removal, each of the depleted cells had a negative voltage on them when measured, which quickly slowly crept up and eventually went positive.

[OutdoorIdiot]

NewBie,

That was a rubbish experiment! Where's the video footage of the huge explosion? I want my money back.

Seriously though, that's possibly good news for owners of the old PM6. Bad news in terms of the attempt to get to the bottom of what has been causing these explosions, though.

The experiments that Kevin and others have been doing, combined with this latest one, would suggest that either the reported incidents are indeed random, or are due to something we haven't properly investigated yet. On the other hand, the fact that Topper repeated the failure first go, would suggest the that the incidents are not random, and can be repeated, if only we knew how.

Oh well.

In an attempt to say something useful:

(1) I still think it is worth Kevin using the old style PM6, if he can borrow one from Icebreak.

(2) Did you notice any deformation of the cells, on account of the spring?

(3) In post #464, Snakebite managed to get a CR123 to vent with flame, by heating the battery up then crushing it. Kevin has tried crushing batteries (to no dramatic effect), but not while they were warm (please correct me if I have misunderstood, Kevin). Perhaps it is now worth concentrating on physical damage to a running/warm battery, since this seems to be a known issue with the old PM6 spring?

[NewBie]

NewBie,
(2) Did you notice any deformation of the cells, on account of the spring?

No deformation with new cells on this light, top or bottom, even when cranked down. I can see decent marking on the rear cell bottom.

I even left the last test going overnight, nothing happened there either.

I'm starting to think about it being due to a certain cell with a defect or condition inside. If this is the case, it may take thousands of cells to find it.

If you can't create the conditions, and repeat the failure mode reliably, you don't have much to stand on. Of course, we may have not created all the conditions yet either...

[batterystation]

In addition to testing, Pelican has also been kind enough to send me a few old style M6 lights. They are in route here. I kinda shifted gears on the tests after talking at length with Lunar Friday night. I was planning all these radical tests that by his opinion did not really duplicate the situation. So rather than do more crushing and puncturing, we were just trying to recreate the sort of thing that happened. Hence, we are going to take a pair of mismatched cells and freeze them first before starting the test and not allow any of the condensation to evaporate prior to turning on the light.

[OutdoorIdiot]

Sounds like more good stuff to come in your experiments, Kevin.

In addition to what you suggest, I'm also thinking:

This post in this thread:
http://www.candlepowerforums.com/vb/...&postcount=165

combined with NewBie's description of seeing no signs of battery deformation, would suggest either that not all cell cans are the same, or not all PM6 springs are the same. Note that the picture of the deformed cell, as Milkyspit points out, came from a single-cell light (just goes to show that one can't be complacent even with a single-cell light).

So I'm wondering what the springs will be like in the old PM6's that Pelican are sending you, Kevin? It might be worth you assessing them when they arrive, and seeing if you feel any are particularly stiff?

Metalhead has already suggested trying to get some sort of quantative comparisons between spring strengths. It is possibly also a good idea, if possible, to get some quantative comparisons between battery can strengths, certainly for a force applied to the base.

Specifically @ Topper and LunarModule: Do you have any reason to believe your PM6's had particularly strong springs? I don't think that question has been asked yet, but I may have missed a post.

{Edit: Just to be clear, Kevin: I know you state that you are off the idea of crush tests, and are thinking of performing the freezing tests with moisture in mind. However, I think the freezing tests will also cause expansion and contraction of metal (and possibly other - must take another look at those battery internal pictures that Newbie posted!) parts, which could concievably cause a leak or weakness that is compounded by an excessive force placed on the base of the can by a spring. So still might be worth exploring spring strengths...}

[Icebreak]

Kevin -

So I don't need to send this? It's sitting in my office...was going to send after work.

- Jeff

[Topper]

Newbie, I was really hoping you would have a reaction, safely of course.
I noticed that your light only reached around 110 and mine hit 122 before it started dropping. Mine was outside in the midday sun. I wish I had thought to place another light near by just to see how hot it would get just sitting in the sun for an hour. Was your test inside or does anybody think it would matter?
I do not recall the spring on mine being extra stiff or not extra stiff. I did not crank the head down as far as it could be; it was where it needed to be for the tightest spot.Maybe on different lights that might meen cranking down further? I do not know.
Topper

[batterystation]

Kevin -

So I don't need to send this? It's sitting in my office...was going to send after work.

- Jeff

Yes, please do. I will send you a new one for it. I was just letting everyone know that Pelican is also sending one over. I would like to compare the two and see if there are any differences up real close.

