At what point do 10440's become dangerous

[LightScene]

I'm planning to use 10440's in a Fenix LD01. Since it has a boost circuit, it can run the 10440 down pretty low. There are 2 issues: 1) the battery may lose some capacity by being over discharged 2) the battery may develop an internal 'short' which causes it to heat up during charging and possibly burn.

It seems that even if the battery gets run down to a low voltage, it may recover to 3V or higher just by being allowed to rest for a few hours.

My question: is it possible to know ahead of time (by checking the voltage, or some other way) that the 10440 is in a dangerous state before I put it in the charger, or do I have to monitor the battery while it's charging to make sure it doesn't overheat?

 

[Fallingwater]

You seem to be a bit confused in relation to lithium-ion cells. You should probably read this; it should answer many questions.

 

[LightScene]

You seem to be a bit confused in relation to lithium-ion cells. You should probably read this; it should answer many questions.

I'm very familiar with RCR123s. Been using them for 4 or 5 years. I've recently done a lot of reading through 10440 and Li-Ion threads in this forum and I'm trying to distill and finalize what I've learned. If you think I don't understand something then tell me what it is.

 

[Fallingwater]

Fair enough.

First up: from what I'm reading, the LD01 will overheat if used on a 10440 for significant lengths of time. I presume you were aware of this, but I still think you should reconsider. LEDs don't like overheating.

Since it has a boost circuit, it can run the 10440 down pretty low.

It can run the 10440 down, but it's not because it has a boost circuit. Any driver can run a LiIon cell down if protection is not built-in in either the driver or the cell. Lights without a driver but with a simple resistor can be counted on to do that as well.

There are 2 issues: 1) the battery may lose some capacity by being over discharged

Overdischarging LiIon cells tends to make them useless, not just to make them last less. You can sometimes recover them, but if you want to be absolutely safe you have to toss every LiIon cell that's gone beyond 3V and doesn't recover if left alone.

It seems that even if the battery gets run down to a low voltage, it may recover to 3V or higher just by being allowed to rest for a few hours.

It widely depends on how low it was allowed to go. If it goes to (say) 2.7V under load, then it's almost certain to recover as soon as the load is removed. Any lower than that and it'll have more trouble. In my experience cells that don't recover in at most half an hour do not ever recover fully, and do not recover at all just by themselves.

My question: is it possible to know ahead of time (by checking the voltage, or some other way) that the 10440 is in a dangerous state before I put it in the charger, or do I have to monitor the battery while it's charging to make sure it doesn't overheat?

If it's above 3 volts it can be charged with peace of mind (you still shouldn't go to sleep). If it's below 3V but above 2.7 or thereabouts it can still be recharged, but you probably should stay in the same room just to be sure.
If it's below 2.7V you *might* be able to recover it, but you'd do well to use a bench power supply with current limiting or a very gentle dedicated charger to bring it back to three volts before you try. Note that this is commonly considered dangerous, and you shouldn't do it unless you a) know exactly what you're doing, and b) are prepared to deal with a lithium fire in such a way that it doesn't damage anything.

Or at least, this is how I do it.

 

[2xTrinity]

Fair enough.

First up: from what I'm reading, the LD01 will overheat if used on a 10440 for significant lengths of time. I presume you were aware of this, but I still think you should reconsider. LEDs don't like overheating.


It can run the 10440 down, but it's not because it has a boost circuit. Any driver can run a LiIon cell down if protection is not built-in in either the driver or the cell. Lights without a driver but with a simple resistor can be counted on to do that as well.

A driver that is strictly buck-regulated, or direct driven, will not discharge a cell lower than the turn-on voltage of the LED. For a white LED that means that even if you left a LiIon connected to the LED indefinitely it wouldn't fall below about 2.8V. The light will also dim considerably from the time battery voltage under load drops below about 3.5V, and would probably take hours to sag all the way to 2.8V as the current at that level will be very, very low.

With a boost driver however, without a protection mechanism, it will simply draw as much current as it needs until the battery itself can no longer sustain the load (long after the point that the cell chemistry will sustain damage). The only time a pure boost driver will fall out of regulation is if it is supplied an input voltage higher than the desired output voltage, or it is forcibly stopped by a low-voltage protection.


Anyway, here's my advice on how to run a LD01 on 10440s:

Buy a LF2x

 

[LightScene]

Anyway, here's my advice on how to run a LD01 on 10440s:

Buy a LF2x

Thanks. I was with you all the way, until that last sentence.

 

[LightScene]

@ Fallingwater: Thanks for your help.

Summarizing what you said regarding dangers: after the battery has rested for a while, if the voltage is below 3 volts, then it requires some attention while charging. If it's below 2.7 volts then there is a real possibility of a fire so it needs to be monitored to make sure it doesn't heat up.

I should also add that I'm planning to run the LD01 almost entirely in Medium mode, which is the mode it turns on in. This will provide around 80 lumens. With the use of 10440s this light becomes a simple on-off light for most uses and I don't think it will overheat at that level, since it will run for nearly an hour to 50% brightness.

 

[Fallingwater]

A driver that is strictly buck-regulated, or direct driven, will not discharge a cell lower than the turn-on voltage of the LED.

