Sudden Cell Death & LEDs

The sudden death of CR123 cells seems to be a fairly common event with high drain, high output, incandescent flashlights. I haven't read about this happening with bright LED lights, yet.

Lights like the L4, PM6 and a few others pull as much current from the cells as their incan counterparts. When used for an extended time, so that the cells heat up, has anyone experienced the "sudden death" syndrome that's common with incans?

The failure mode I'm refering to is when the light seems to be working normally but won't turn on again once it's switched off. This is a different event than normal, rapid, dimming at the, expected, end of the cell's life.

I've only noticed this failure to turn on again, after the cells got quite warm from an extended use. It seems, at this point, to be the cell heating that causes the sudden death.

What has your experience been and do you have a different theory? Thanks for any replies.

Same experience here with higher drain LED lights like the L4 on multiple occasions. It is only one of the two cells which is dead, the other usually can work normally for another significant amount of time.
No idea about the theory here /ubbthreads/images/graemlins/icon3.gif
bernhard

Incadescents tend to draw more current as the cells start to drop voltage.

...but assuming the light wasn't dimming when first turned off... dunno.

With high-draw lights, perhaps the thermal protection in one of the cells kicks in when switched off.

[ QUOTE ]
idleprocess said:
Incadescents tend to draw more current as the cells start to drop voltage.


[/ QUOTE ]

I'm pretty sure that it's the reverse. Incandescents have a fixed resistance, so as the battery voltage drops so does the current. That's Ohms Law. The resistance drops when the filament is cold, but that is before it starts glowing.

Regulated lights are the opposite. As the input voltage drops the converter has to draw more amps to create the same output power. Again, Ohms Law. At some point the battery can not supply the required current or the the voltage falls below the level that the converter needs and the output decreases.

At a certain voltage the LED just goes out, even though it has 3/4 of it's normal voltage at the LED. This is tha threashold voltage (IIRC) and would typically be around 2 volts (I think) for a Luxeon.

Converters need a certain voltage in order to start up after being off. That's the start-up voltage and it might be higher than the voltage at which it quits working altogether.

Sudden cell death in lights with converters may be due to the battery voltage falling below the startup voltage, even though the voltage was high enough to keep it running once it was started. That would result in a light that worked last night, but is dead today. This should be more prevelant in lights that use a single cell that are already close to the minimum requirements for the converter.

Daniel

Daniel, the sudden cell-death syndrome has the unique characteristic that in a 2 or more cell light, one of the cells would be dead (usually totally dead, like 2V or less) while the others still have plenty of power left.

Lately, some folks have been taking note of which cell it is that turns up dead and it seems to invariably be the one closer to the lamp array.

Also, the tendancy with the lights where sudden cell-death is experienced, even those with good regulated converters, is that the light output will dim as the cells die. Whereas the experience with sudden cell death is that the lights appear to be working strong without any evidence of dimming. Though once the light has cooled down, it will no longer work.

On the first point alone, your explaination is unable to cover. Why is only one cell dead while the others, which have been subjected to the same rate of drain, are nowhere near as spent? Why does it always seem to be the one closest the lamp assembly?

Funny, I thought it was the one nearer the tail, unless my monday-dazed brain has misremembered somehow. I should go dig out the thread. The cells aren't quite a case of sudden death, rather they're the result of it running down.

Rotating the cells helps reduce the difference between the two cells.

[ QUOTE ]
idleprocess said:
Incadescents tend to draw more current as the cells start to drop voltage.

[/ QUOTE ]
[ QUOTE ]
gadget_lover said:
I'm pretty sure that it's the reverse. Incandescents have a fixed resistance,

[/ QUOTE ]
The actual situation is, that an incandescent bulb is a little bit kind of a constant current device.
Actually the current changes much less than the voltage. There are nice diagrams available in the web (don't have any of the urls at hand), but it was something in between that the current is changing at square root or 3rd root of the voltage change. Different bulbs are different, of course.
What may kill cells (especially rechargeables) is the inrush current at start. This one could be 7 to 14 times the nominal current.
As I wrote somewhere else, I do have a LiIon cell which is good for 6A but must not be used with incandescents of more than 0.85A.

To stay a little bit on topic in the LED forum: LEDs fortunately behave a little bit more polite.

Based on steelwolf's comments, there is only one thing that comes to mind, and it's pure conjecture.

The Duracell site has technical spec sheets. The MSD (safety sheet) shows the warning that it may degrade with heat. The spec sheet for the 123 shows a designed operating temperature range of -4 F to 167 F.

Maybe the chemical composition is breaking down from the heat, and forming different compounds as it cools.


Daniel

read my post "MICRA DEFECT????"
sounds very similar...it might be one & the same thing.
though i call it "MOMENTARY DEATH".

I'll betcha you are right, Gadget Lover. The core temperature could get over 167F during extended runs in some lights that have more insulated bodies. My rubber covered Scorpion had sudden cell failure. I wonder how thermally conductive the G2's metal sleeve/Nitrolon jacket is?

I had the same thing in lights that weren't running hot at all. At least not terribly hot.
bernhard.

P.S.: in the back of my head I have a faint memory of someone stating that the internal resistance of lithium cells actually DEcreases when they get hot /ubbthreads/images/graemlins/icon3.gif ... which would explain a sudden death after cooldown. Although this sounds very strange to me now, don't laugh /ubbthreads/images/graemlins/icon15.gif

My friend had the same thing happen with his new SF 6P while out on the trail. He ran it steadily for about 15 minutes, switched it off for a moment, and it wouldn't turn back on. He took it apart, fiddled with it, put it back together...nothing. Half an hour later, after it had cooled down, he turned it back on and it worked fine. He said it had gotten quite warm while continuously running. This was nearly three weeks ago, and I think he's still using the original batteries.

- Tim

Wow, amazing, it appears that CR123's have several failure modes. The very first time I used 123's this happened to me so I always carry spares.