http://www.candlepowerforums.com/vb/showthread.php?t=287906
EPIC. (While not exactly a budget Light)
WOW!!! That's pretty much all I can say. WOW!!! 
Certainly isn't a budget light - that is $250 worth of emitters!
More like completely insane!
I hope those cells are in parallel or he's just built a stupidly expensive pipe bomb. An SST-90 needs 12 amps for peak output, so 8 of them are going to need 96A! That is pushing even IMR cells too hard - 10A is their maximum so 8 of them in parallel is 80A at most. An SST-90 is a 30 watt device so 8of them make a 240W device. Bad idea!!!!!!!!!!
He says he's driving them at 9A which is not going to give peak performance - my single SST-90 runs at 12A and gives 1500 lumens, so 12,000 lumens is the absolute maximum for that configuration.
I suspect that is in a 3S2P configuration so he's pulling more than double what is safe from IMR cells. There is a thread somewhere on cpf of an explosion of a similar (but incandescent) 250W device with IMR cells in it.
I think he's built a device that could kill him. At best, it is going to quickly destroy a lot of expensive cells. At worst it is going to kill someone. A safe battery pack for a 240W device is very, very large. And even more expensive.
Should I want to build bombs, I could get a lot more explosion for a lot less money. And probably more light.
For a few milliseconds anyway.
From the pictures and description I get 8S emitters powered by 8S batteries in DD and <13A draw for safe chemistry cells maxing out at 16A continuous. Worst thing that could happen is ruining a lot of $$$ or perhaps, though much more unlikely, getting a short circuit that turns the entire light into a puddle of molten metal and sets something expensive on fire. His current pack should be good for up to 480W (8 cells, 3.7V nominal 16A (10C) max).
"LiMn" IMR cells does not have the same tendency of going kaboom like regular "LiCo" Li-Ion cells does, they may vent but will not turn into rockets/firecrackers/pipe bomb (or at least that is what I've heard).
18k LED lumens initially (my guess is 12-14k OTF) , dropping as the batteries sag and the emitters heats up.
BTW I'm absolutely jealous, wish I had the budget for building a light such as that one even though it is a really unpractical light : )
Lights like this really need a power tool type pack (ie dewalt). They're not that expensive (compared to that light at least).
"LiMn" IMR cells does not have the same tendency of going kaboom like regular "LiCo" Li-Ion cells does, they may vent but will not turn into rockets/firecrackers/pipe bomb (or at least that is what I've heard).
Look up the thread don mentioned. Venting in that sealed light caused a significant explosion.
On the lights themselves, he seems to be draw 13amps (see meter image), with all the batteries in series (and all the led's in series).
You mean the TK Monster thread? I had forgotten about that one. Goes to show that one should never mix good and damaged cells or even "safe" cells becomes unsafe. A pressure release valve or a burst disk integrated in the battery tube should help avoid turning the light into a rocket, shouldn't it?
You are right Xrayboy this is not a budget light unless you have a very big budget, and
I dont. Would be nice to have though. Im sure the neighbors would be very "happy" if
I used it in the garden.
And the TK Monster cells were Sony's IMR cells. They are not immune to explosion though it is a lot less common. As far as I remember the device did have a blowout vent but it wasn't close to adequate. Better a rocket than a bomb though.
AW specs his IMR cells for 16A which is within the limit but those cells had better be properly balanced every time or bad things will ensue.
I gather from the TK Monster thread that it did not have an overpressure vent and the light in question failed at its weakest point. When the cells vented the pressure inside the light built until threaded section between the battery tube and the battery tube extension burst and the one part of the flashlight shot off in one direction and the other part shot off in the other direction. A pressure vessel bursting can be deadly, a pressure relief valve would probably have turned the light into a spinning rocket but would have prevented it from bursting.
I'm thinking of drilling small holes in my multi cell lights and filling the holes with wax or some soft silicone. I think that it would be enough to prevent flashlight bursting events while not compromising the moisture seals, any ideas?
If it were not for their limited capacity A123 systems LiFePO4 cells looks very promising as a safe alternative, although they can still vent in extreme cases they won't vent with flame nor produce overly toxic gases upon venting.
If the area where you drill is flat, simplest way is just put high quality weather resistant tape over it. Or if you can buy a silicon plug of some sort just drill a hole a tiny bit smaller than it.
Simple, elegant and functional, I like it! Thank you : )
If it were not for their limited capacity A123 systems LiFePO4 cells looks very promising as a safe alternative
The problem I have with those is that when you use them in multi-cell applications (like 2x 123 in 18650 light, or 2x 14500 in L2r), there is essentially no over-discharge protection for the cells, which is fine if you're very diligent about charging, but I'm not.
Also, in single cell configs like AA lights which can use 14500, boost drivers I think will run them until they die.
Are there undervoltage protection in chinese lights designed for single cell lithiums?
Dunno, but if they are, then the undervoltage level will probably be set for li-ion and not lifepo4 since lifepo4 isn't as commonly used as li-ion.
The only one I know of that does is the programmable Akoray K-106. And in my experience undervoltage protection on most cells does not cut in soon enough - it should probably do so at 3V but I have discharged cells to below 2V that claimed to be protected.