Emisar D3AA, custom Triple w new AA/14500 driver w NoPWM

[jon_slider]

Haha that was me! JacobDock on Reddit.

LOL, snap!.. great idea ;-)

 

[aznsx]

Everyone should love efficiency, who wouldn't want to know their things are performing at their very best?

Sure, more is better, but prioritized against other performance parameters, etc.

I do have a question for anyone who might have some guidance though, and while it in no way relates directly to the future product under discussion in this thread, it does relate to 'efficiency', which seems to be of great interest here. That said, it will be nice to see that efficiency quantified with actual numbers in % (vs words) when the prototypes are tested, and perhaps compared against other benchmarked driver designs. How much increase in efficiency is definitely a factor to consider.

While we all like efficiency, there's a big difference in where (in terms of optical output power) that efficiency is most important to different users; with some valuing it most at minimum output levels, and others at higher outputs.

The question I have is: To what extent is efficiency (%/ratio) related to relative output (as a % of max output), and is that relationship somewhat linear across the output range, or actually rather non-linear (as I suspect just guessing)? (I'm sure this varies with the specific driver design as well(?). I guess I'm trying to envision a graph plotting both relative output and efficiency as its inputs. Anyone have knowledge of such things &/or an educated guess? My guesses aren't very well-educated, so I'd appreciate any help offered ;-)

 

[jacobd]

Sure, more is better, but prioritized against other performance parameters, etc.

I do have a question for anyone who might have some guidance though, and while it in no way relates directly to the future product under discussion in this thread, it does relate to 'efficiency', which seems to be of great interest here. That said, it will be nice to see that efficiency quantified with actual numbers in % (vs words) when the prototypes are tested, and perhaps compared against other benchmarked driver designs. How much increase in efficiency is definitely a factor to consider.

While we all like efficiency, there's a big difference in where (in terms of optical output power) that efficiency is most important to different users; with some valuing it most at minimum output levels, and others at higher outputs.

The question I have is: To what extent is efficiency (%/ratio) related to relative output (as a % of max output), and is that relationship somewhat linear across the output range, or actually rather non-linear (as I suspect just guessing)? (I'm sure this varies with the specific driver design as well(?). I guess I'm trying to envision a graph plotting both relative output and efficiency as its inputs. Anyone have knowledge of such things &/or an educated guess? My guesses aren't very well-educated, so I'd appreciate any help offered ;-)

I think the main thing that will stop direct runtime graph comparisons between this light specifically is the of DIRECT competition.

For direct and fair Lumen/Runtime (say how long can it hold 200 lumens) comparisons to compare drivers you'd need to completely isolate all of the variables outside the driver, so;

• the same emitter and temp
• same optic setup (to ensure no lumens are lost in TIR vs reflector type debates)
• same battery
• same material/ weight (to nullify differences in heat sinking ability)

I cant think of many lights that share that many similarities that use different drivers. That's probably the MAIN reason we don't see that much data comparing drivers like that, compared to standard runtime graphs. You could likely rig a setup to do these tests with the driver outside the light, cooled and connected to the same batteries, but I don't think I know of anyone doing that.

 

[jon_slider]

You could likely rig a setup to do these tests with the driver outside the light, cooled and connected to the same batteries, but I don't think I know of anyone doing that.

actually, the designer of the driver in the upcoming Emisar D3A tests basically the way you describe, outside the light, using a regulated power supply

when he gives driver efficiency numbers, they are not subject to variations in batteries, optics, LEDs, or other host specific characteristics

for reference, he has tests of multiple Zebra models here,

and to give some perspective on his design strategies, here is one of his posts, about an AA/14500 driver he designed for the FWAA, which is not commercially available. If you want one, you have to assemble all the components by hand, onto the mcpcb he designed

otoh, with the Emisar D3AA, we have our first opportunity to own one of his designs, by buying it off the shelf, preinstalled in a light that is a complete, finished, ready to use light. No modding required.

 

[jacobd]

actually, the designer of the driver in the upcoming Emisar D3A tests basically the way you describe, outside the light, using a regulated power supply

Oh there you go! Thefreeman is so so good for the industry.

What configuration are you going to look to get first Jon?

 

[jon_slider]

What configuration are you going to look to get first Jon?

I like dedomed 4500K 519a lately.. I might go with a Red host.

Ive been wanting a Red host to keep on top of the fridge, as an emergency light if the power goes out (hows that for a self serving justification? lol). The power actually never goes out here. (and I already have 3 other lights on my fridge)...

how about you? ;-)

 

[jacobd]

I like dedomed 4500K 519a lately.. I might go with a Red host.

