That’s what I was implying though, without some factual input it’s just not possible to decipher what’s going on there. They may have creatively implemented a plastic subdivider for the fan compartment that does indeed shift air across. I know Nitecore has done some amazing plastic dividers in their lights that work but seem to have made their lights overly complicated. Something like that here could, hypothetically, create a pathway for the air movement. Won’t know until we see one I guess.
Are we talking about 360W here? 4416 lumens to 12 emitters, what’s that, 5A x 6V? 30W each x 12? Powered by 8 20700 cells, that’s a heck of a load on cells! And to cool that, we can’t even think about a puny little computer cooling fan pulling tiny current at 12V, no, that’s not gonna get the job done here! We need a serious fan! So yeah, the fan will indeed cut into the battery pack, as if the emitters weren’t already kicking it’s arse!
360W is nothing for 8x20700s. If the cells are rated for 30A, being Sanyo NCR20700A/C, running at 25A, that would mean 3,5Vx25Ax8 cells = 700W max draw pack.
The main problem as you said is power dissipation.
I think they should have just taken a Threadripper sized air cooler with a massive head and a 120mm fan. Would that be overkill? No, not at all. Although I don’t think anybody would like a 1,3kg heatsink/3 pound heatsink, but in this light, that would be liked.
What is the capacity of a high discharge 20700 cell? 3000mAh? You see the problem? If those 8 cells are in 2S4P, which they likely are, you effectively have 4 8.4V batteries at 3000mAh to run the 12 emitters. I can’t see run time being very long at all, and the heat of course would overpower whatever fan they could cram in there. It’s all well and good that a cell can do 10C, but the simple reality is run time is relegated to mere minutes at this kind of discharge rate. Believe me, I know! I actually used 3 pounds of copper for the head of a light, and then some, and with only 3 70.2’s pulling 48.6A from 2 Samsung 30T 21700’s the effect is fairly devastating, both in heat and cell drain. That’s “only” 20,000 lumens…
More less-efficient emitters, more less-effective cells, less heat sink with the addition of a fan, I don’t see it boding well for the monies spent…
Edit: Some might see my comparison as irrelevant, but 12 x 6V emitters run by 8 cells is still 3 6V emitters run by 2 cells, just 4 times in one light. If the heat in my single light with 2 cells and 3 emitters is bad, imagine it multiplied x 4?
I think it’s over 500 watts.
I was recalling that the L6 makes around 4000 lumens with a ~5A draw, so I did simple math to get 30W per emitter. 5A x 6V = 30W in the L6. Multiply that times 12 and it’d be some 360W. I’m probably overlooking something as I usually do.
53,000 lumens divided by 12 emitters would require each emitter to make 4416 lumens, or effectively 12 Convoy L6’s running in a single head.
@DB Custom, don’t forget your 20k lumen load was spread out across 3 emitters, and not 12. XHP70.2s are very efficient below a 5000 lumen load.
Also, do you have a thread of you making that 21700 light? I would love to see it.
XHP-70 emitters make more fuss about being pushed hard. 
Times 12. Still, it’s not all about the heat, power consumption is problematic for the limited power supply. The number of places that require compromise is the real issue.
Even with 9V MTG2 emitters, 4 of them make a lot of heat. I have a TrustFire TR J-20 that uses 3 32650’s to power 4 9V MT-G2’s in parallel and again, in spite of a ludicrous amount of additional heat sink the light does get hot pretty fast. And it “only” makes around 14,000 lumens. This light is monstrous, with very deep fins and a LOT of them. Insufficient. Big power is just difficult to do on battery packs that we can hand hold. We need to use the flashlight to house the cooling system and a belt mounted battery pack to supply the power, this would remove the pack from the heat of the operating light and allow for much higher efficiency. Keeping the cells so close to the heat source is just folly, when maximum lumens is the game.
My big light, or the triple 70.2, is my 5th scratch built light… hence DBC-05. I started this literally the day before I blew out 2 discs in my neck, finished it (as a different plan) after neck surgery. It was satisfying coming back and achieving the results I saw. I don’t take pics while on the lathe, sorry. I’m not a machinist, I’m a photographer… I do NOT want my equipment out in the shop and I DO want to keep all my fingers…
Blue, remember that Imalent is using the older XHP-70 in the big light.
Ok wut. Why? XHP 70.2 are quite a bit better than XHP70s, except for tint shift obviously.
Perhaps they have a problem with their source of cheap XHP70.2s not passing quality testing at such high power levels, or maybe they want higher profit margins by using old XHP70s left by streelight/big light manufacturers since they are way cheaper.
Martin says that he was told the 70.2 have some issues so they went with the more reliable 70 series. I am not sure what they were referring to but that’s what they told him. Obviously, it’s not going to be easy to push the 70.2 to challenging levels when using 12 of them, and at under 5,000 lumens per emitter there just isn’t a problem. The use of 70.2’s in this light would yield even more than what they’ve tested, if indeed the 70’s do 53,000 then the 70.2’s would do more than that without changing anything else.
Now, are we going to pay $600 for a flashlight then buy 12 new emitters at over $12 each to upgrade it? Well, somebody here might. Somebody with deeper pockets than mine…
Edit: Several of us here have pushed the 70.2 to over 9000 lumens in lights like the L6, maybe even in the GT…
:D
LOL! That’s giggleworthy for SURE!
I can see the bones in my hand! :person_facepalming:
Kawi, order me one when you get yours. 
A tangent question on XHP70 vs XHP70.2. have any of you guys that have modded with XHP70.2 experienced the emitters failing? I have read of several cases with the DX80 where 2 of the 4 dies lose connection and go dark. Similar to the common problem with XHP50.2. it can be easy to miss because 2 of the 4 dies would still be putting out a lot of light. You would need to look at it with eclipse or welding glasses to spot the dark dies.
that's the reason why they used the xhp70 .
70.2 IMHO It’s all in the (bad) re-flow too much heat during the flow, or not enough solder with the ones I maimed/killed, never killed a 70 unless I was doing a total clean de-dome breaking bond wires. The 70 is a really tough emitter compared to the 70.2, but what do I know…it’s just from my limited non scientific experiences….
I sent the pic and link to this thread to some of my friends. I personally will not be getting one, but maybe they or some one I know will, so I might have a chance of handling one or two, Pat/Ross are you lurking…
It’s a crazy beast, hope it works out well for Imalent and their customers…… :+1:
Well dang, am I gonna have to build something in an attempt to see those numbers? I have 50 Samsung LH351D’s…