http://www.candlepowerforums.com/vb/showthread.php?t=260499
Specifications:
1000Lumens at 150mA
1500Lumens at 250mA
This is true? Would be suitable for flashlights?
You forgot to add 25v @ 250ma (per string)
That is more reasonable.
I understand the 25v..... but "per string" what means?
Someone on cpf estimated the efficiency on par with an r5 emitter. With that info i really don't get the amount of excitement focused on that product.
I think the "per string" thing is that the dies internal to the thing can be addressed in groups - probably of 6-8 because that would give a forward voltage in the 25V region.
I'm not at all clear from the datasheet, but there appear to be around 24 dies in the thing, at a guess wired up in 4 strings of 6 dies - a bit like a giant SSC P7 where the four dies can be powered in series or parallel.
Can't see it being useful in a flashlight, the power requirements are pushing what is sensible with batteries, this is a 20W device.
People complain about the "big" die of the XP-G LEDs at what 3 square millimetres, this thing has a "die" of 154 square millimetres. You'd need an enormous reflector to get a torch beam out of this thing.
Not to mention the trouble getting rid of the heat in a reasonable fashion... That would be my first barrier to evercome. two li-ion 32xxx cells might power that thing if some form of voltage boosting is added. Not entirely sure if it is a correct approach but i would try that instead a load of 18650 combined in a fantastic bin laden worhty dirty bomb.
The 6 die Osram Ostar LED has similar voltage requirements and there are quite a few custom devices sporting them so it must be feasible. The heat needn't be too big an issue at least for relatively short runtimes - my Varapower 2000 is a 30W device and I get twenty minutes on full power out of a fresh charge on its cells - which is about when it gets too hot for its own good, so extending the runtime would involve more expense and complexity. But it does have a less than 4 volt forward voltage.
No doubt someone will mount one of these to a big lump of copper and bung it in some sort of Maglite. A 6d running 6 32600 cells might let you get away with direct drive and only a small resistance since it is only pulling an amp at full power. You probably need 7 cells to get any runtime out of it though.
7 series lithium cells is not a configuration for the timid. Or the sane. At least the current draws aren't huge so it might be survivable in the short term.
A 21 cell serial NiMH pack might be interesting. 21 Sub Cs (22.2mm dia - 44mm long)would give you the voltage in a manageable volume and give you 3-4 hours runtime. But that isn't going to fit into a Mag mod. And it is still going to be a hugely floody device unless your reflector is 60cm across.
NiMH A cells (17 x 50mm) are easy enough to get and it should be just possible to cram 21 of those in series into a 5 or 6D Mag body. They give 2200mAh in theory so you should get 2 hours out of them.
That would look more of a weapon than flashlight. With Li-ion cells it could easily be a weapon but with innacurate explosion time. :)
With the weight of NiMH cells needed it's make a pretty good club.
I guess, but try to find a suitable strap-on-belt holster for that. :D
An idea if some chinese manufacturer would do that is to put the name: UltraBaton "flashlight" holser with motorized exoskeletal support mechanism for easier carrying.
The worst thing in the world then would be to charge the nimh cells by http://fr.digikey.com/1/1/725780-battery-charger-usb-2-nimh-in3006cg.html
Runtime 2h, charging time 2-3 weeks.
Pls ban that spammer
Oh no!!! again!!
It's actually pretty clear from the datasheet that there are three strings (the package has three cathode pads and three anode pads, and 3 * 0.25A * 26.5Vf[typical] = 19.9W[typical]).