Emisar D18 introduction

Hi Hank, for me there is no interest in a 18350 Tube but a big interest in having a 3x 21700 tube. Ist there a chance to get it? 3x 18650 instead of 3x 21700 is the only reason for me not to
buy this great light.

Right, I remember that now. That was weird. I don’t think that sort of thing is likely to be common though.

Yeah… it’s one of Cree’s new rainbow emitters. Someone on reddit posted a 70.2 beam shot recently:

… which is pretty much the entire reason why flashlights are using emitters like SST-20 and LH351D now instead of Cree.

The issue TA ran into is something only TA has had. I’ve never seen it happen, and I’ve tried. More details are in another thread. All tests so far suggest the D18 doesn’t have any issues like that.

I think it looks silly.
Would love to have one.

If you put a space after the final exclamation mark, the picture will show up. :wink:

Some might like the smaller battery tube but to me it sounds like having a sports car & taking the fuel tank off & replacing it with a petrol can.

Pocketable hotrod is what makes Emisar lights unique. You can get the MF01S, which also has 18 emitters, 4x 18650, should be higher output, more focused throw, better heat dissipation due to larger mass and more finning. But it is not pocketable.

Edit: Sorry I misread your post. I thought you wanted 3x21700 for some reason.

I have concerns that 3x18650 can’t properly power this light, can someone tell me how 3x18350 are gonna do anything but get murdered by what is essentially a short circuit’s worth of current?

The short tube makes me want this light, it is cute. Can we get light creeping out of the fins like in the picture? :party:

But seriously, I think the short tube may only appeal to a bunch of flashoholics, but if there’s enough of them it may be worth making it.

IMO it’s the diameter here that limits the ‘pocketability’ of this light (and soda can lights in general). I personally don’t find 3x18350 interesting at all. Length here is not an issue (102mm!), and to have the same diameter and way less power available just doesn’t make much sense to me. YMMV. Also, those wishing for 21700 or 4x18650, lets all realize that the peak power that even TWO of the best 18650s can deliver will make a light this size hit thermal stepdown in a few minutes. Battery power is not a limitation. Also, anyone that may support the idea that that extra 33% runtime you get on lower modes is so valuable, I ask this question: when have you ever used a light in this category on lower modes (of which are achievable in much more pocketable form factors) long enough to deplete the batteries and require a change?

Hank,
I think you did a great job on this light. Proof of that (in my world) is that I was considering myself done with lights of this general type and I’m very likely to purchase one of these. I’m guaranteed to buy this with the availability of HiCRI very near or below the BBL at lower currents. The problem with 18x emitter configuration is that on any sane output level the Amps/LED is in a range that the newest hiCRI options have so much green shift. Turbo should be great though at a ballpark 3A each. Too bad it’ll be over so fast :cry:

PS:
Considering this problem that I just mentioned, I’m thinking that the use of a multi-channel driver that turns LEDs off as you ramp down (for example FET runs all 18, 13x7135 runs 4, and 1x7135 runs one), effectively reducing the tint shift across the output range would be an interesting solution. Complex and bulky though I realize… 2cents spent :beer:

IIIIIIIIIIIIIIIII LIKE IT! :smiley:

And in the gold color.

And with a short tube.

All schoolgirl giggles and grins with this one, great work Hank!

Question?

Is it Frankendaleable?

:smiling_imp:

I think we have found out that EVERYTHING is Frankendaleable.

Depending on the emitter Vf, full circuit resistance, and cells chosen, there is WAY more power available in 3x18650s than is used here. Lets say you pop in 3 Samsung 25S (0.03Ohm) cells and are running L2 flux bin SST20s. Knowing Emisar/Noctigon designs, resistance is pretty damn low, lets say 10mOhm this is the most sensitive variable and also the least informed. Rough calculations show 46.5A against these assumptions and approximations. Comparing against Maukkas output test and assuming 15% OTF losses, these numbers line up with output specs as well. 14000 OTF would require ~915 per emitter which requires about 2.6A each for a total of ~47A and ~150W. Battery pack wise, the cells are seeing ~16A draw a piece, which is easily handled by even 30Qs for short periods. Adding one more cell only increases total output by ~18W (12) and decreases cell load to ~13A each. Lumens increase of say 10.

Is 150W not enough power for this light? If not, does 168W pass? Is 16A the short circuit current of a high drain 18650? Does 10% more output (and heat) seem like an appropriate trade-off for a meaningful difference in diameter?

I think your concern is that each emitter isn’t maximally driven by only 3 cells. That is true. And here the exact point of the design. Getting very close to 100lm/W OTF is the main attraction to a 18x emitter light and something that should be appreciated for what it is. I think HANK balanced the design here very well. KUDOS :+1:

The Samsung 25S can supply 28 Amps to 17 emitters in my Ham’r. At that level it’s making, surprisingly enough, just over 14,000 lumens. :wink: (Yes, one single 25S!)

Plugging a single cell into the same calculation as above I got 27.31 amps. Probably luck that everything comes out that way. What emitters are in the Ham’r? Link to your build?

Great work Hank!

Hank, how much will it weigh?

Make a 26650 attachment!

Totally agree… as lumen increase, bigger capacity are needed…

I have 15000 lights and 4 x 18650 doesn’t provide enough capacity.

26650 would be absolutely monstrous. Way too large IMO.

What could be actually done is an 6x18650 extension, in parallel of course.

That will probably have more probability than the BLF Q8/Lantern extended tubes.