True that, but I haven't seen many flashlights that run on a single LiIon cell and have strictly a buck driver... or maybe I just haven't looked hard enough. :candle:

Summarizing what you said regarding dangers: after the battery has rested for a while, if the voltage is below 3 volts, then it requires some attention while charging. If it's below 2.7 volts then there is a real possibility of a fire so it needs to be monitored to make sure it doesn't heat up.

Yup. Some say that anything below 3 volts is enough of a reason to immediately trash the cell, but I tend to be a bit less drastic.

I should also add that I'm planning to run the LD01 almost entirely in Medium mode, which is the mode it turns on in. This will provide around 80 lumens. With the use of 10440s this light becomes a simple on-off light for most uses and I don't think it will overheat at that level, since it will run for nearly an hour to 50% brightness.

You can't do that, I'm afraid. Running on a 10440 the LD01 will lose regulation and all modes and run in full-power turbo mode all the time.
http://www.candlepowerforums.com/vb/showpost.php?p=2687303&postcount=5

Edit: I'm reading another thread right now, and it seems it retains the modes. I can't quite figure out who's right, so I'm guessing we have to wait for someone who has more experience with the LD01...

 

[moldyoldy]

<snip>

You can't do that, I'm afraid. Running on a 10440 the LD01 will lose regulation and all modes and run in full-power turbo mode all the time.
http://www.candlepowerforums.com/vb/showpost.php?p=2687303&postcount=5

Edit: I'm reading another thread right now, and it seems it retains the modes. I can't quite figure out who's right, so I'm guessing we have to wait for someone who has more experience with the LD01...

Your edit was correct. I have an LD01 and 10440 cells. All modes are retained regardless of 10440, Eveready Lithium L92-FR03, or Ni-MH AAA cells. That is also true for it's predecessor the L0D-Q4 or whatever LED happens to be inside. I have used 10440 for years in the L0D etc series and lost only one L0D (original).

FWIW, the L1D-Q5 is the Fenix light that loses modes and stays on turbo power most of the run-time until the Li-Ion 14500 AA cell discharges sufficiently to come w/in range of the regulation circuitry. For me, none of the L1D-Q5 units have failed yet, including one I sold to a janitor who uses it all the time with a 14500 (not really recommended though).

 

[LightScene]

Don't worry. The LD01 is designed to run on 10440s and all modes function correctly. People have been using 10440s in Fenix L0Ds for a couple of years. There is a lot of misinformation and disinformation disseminated on CPF about this topic. That's why I've been spending time to figure out what the facts are.

 

[Fallingwater]

Sorry I haven't been able to help you more, I don't particularly appreciate high-power AAA lights and so haven't taken the trouble to get information on the LD01 myself.

 

[LightScene]

You've helped me a lot. I appreciate it.

 

[2xTrinity]

Regarding whether the LD01 retains the modes:

The LD01 does not use variable-current dimming. It uses a boost driver with fixed output current*. Dimming is accomplished using a separate low-frequency PWM circuit that opens and closes the circuit using a fixed duty cycle. IIRC this is somtehing like 30%/10%/100%.

When running on LiIon cell, the boost driver DOES go out of regulation -- and the emitter is directly driven from the LiIon cell. However the PWM circuit, which is separate, still works. So, for example low will still be on 10% of the time compared to maximum setting. The "levels" are preserved.

Lights that use variable output current, such as the Fenix L1x series, DO lose modes when a 14500 LiIon is used. This is because the dimming is accomplished by the boost driver changing its output current. The boost driver once again is completely bypassed when the LiIon cell is used, and since that also manages dimming in a constant-current light, levels are lost. Familiarity with lights like this is likely the source of the confusion.



* The driver has close to fixed output current, but current actually does depend partially on the voltage of the cell. Maximum output will dip gradually as a cell wears down. This is why this light is often considered "semi-regulated", as this effect is still minor compared to a true direct drive light, or a light using a "dumb booster" (simply 2x multiplication of the cell voltage).

 

[Fallingwater]

Thanks for clearing that up. I guess this is a point in favour for PWM lights in the eternal PWM-vs-constant-current struggle...

 

[bigbear]

When running on LiIon cell, the boost driver DOES go out of regulation -- and the emitter is directly driven from the LiIon cell. However the PWM circuit, which is separate, still works. So, for example low will still be on 10% of the time compared to maximum setting. The "levels" are preserved.

hence in the context of a LD01 w/ 10440, i would read this as the boost driver will _always_ be out of regulation as the voltage will always be > 1.5V.

would this mean that there would be a very diminished chance the cell will be over-discharged because of the boost driver?

(of course, it will still be over-discharged if direct-driven to < 3V)

 

[LightScene]

would this mean that there would be a very diminished chance the cell will be over-discharged because of the boost driver?

I think your question was answered in post #5
"A driver that is strictly buck-regulated, or direct driven, will not discharge a cell lower than the turn-on voltage of the LED. For a white LED that means that even if you left a LiIon connected to the LED indefinitely it wouldn't fall below about 2.8V. The light will also dim considerably from the time battery voltage under load drops below about 3.5V, and would probably take hours to sag all the way to 2.8V as the current at that level will be very, very low."
It's the boost regulator that keeps supplying voltage above the turn-on voltage even when the battery voltage is too low to turn the LED on.