Ive been wanting a Red host to keep on top of the fridge, as an emergency light if the power goes out (hows that for a self serving justification? lol). The power actually never goes out here. (and I already have 3 other lights on my fridge)...

how about you? ;-)

I've been wanting a dark grey Hank for a little while, and I've been loving the 5700k DD lately, plus I think if he releases them in the MAO white I'll grab one in 2700k DD as a bedside candle.

Although now I better grab a red one as an emergency light! 😂

And of course I'll get the Ti when that makes it to the store. I think I'll wait a few weeks though, I'm always apprehensive about new releases *cough cough* Skilhunt H150 hand grenade *cough cough*

 

[LRJ88]

And here i was thinking that maybe this month wouldn't have something tugging at my wallet.

 

[vicv]

This'll be the first time I consider a dual fuel light. Usually I don't like 1.5v compatibity with my lithium ion lights. As it makes too many downsides due to the terrible performance of alkaline. Even here, eneloop performance will be poor due to the light needing to be able to use an alkaline. But still, this will be a good one. I'm really hoping he follows this with a single emitter and tir version. But he seems to only be interested in multi emitter lights, which are just so inefficient. And yes has a very efficient driver. Still, it's nice to see one of the freeman's drivers commercially available. It's just too bad it's not his buck/boost + fet driver meant for a single lithium ion

I was just reading up on the it more, and it looks likes it's just a boost driver. That's why it's a triple, they're in series. This light also looks likes it's going to be pretty chunky. To the point that an 18650 would barely make it bigger. I think I'll pass on this one. But thanks Jon for bringing it to CPFs attention

 

[jon_slider]

MAO white

Im also a fan of MAO white

it looks likes it's just a boost driver. That's why it's a triple, they're in series. This light also looks likes it's going to be pretty chunky. To the point that an 18650 would barely make it bigger. I think I'll pass on this one.

I agree it seems as large as an FW3A or Zebra SC64.. Im trying not to be an early adopter.. but.. sometimes my Buy it Now finger twitches.. we shall see ;-)

 

[ilikeguns40]

Im also a fan of MAO white

I agree it seems as large as an FW3A or Zebra SC64.. Im trying not to be an early adopter.. but.. sometimes my Buy it Now finger twitches.. we shall see ;-)

I'm purchasing for the main reason it's a duel fuel AA Anduril light that I can mess around with at home and with my kids. Can get it in any of the nice LED options, aux LEDs, freeman driver is a nice touch. Simply a light to mess around with, nothing more, nothing less. Also the price with Hanks I believe is a great value for what you will get. Seems like a fun little light. Will it replace my potted Zebras, Malkoffs, and Peaks for daily carry? Probably not.

 

[jon_slider]

AA Anduril light that I can mess around with at home and with my kids

a kid light using AA sounds like a good idea ;-)

 

[LRJ88]

The preorder is up now, just in case you're interested.

 

[ilikeguns40]

Yup. Just ordered one a while ago. If I like it, I might order another one later

 

[jon_slider]

Im waiting for my wallet to recover from a couple of recent purchases.

fwiw, Jackson offers an LED that Hank does not, the Fireflylite 3750K 351A.. Im very tempted to try it, since I already have an abundance of lights w 3500K 519a (DeDomed 4500K).

The appeal, for me, of the 351A, is that the tint seems to be more pink (based on duv tests by SKV9), than the 519a. I come from loving the pink tint of sw45k, something the 519a does not do for me (I perceive it as more orange than pink)..

temptations abound.. lol

 

[ilikeguns40]

May have to try that one out as well. I've got a bunch of points saved up at Jackson's website. Might be where those points go haha

 

[LRJ88]

It's supposed to use a whole new driver system and all so it'll be fun, i nabbed another KR4 as well since it seems all new flashlights from now on might be getting the new drivers as well.

 

[jon_slider]

May have to try that one out as well

I suspect it is a throwyer LED, although that is somewhat reduced by the use of optics in the D3AA.

The 351A is smaller than the 519a, contributing to a tighter beam focus (smaller more intense hotspot when used in a reflector)

 

[Burgess]

I am *SO* tempted ! ! !

If only I didn't ALREADY have
enough flashlights for the rest of my life,
plus several reincarnations . . . .
< sigh >

Must RESIST . . . .

 

[jon_slider]

Well, here is a reason to resist (maybe futile.. lol):

The D3AA is not more efficient than an AA Zebra, at outputs below about 100 lumens. (based on this post)

otoh, the D3AA still has the claimed advantage (unverified), of no flicker at firefly levels.

this is what I know so far about AA output and runtime for the D3AA