Hi,
Found this thread. I think I want to try this mod. The only snag is that there doesn’t seem to be any source for 10mm LEDs… anywhere.
Does anyone know where one can acquire a 10mm star LED?
Thanks!
3535:
https://mtnelectronics.com/index.php?route=product/product&product_id=436
5050:
https://mtnelectronics.com/index.php?route=product/product&product_id=433
1 Thank
1 Thank
Wow, thanks!!
Any advice for what emitter to use? I’m looking for something low-powered that can directly run off a 10440 and is maybe a bit brighter than my current EDC light (Maglite Solitaire LED)
I’m not familiar with your light nor 10440 batteries but a 10440 isn’t powerful so I’m thinking any 3535 or 5050 emitter would handle a 10440 easily.
It would come down to what you’re looking for.
I’m a fan of the Nichia 519A 5000K, to me they are pure white, bright and High CRI.
1 Thank
Seems lots of people like the 519.
I’m not super familiar with modding flashlights.
Don’t most LEDs need some sort of driver circuit? Or can they just be powered straight off a battery?
Maglite claims 47 lumens of output for the Solitaire LED.
Also, should I be concerned about thermals if I use a 519A? It’s going to be mounted without any sort of heatsink
Direct drive will be bright/high powered in all instances.
Have you considered a Skilhunt E3A? Quite good value for money, single mode, 100lm twisty light runs on Alkaline (“alkaleaks”) or NiMh; no “maglite lantern functionality” if that’s important to you though.
Yes if you want regulation or modes. “DD” in this context is direct drive, it works because the resistance in the wires and cell is sufficient to bring the voltage low enough that the LED can handle the current that passes through.
However, if you have a cell with high current capability (i.e, can “support” high voltage while pushing large current), low vF LED (needs “more” voltage drop to prevent it burning up), or even low resistance wires (not enough voltage drop, burns the LED), then direct drive doesn’t work.
1 Thank
not to mention they appear to be rather accomodating to direct drive LiIon’s…I run one in my FrazLabs light 
1 Thank
whatever you mount it to will essentially be the “heatsink”. You could try to wire a resistor to the circuit to reduce the output a little bit
OK
So, the current plan is to use a Nicia 519a
My understanding is that (Battery Voltage - Target Voltage)/(Target Amperage) = Pull-Down Resistor Ohms
So to achieve 25% light output, I’d have a 2.75v target voltage and a 0.25a target amperage, so I’d only need a 4 ohm resistor.
Any issues with my calculation?
Is 4.2v going to burn up a 519A ? Isn’t 2.5volts a dead 10440 lithium ion cell asking to be recharged .
How many amps can a efest 10440 discharge in direct drive ? Isn’t a 519A direct driven going to put out 600+ lumens , depending on the battery .?
2 Thanks
Theoretically, correct. However, I believe you actually CAN direct drive a Nichia on a LiIon…I would look at the 519A data sheet, ballpark the lumens you want, then sub in that respective amperage. So for 90lm, roughly your 0.25A, (4.2V-3V)/0.25A= 1.2/0.25A or ~5ohms. If we go by nominal voltages, then 0.7V×4, or ~3ohms…but I’d err on the side of caution and use the higher value as it represents a max case scenario.
**At 2.75V target, that’s basically the lowest voltage the LED turns on, most cells with decent protection circuits should cut out around that value, if not a bit higher. If you’re using koef3’s 519A graph as a resource, the two splines exclusively represent either lumens, or voltage. I believe that if you draw a vertical line crossing the two splines, you can estimate how much power is required for the corresponding lumen output.
The cells I plan on using can only output 4.2 volts for about 2 seconds. They then drop to 3.9 volts, and voltage declines linearly from there. Thus, using nominal voltage is close enough.
Another reason for me to not direct drive is because I plan on using unprotected cells. My understanding is that the LED should cut out at around 2.6v, so with the added voltage drop of the resistor, the LED should give up well before the battery is damaged.
For some reason, I’m convinced that the LED will still be drawing power while apparently off in such a config, because it’s still a direct drive…
**Such a circuit is essentially a potentialized tank (pressurized or gravity) that only has a flow valve with no off/cutoff setting. Imagine that it’s passively emptying while turning a turbine or dynamo with a specific resistance to power an LED. As it gets near empty, the current might drop, but the tank is still dispensing whatever droplets it has left at an insufficient pressure to power the LED.
It shouldn’t draw power when turned off - the switch physically separates the negative terminal.
There’s tons of small leds you can run off a 10440.
Luminus SST20
Luminus SFT-12R
Luminus SFT-25R
Cree xp-g4
Osram W1.1
Nichia 519 (although for efficiency there’s better choices)
The 10440 battery is good for low-drain, and can feasibly provide up to 5 watts with good output.
1 Thank
Just for reference.
A Maglite Solitaire cannot dissipate 5 watts of heat.
I am hesitant to put anything that draws more than 1w of heat in it.
I wouldn’t either
1 Thank