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

[lexina]

In the other pelican thread, the batteries exploded WHEN THE LIGHT WAS OFF.

If I recollect correctly (have not gone back to check the thread), this was not conclusive. I think one of the theories put forward was that the user, after discovering that the light was very dim, might perhaps have inadvertently left the switch on before returning the light into the storage box.

[enLIGHTenment]

Just found this, glad youre OK Stephan. It could have been alot worse for sure, kinda makes me wanna only use singled celled lights....

Don't be sure they're safer.

http://candlepowerforums.com/vb/showthread.php?t=121303

The other day, a BatteryStation lithium AA exploded in a single cell device and set several fires.

[dyyys1]

While the fact that there have been two major battery explosions within days of one another is an incredible coincidence, we have to remember that it is just that: a coincidence. Considering the number of batteries used by the thousands of people who use CPF, there are very few accidents, explosions, or fires that occur. While these events are important to discuss in order to prevent them and to warn us to be careful with batteries, the risk you take in using these batteries is extremely low.
We all take risks daily. These risks are part of life. Think of driving, for example. Driving is very dangerous. How about stairs? Stairs have the potential to be deadly if you fall. Some things that we wish to do are risky. Does that mean we should never do them? Absolutely not! In many daily activities, the reward is worth the risks. Some risks are low enough that I am willing to accept them. If I made a list of all the risks I take every day, using batteries would be near the bottom.
I will say again that we should examine these stories to stay informed and prepared for the worst case scenario (the Boy Scout Motto is "Be Prepared") and to prevent their recurrence. However, this does not necessarily need to have a profound impact on the flashlights you are willing to use. These explosions we read about occur only in isolated, unique, hard-to-duplicate circumstances. Until a risk becomes greater than the reward, there is no reason to stop doing what you are doing. However, I understand if you are a safety freak (like my mother). If you are uncomfortable using a tool that is even slightly dangerous then don't use it.
Always make yourself aware of the risks and the rewards in whatever you do and make an intelligent choice based on the information you gather. Personally, I am going to be careful and treat these batteries with the respect they warrant, but I am still going to use them. I have a SFA2 coming in the mail, and I am very excited, not worried about batteries.

[SilverFox]

Hello Topper,

David Campen has repeatedly pointed out the shortcomings of the ZTS tester, so I just want to make sure I understand your results correctly.

You took two cells that tested at 100% on the ZTS, ran one until it was at 40%, then put the mismatched pair into your Pelican M6 and turned it on. It ended up blowing up. Is that correct?

This is a data point that illustrates that the ZTS may be able to sort cells by matching their test results. This should also increase the safety of using multi cell lights.

Please understand that I am not picking on David. I happen to agree with a lot of the points he brings forth, however I also believe that in spite of its shortcomings in determining the remaining capacity of a lithium primary cell, the ZTS tester can still be very useful for matching cells in a multi cell application.

Tom

[RA40]

My question now: at what differential between batteries may result in one of these reactions? 100:40-proven, 100:50, 100:60....

While it's dangerous, this type of testing can give us a guidline as to what levels may pose greater risk in use. I don't think it needs testing below 40% but if this might occur with a 100:80. Are there other testers than can give more precise reults other than the ZTS 20% increments?

[AlexGT]

Another BatteryStation lithium exploded! This time it's AA

http://candlepowerforums.com/vb/show...96#post1453696

Kevin you might want to go check it

AlexGT

[TENMMIKE]

My question now: at what differential between batteries may result in one of these reactions? 100:40-proven, 100:50, 100:60....

While it's dangerous, this type of testing can give us a guidline as to what levels may pose greater risk in use. I don't think it needs testing below 40% but if this might occur with a 100:80. Are there other testers than can give more precise reults other than the ZTS 20% increments?

the ansmann energy check goes at 10% incrementsand displays volts as well, it only available in europe unless you have a euro friend send you one

[Geologist]

the ansmann energy check goes at 10% incrementsand displays volts as well, it only available in europe unless you have a euro friend send you one

Is that for lithiums? If so, have you had direct experience? Being on the other side of the pond, I might have to pick one up...

Edit - just found the model from Ansmann - anyone know how it compares to the ZTS tester? Is the Ansmann just a glorified voltmeter? (no offense, but ZTS claims have always been much more, and the Ansmann is ~ half the cost).

Not trying to go too far OT, but for those of us with large stock piles of Lithiums, it is maybe a good idea to have these type of testers available at home...

[TENMMIKE]

this is MY understanding of things ... ansmann has in some way or form used the ZTS testing criteria or technology, ZTS is suing ansmann ( that's why no ansmann in USA), ansmann therefore is using basically if not the same load test that zts uses, but gives a higher resolution and added volt display, the lithium criteria is still a issue as we are unsure to how well the litihium testing is correct from zts factory relative to how we use the cells, this is to the best i know still in investigation between some of us that have the various said instruments, clear as mud huh?