[Icebreak]

PM6 sent priority box with confirmation number at 4:55 CST at a local post office.

[Art Vandelay]

Art. Check out post 509.

Icebreak, thanks for pointing me to post 509. In it, Newbie makes a good point about the thin membrane that seals the exit holes. He also has some great internal pictures of Duracell, Surefire and Energizer batteries.

However, I want to know whether Battery Station batteries have PTC (positive temperature coefficient) protection. If not, this would be a significant difference between Battery Station batteries and most of the others mentioned in this thread.

I think not having PTC could be an advantage in some high powered flashlights.

Does anybody know if BatteryStation batteries have PTC protection?

Thanks

[tvodrd]

Yes.

Larry

[Icebreak]

Art Vandelay -

Apologies for the less than complete answer. Really I was just looking at the photo. I've never seen PTC protection until I saw that picture.

Looks like tvodrd fielded the query for us definitively.

A little off topic but I applied for an internship with Vandelay Industries. They gave the job to another guy. I was pretty rattled about the decision. However, after a couple of sessions with Dr. Von Nostrum, I felt all better.

- Jeff

[milkyspit]

This post in this thread:
http://www.candlepowerforums.com/vb/...&postcount=165

combined with NewBie's description of seeing no signs of battery deformation, would suggest either that not all cell cans are the same, or not all PM6 springs are the same. Note that the picture of the deformed cell, as Milkyspit points out, came from a single-cell light (just goes to show that one can't be complacent even with a single-cell light).

So I'm wondering what the springs will be like in the old PM6's that Pelican are sending you, Kevin? It might be worth you assessing them when they arrive, and seeing if you feel any are particularly stiff?

I can state with some confidence that not all PM6 springs are the same! For starters, the super stiff springs I mentioned many posts above were found in the first batch of hard anodized PM6 LED lights made for Battery Station. All those lights had/have clicky tailcaps. I had one of these lights, as did Luxlover, and both of us independently crunched some cells... mine primary, his Li-ion rechargeables (which have stronger cans, if memory serves). The springs looked like they had been made from unusually thick metal... easily equal in thickness and stiffness to the metal used in Mag 2D tailsprings! Using my thumb the spring was notably difficult to compress. When tightening the tailcap all the way, I recall a very scratchy tactile sensation as the tailcap got tighter... this turned out to be caused by the spring pressure pushing the threads with tremendous pressure against the backside of the mating "sidewalls" (the thread trenches... don't know the technical term for these, sorry). I literally cut my spring in half to make things fit without the excessive cell deformation, and it's still plenty snug in there.

Luxlover eventually had his light replaced and told me that Pelican had switched to a less stiff tailspring... still too tight in his opinion, but apparently not as oppressively tight as the previous generation. I didn't examine the newer spring myself, so I'm going on Luxlover's description, though on the other hand I have no reason to doubt what he said.

Had Pelican changed springs yet again since then? Do any twisty tail lights, or any incandescent PM6 lights at all, for that matter, have the super stiff spring? I really can't say. In the narrow domain of the hard anodized LED PM6, though, I know for sure there were at least two types of spring: tight and tighter.

If I have a moment, I'll try to snap a couple photos of my tailcap... remember, though, that the spring has already been cut in half!

BTW, for what it's worth I have no plans to get rid of my PM6 lights (have three of them)... if there's a design flaw in there, it seems that the explosive result still needs some cofactor, such as a defective cell, condensation, pressure differential, or whatever... there are just too many PM6 lights out there that function flawlessly for their owners to believe this light is a death trap.

That's not to minimize this particular incident either! It's in all our interest to be safe, and the more info we can collectively gather, the better off we'll all be. All the folks with great ideas, endless testing, references to useful technical data, etc., have been greatly inspiring. Keep at it, guys!

[OutdoorIdiot]

Milkyspit,

Thank you. Detailed information doesn't get much better than that!

I'm sure photographs of your tailcap springs can only help matters, if you can find the time.

I also happen to agree with your ethical observations about the value of this thread, tempered with the need to not get carried away with assumptions about certain brands/models.

[NewBie]

Newbie, I was really hoping you would have a reaction, safely of course.
I noticed that your light only reached around 110 and mine hit 122 before it started dropping. Mine was outside in the midday sun. I wish I had thought to place another light near by just to see how hot it would get just sitting in the sun for an hour. Was your test inside or does anybody think it would matter?
I do not recall the spring on mine being extra stiff or not extra stiff. I did not crank the head down as far as it could be; it was where it needed to be for the tightest spot.Maybe on different lights that might meen cranking down further? I do not know.
Topper

My test was outside, inside a five gallon bucket, with a 1/2" thick polycarbonate clear plate over it, in the late afternoon into the night. I also just went out and removed the flashlight from yesterday, it has been on all the time, since the test. One of the cells was still a bit negative, climbing slowly to positive, the other was down to +1.595V.

