FW3A, a TLF/BLF EDC flashlight - SST-20 available, coupon codes public

I’m down for 2, and 1 each of those sounds good, maybe 4500 for the warmer one.

That has been mentioned before. If it can be done / accomplished, I can’t see that it would be anything but a win-win for all involved.

Cause your right I think…. no matter which way it goes there will be people who will not buy this great light because of the emitter preference, as you mentioned above. :frowning:

Brilliant.

lol
that was not one of the choices I gave
here is why

however, I expect some people will swap in N219b 4500k

yes, and that would be unfortunate. I hope most will find a way to have the LED swapped to the flavor of their choice.

fwiw, Some modders have stock of exotic LEDs that can be had for a nominal fee. I sometimes choose modders based on whether thy have an LED I want, and Ive recently had mods done by both Texas_Ace (4500k N219b), and Clemence (2000k E21A quad). I recommend both of them.

Your FW3a’s could be drop shipped to the modder of your choice, and you could end up with a grail LED for less than half the cost of an HDS

Both 219b and e21a are pretty much not usable on a fet light.

I have some 219b FET lights. The tint isn’t as good on direct-drive modes, but it doesn’t generally have problems otherwise. On a FET+N+1 like this, it could work pretty well. The highest regulated mode would be only about 800 lm though.

If the FW3A ends up using the LH351D, please make sure it goes back to the clear 10507. There’s no tint shift whatsoever with the Samsungs, and at least I wouldn’t want it to be even floodier than it already it is.

So far, LH351D and 4500K have held a lead for quite a while. 4250K and 4750K keep swapping for second place, with 4750K currently in the lead. XP-L HI and 219c keep swapping for second place, with XP-L HI currently in the lead.

I haven’t done emitter-CCT correlation yet, but from how things look now, it seems like the best option would be to ask Lumintop to make two flavors:

  • Floody: LH351D 90CRI in 4000K or 5000K (not sure which)
  • Throwy: XP-L HI in … 4A? (~4650K)

I’m not sure if they’ll be willing to do this, but if so, it seems like it would cover the widest range of preferences.

No max. specified length 67mm.
Aim is as compact as possible. Batteries with Micro-USB are ca. 69,7mm.
Maybe if you replace the tail spring.

Would it not be easier to chose one emitter option and give the option to order without leds?

Sombody mentioned only the 4000K LH351D is 90CRI. The 5000K is only 70CRI

The tree only 10m away under in the beamshoot is an important information.
How far is the house in the background away?

Can you add this to the description in the poll?

I guessed the tree on 20m. Being only 10 away I think the floddy one pisses away light has to much unused light.

I will vote for the XPL Hi. Having no dome it has less colored rings at close distance.
Seems for me best of both worlds for an EDC.

For the temperature, I want less scatter with fog, mist and dust in the air. 5000K isn’t good for this, 3000K would be good for this, but will never win.
So 4000K.

On the other hand you can polish the frosted optic. And if you want to swap LEDs frosted would be better.

How floddy is the Samsung LED compared to an Astrolux S41 / S42 ?
The first is quite popular so people can relate to.

That’s what I thought, thanks!

There is a 5000K 90CRI version, but BLF hasn’t tested it yet.

I went out to measure it… that tree is only 8.5 meters away. The house in the background is 53 m away.

So, yes, it’s very floody.

Both lights were set to the same amount of lumens according to my light box, about 107 lm, to show how things look at a medium level.

Interested in one, please add me to the list.

Interested in one. Thanks

But it was a “simulation” using a different LED? Until you have a prototype with Samsungs, or Nichias in it, it’s just an approximation, and potentially mis-leading.

Besides the various optics are inexpensive and (unlike the LEDs) trivially easy to change, maybe offer a set of alternatives for those who would like to experiment, don’t already have some in our bits box, and don’t want to buy them individually at higher prices.

That’s one of the great attractions of a torch using “standard” TIRs, the ease of experimentation.

Or at least show us how the beam looks once the frosted optic is polished, or the tighter one is fitted.

Or set up a third poll for optic preference :wink:

PS: Please clarify how many 7135s are in the “N” bank. I think you suggested that the max. regulated level (i.e. the max usable level) is equivalent to a fully loaded Nanjg ?

PPS: whilst I agree that using linear drive is more efficient than a crude FET driver, the low Vf of the Nichia still means inefficiency, just that the waste heat is dumped in the driver rather than the LED, which is why I am keen on seeing the Samsung used.

PPPS: what is the mass of proto2 (and the original design) ?

I should mention that in my image, which has similar cd but less angle covered compared to the samsung fw3a, the distance to the furthest treetops is 60m, and they are clearly lit. The comparison shot shows the approximate difference in cd/lm (though it should apparently be 2.25 cd/lm instead of 2.1) not the ultimate throw. I think the key thing here is to be sure the level of lumens at which it can continuously run is higher than what is needed to produce sufficient cd, whether the multiplier is 2.x or 4.x, and I think the lh351d is more than enough.


(rem: this image was unedited and the camera rendered darks as darker and lights as lighter than my eyes. The flashlight is the only source of light here; anything visible is easy to see without dark adapted eyes.)

I’d just like to quote our resident expert djozz. From two days ago.

Though he made a typo, I think he meant to compare with LH315D.

ISTM that the LH351C (not D) could be perfect for those of us who value good CRI, with less flood.

Surely worth evaluating. Available. Well priced. Let’s look forwards, not fall back on safe but unimaginative choices.

Interested in two. Thanks