Is that for lithiums? If so, have you had direct experience? Being on the other side of the pond, I might have to pick one up...

Edit - just found the model from Ansmann - anyone know how it compares to the ZTS tester? Is the Ansmann just a glorified voltmeter? (no offense, but ZTS claims have always been much more, and the Ansmann is ~ half the cost).

Not trying to go too far OT, but for those of us with large stock piles of Lithiums, it is maybe a good idea to have these type of testers available at home...

[Topper]

Hi Tom, that is correct. I ran one down in a single cell light and waited several hours so both cells started out at room temp. I did make sure to put the cell showing 40% in first (closest to the head) for what its worth. I was using a La Crosse Infrascan to check the temp (as an after thought). The light hit 122 then started to cool down and was dim so I figured no go on my test so I went back in the house for about 20 minutes. I was surprised when I came out to get it and realised I missed the part I wanted to see (from a distance not up close) now I am not in the least sure that anytime anyone
uses a 40/100 combo it will blow off the tailcap I am only sure it did this time.
If I had been holding it it would not have gotten as hot (I think) and if it did get too hot to hold I would have turned it off so perhaps it would have been fine. I am not about to toss all my 2,3 and 4 cell lights out but I will be careful.
Topper

Hello Topper,

David Campen has repeatedly pointed out the shortcomings of the ZTS tester, so I just want to make sure I understand your results correctly.

You took two cells that tested at 100% on the ZTS, ran one until it was at 40%, then put the mismatched pair into your Pelican M6 and turned it on. It ended up blowing up. Is that correct?

This is a data point that illustrates that the ZTS may be able to sort cells by matching their test results. This should also increase the safety of using multi cell lights.

Please understand that I am not picking on David. I happen to agree with a lot of the points he brings forth, however I also believe that in spite of its shortcomings in determining the remaining capacity of a lithium primary cell, the ZTS tester can still be very useful for matching cells in a multi cell application.

Tom

[batterystation]

I spent all day Saturday trying to blow up CR123A batteries of every brand in numerous lights with no luck. Only discovered my new Sanyo arrivals are now made in China too. Now I see after a weekend of trying this that yet another battery has erupted. Another UL rated product. I just don't get it. We never let anything out for sale until it passed the coveted test and still.
The AA project will now be scrapped. I am more comfortable believing the CR123A thing was a fluke given the huge numbers sold worldwide and the overall track record. I am not sure what has caused this problem. I am not sure I will ever know. The AA is too new. I was just trying to give the bunny some competition. Unbelievable............

[dyyys1]

Surely one Li AA explosion is not enough for you to scrap the whole project! Have you had other problems as well? If not, then please don't cancel the AA project. That other explosion could be a fluke just as easily as this one.

[Lunal_Tic]

The AA project will now be scrapped. I am more comfortable believing the CR123A thing was a fluke given the huge numbers sold worldwide and the overall track record. I am not sure what has caused this problem. I am not sure I will ever know. The AA is too new. I was just trying to give the bunny some competition. Unbelievable............

Would it not be possible that it was a mechanical failure caused by the product in which it was used? It would be a shame to bag an entire project without a bit more research.

Either way you decide, good luck. Keep sane.

-LT

[Knight Lights]

The AA issue is completely different, Kevin.

Please don't cancel it. That use in a charger is not typical or expected. While the charger may say it is OK to use with Lithium cells, it is completely different than typical uses, and should be viewed as such.

Bill

[David_Campen]

The AA project will now be scrapped. I am more comfortable believing the CR123A thing was a fluke given the huge numbers sold worldwide and the overall track record. I am not sure what has caused this problem. I am not sure I will ever know. The AA is too new. I was just trying to give the bunny some competition.

Sigh, sorry to hear that you are scrapping the Li-AA project. I have used your Li-AA cells in a number of devices at up to 1 amp continuous drain without any problems. This one reported case of a fire - who knows what sort of drain that battery was subjected to. It was nice to see Energizer having some competition.

[daloosh]

Topper, thanks for the test, it adds data to the pile. Maybe you can borrow a digital camera from a friend or neighbor to take a picture before you clean it up?

Kevin, thanks for staying on top of all this!

daloosh

[NickelPlate]

Let's face it. The problem IS with the Pelican M6.

Really? And what is the basis for your conclusion?

It is a dangerous, unsafe light.

Every light that uses CR123A cells is potentially dangerous and unsafe. It's up to the end user to decide whether the inherent risks are worth using it or not.

There can be no other explanation.

And what test data do you have to back up that statement?


Lots of CPFers here own and use a PM6 (myself included) with no problems. Until someone does some controlled experimentation that can produce repeatable results, all the qualitative speculation is just hearsay.