As I understand it, Topper's test was conducted on *very* hot asphalt, with the sun blasting down on the asphalt and the light at the same time.

So in Topper's case, multiple problems were created that the cells had to deal with:
-Mis-matched cells (one purposefully discharged by 60% with 40% remaining)
-Extremely hot environment
--Light was already hot
--The black asphalt was *very* hot
--The air temperature was hot
--The hot sun was bearing down on the flashlight, heating the body further
-Battery deformation due to old style PM6 light springs.

I don't think any of the actual normal usage failure modes that folks- who have had their lights blow up, were under all these conditions simultaneously.

My makeshift PM6 tester:

[milkyspit]

Milkyspit,

Thank you. Detailed information doesn't get much better than that!

I'm sure photographs of your tailcap springs can only help matters, if you can find the time.

I also happen to agree with your ethical observations about the value of this thread, tempered with the need to not get carried away with assumptions about certain brands/models.

My mistake! The super stiff spring is at the HEAD end of the light. This is a technicality, though... it had the same effect as described in above posts.

Here are a couple photos...

My PM6 HA3 LED disassembled. Super stiff spring at the head end. Remember, I cut off AT LEAST HALF of it... taking another look while shooting these photos, I'm thinking I actually cut away more!


Closer look. Note how thick the wire is relative to the positive battery contact point, for example!


Not sure if this will help anyone, but hopefully the photos will give some idea of just how thick and heavy this spring was. Again, no intent to malign Pelican, Battery Station, Kevin, ...or anyone else! Just sharing.

[Icebreak]

Topper -

I got one of your data points wrong so (maybe trying to make up for that mistake) I'm posting some enviromental data points from you to gather that up a bit.

Newbie -

I guess it was concrete not asphalt.

Stuff Topper Typed:

"I ran one down in a single cell light and waited several hours so both cells started out at room temp."

"The PM6 was on the ground and about 6 feet from a white plastic fence."

"Mine was placed on a concrete driveway."

"I got tired of standing outside in the sun and heat."

"It gets pretty humid here…"

All -

I'm still leaning toward the idea that this event is a random cell that has a chemical or structural flaw with chances of 1 in 100,000 or greater of exploding under just the right circumstances.

However, the fact that Topper could replicate it using ZTS to get a 40%/100% 2XCR123 combination makes me think there can be some predictability. Whether other methods are better or not, that BS is using ZTS on all their CR123s is a fairly massive effort. Good on 'em.

It's also possible that the chances are greater than 1 in 100,000 that in flashlights because we simply don't hear about all the events. Joe outdoorsman finds a blown up CR123 driven flashlight in the back of his truck and it's likely he'll contact the company, get a nice explanation and a new light then go on about his business.

I continue to enjoy this community effort to indentify the culprit(s).

Side Note:

Many members have noted all the different chemical make ups of lithium batteries. I'm a little concerned that eventually the gov may insist on all lithium batteries being banned from air frieght. I can't remember which one it is but one of them has an active ingrediant that can eat through aluminum. It's the big one in a hard can. Well, I can see that but not all lithiums should be banned.

Remember Kevin at BS said we could turn in Lithium AAs and trade them out for BunnyBatteries? Well I've got them here, there and younder installed in various appropriate lights. I had 8 unused so I thought I'd send them back to try the bunnybats. The lady at the post office asked about them.

"Are these regular alkalines or something else?"

I replied, "No ma'am. Those are lithiums."

She stated. "We'll have to ship them ground."

I offered, "Would it simplify things if we subtracted the lithiums?"

"Why, yes." she said. "We can get this flashlight right out".

I felt eyes on me and noticed some strange looks from people in line next to me...like I was doing something dangerous.

I told a mutual flashlight buddy (CPFer) and they reminded me that lithiums were used in meth production...that's why I was getting the looks.

Of course I had no confidence in describing AA Li-FeS2 as being airship worthy. I'm thinking of making a major purchace of CR123s, CR2s, Ns, Li-FeS2s and RCR123s while I can.

[NewBie]

My mistake! The super stiff spring is at the HEAD end of the light. This is a technicality, though... it had the same effect as described in above posts.

Here are a couple photos...