That said my condolences to LM and I'm glad you weren't seriously injured.

Regards,

Dave

[chevrofreak]

Two things I've noticed. It seems that all of the Pelican M6's that have had problems were the older models that had the focussing ability, and had the older momentary only switch.

[Spongebob]

Seems like the majority beleive this was the effect of miss matched batteries in a high drain light. Does this mean it would be less likely to happen in a low drain LED light even with miss matched batts?

If so the safest use for a CR123 would be in a low drain single cell LED light.

[bwaites]

chevro,

Interesting point. Is it possible the older tailcap has developed high resistance and is creating a problem that increases the battery temperature? This might predispose the venting circumstance, but that is just a "what if?" kind of idea.

It might also explain why I couldn't recreate it using a lamp that pulls even more current than the Pelican.

I don't see how the LA's can have any impact, since the only thing that changed was potting the bulb into the reflector. Same lamp, same spring, same reflector, same everything otherwise.

Bill

[greenLED]

Topper,
That's a great test you did (glad you're safe).

IIRC, BatteryStation is doing tests with 80% capacity cells. Maybe this state of charge is not low enough to trigger an event similart to what LunarModule had (which Topper seems to have replicated). I think this type of mismatch in charge is what testers should be focusing on, not relatively "fresh" cells.

If I had my Peli M6 I'd sacrifice it in the name of another data point. I'd use one fresh cell and one at around 40% charge (or completely depleted).

[chevrofreak]

chevro,

Interesting point. Is it possible the older tailcap has developed high resistance and is creating a problem that increases the battery temperature? This might predispose the venting circumstance, but that is just a "what if?" kind of idea.

It might also explain why I couldn't recreate it using a lamp that pulls even more current than the Pelican.

I don't see how the LA's can have any impact, since the only thing that changed was potting the bulb into the reflector. Same lamp, same spring, same reflector, same everything otherwise.

Bill

The momentary switch on the older M6's was really very poor. After a while mine stopped completely resetting to its off position, the parts were binding inside.

Pelican made it far more complicated than they needed to, with two springs and two plungers.

Pictures here http://lights.chevrofreak.com/roger/Pelican_M6_switch/

[Icebreak]

I don't think Topper said if his was focusable or fixed focus.

[chevrofreak]

I don't think Topper said if his was focusable or fixed focus.

He was using cells of vastly different capacity. That alone would cause what he witnessed.

[batterystation]

The AA issue is completely different, Kevin.

Please don't cancel it. That use in a charger is not typical or expected. While the charger may say it is OK to use with Lithium cells, it is completely different than typical uses, and should be viewed as such.

Bill

There have been supply issues too. We have been fighting a legal battle over these just to get them here. That is what makes this worse. Believe me, this was not a rash decision. I really want to sleep at night. I am confident with all the numbers and evidence that the CR123A is inherently safe. Obviously something went wrong. I don't dispute that. It was horrible but I think it was a fluke combination of a bunch of things that just went wrong somehow. We are trying to duplicate this. The L-AA should NOT have done what it did. We clamp them in vices here repeatedly and have not had this problem. UL does all sorts of tests. I just feel better about the CR123A due to the history, numbers, and track record. Several million sold since the upgrade for load current. Please understand, I am NOT aware of any known defect of the AA but am being overly cautious. I do have Energizers at near cost on the CPF page now for anyone that needs them.

[dyyys1]

He was using cells of vastly different capacity. That alone would cause what he witnessed.

You may be right, but the issue has not been conclusively resolved, as Kevin from batterystation has told us. However, the use of different cells seems to make a big difference, at least in the Pelican M6.

[batterystation]

Topper,
That's a great test you did (glad you're safe).

IIRC, BatteryStation is doing tests with 80% capacity cells. Maybe this state of charge is not low enough to trigger an event similart to what LunarModule had (which Topper seems to have replicated). I think this type of mismatch in charge is what testers should be focusing on, not relatively "fresh" cells.

If I had my Peli M6 I'd sacrifice it in the name of another data point. I'd use one fresh cell and one at around 40% charge (or completely depleted).

I am also attempting this with ZERO and 100% charge batteries too. Thus far, nothing but some heat and dead batteries.

[dyyys1]

He was using cells of vastly different capacity. That alone would cause what he witnessed.

You may be right, but the issue has not been conclusively resolved, as Kevin from batterystation has told us. However, the use of different cells seems to make a big difference, at least in the Pelican M6.

[Codeman]

Regardless of all else, I'll feel better when I place my next order for 123's from BatteryStation, knowing how hard you've work to nail this issue down, Kevin.

And a belated thanks to Lunarmodule for posting about the accident. I truly regret that this happened to you, but this thread will definitely make me more cautious. A small consolation, I know, but I am grateful nonetheless.

[Topper]

Sorry about that. It was an older model that was focusable.
Topper