My PM6 HA3 LED disassembled. Super stiff spring at the head end. Remember, I cut off AT LEAST HALF of it... taking another look while shooting these photos, I'm thinking I actually cut away more!


Closer look. Note how thick the wire is relative to the positive battery contact point, for example!


Not sure if this will help anyone, but hopefully the photos will give some idea of just how thick and heavy this spring was. Again, no intent to malign Pelican, Battery Station, Kevin, ...or anyone else! Just sharing.

Here is a monster spring wire diameter, but full length, does this put my PM6 at the right generation (it does have the focusable reflector)(I do know it is at least as old as 2003, or older...):



This one does look more like the pieces I have seen on other PM6 pictures of the explosion remants.

[PlayboyJoeShmoe]

Indeed that is an old style Lamp Module. On my old PM6 (and many others) a good portion of the spring is dremeld off and smoothed up. It lets the tailcap screw on farther without really squeezing the cells.

Reference Milkys post for detail.

[photonhoer]

LM - I want to join the chorus of well-wishers; your misfortune has uncovered a REAL need to look more closely at how and where we use these CR123 storage devices. Also, major kudos on your attitude of "let's find out" rather than "who can I sue"!!! I hope your complete recovery is on the near horizon.

Also, I want to echo OutsideIdiot: "But we need to focus on ideas, rather than jumping to poorly-founded conclusions about brands." There's a lot to be figured out here, and I offer my THANKS to all of the many people who are hard at work developing hypotheses and testing them.

One small effort at contribution: The heat variable seems an important one to pursue; it's hard to fathom this catastrophic outcome without it. Is the PM6 not noted for its illumination? ...and this correlated with current draw? Therefore might other lights with lesser current draw - and indeed the same light with a different light source - fail to replicate the explosive failure? Perhaps worth checking out the current draw of the older light module in comparison with newer PM6 modules, and with other 2x123 lights. If there is a significant difference, then testing with other than a very high draw module would of course contribute nothing to our understanding.

Finally, a question: what do we know about the behavior of the CR2 and the rechargable configurations of the Li-Ion batteries? Are they potentially just as vulnerable? Is the chemistry the same in the reg. CR2 as the C123? Are the rechargable versions different enough to side-line them from concern?

Thanks for listening,

John

[brightnorm]

Link to yet another explosion involving lithium primaries:

http://candlepowerforums.com/vb/showthread.php?t=122500

Brightnorm

[Hans]

Finally, a question: what do we know about the behavior of the CR2 and the rechargable configurations of the Li-Ion batteries? Are they potentially just as vulnerable? Is the chemistry the same in the reg. CR2 as the C123? Are the rechargable versions different enough to side-line them from concern?

There's a sticky on safety considerations with li-ion rechargeables. It makes sobering reading:

http://www.candlepowerforums.com/vb/...d.php?t=106242

That thread doesn't answer *all* the questions, but it answers quite a lot.

However, I think it's a good idea to keep the question of rechargeables separate from the present discussion about the safety of lithium primaries. Even though the probability of this sort of accident happening to anyone who sticks to the main safety measures mentioned here and in other threads here, such as

- using only well-matched batteries in multicell lights,
- using only batteries with a proven track record like Batterystation, Surefire, Duracell, Sanyo and so on,
- being careful with the batteries, that is not dropping them or damaging them in any way

is probably fairly low, I feel it's nevertheless essential that someone finds out which set of conditions is responsible for such accidents. I'm not so much worried about my personal safety, after all I know about the potential risks now and I decide whether I continue using lithium cells, but I am worried about the safety of others who may be affected should one of my lights blow up.

I also feel the likes of Surefire, Inova, Streamlight and so on must be interested in getting to the bottom of this. Should there be a general problem with lithium cells in high-draw lights, they may well have a nasty problem on their hands as soon as the first person gets killed. And without wanting to sound too pessimistic, given the dangers of lithium cells outlined in some of the recent posts that may well happen sometime.

Hans

[batterystation]

Does anybody know if BatteryStation batteries have PTC protection? Not that there is anything wrong with that.

Thanks

Yes they have PTC protection. They would never pass UL or any other tests without it. I just short circuited one here on my Fluke DMM and at about 7 amps it got hot and the current slowly dropped to under 1 amp until I let the cell cool off. In another thread I noted that out of 40 samples from different companies three years ago, 13 did not even have PTC protection. I would imagine a bunch still don't but all the good names do.

[MikeF]

Does anyone have email addresses for Paul Kim amd Dr. John Matthews they could PM to me?

[MikeF]

Double post, my first one ever I think.