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

Please Add me to teh List


Please add me to the list. Thanks!

Please add me to the list too. I created an account here just for the F3WA.

I’m curious why XP-G3 is used instead of XP-L2? Is there a chance the more efficient XP-L2 V6 bin will be used or nichia 219c?

Consider Nichias plz.

Looks like I’m the first one to get a prototype. Here are some images and early observations about it.

Note: This is only a first prototype. The FW3A is still in development, and all issues mentioned here should be fixed before release.

The prototype appears to use new host parts designed for this light, but some of the components are probably just whatever Lumintop had laying around. For example, it has cool white XP-G2 emitters instead of neutral white XP-G3, and the driver’s 7135 chips are the wrong brand. These things are not cause for concern.

First, some pictures.

FW3A prototype:
(yes, the tail anodizing doesn’t match… Lumintop was experimenting)

With some other 1x18650 lights: (Convoy S2+ silver, FW3A, Emisar D4 grey, BLF A6 bare)

The beam (XP-G2 cool white, 10507 optic):

The business end, next to an Emisar D4:

Without the bezel or optic: (looks like XP-G2)

The battery is replaced by removing the front half, not the tailcap:

What I can see of the driver without removing it: (not sure if people have seen Fritz and DEL’s driver design, but it’s a thing of beauty)

Some initial thoughts / observations, in list form…

The good:

  • The host looks and feels really good. Will probably become my new EDC.
  • The FW3A is slightly smaller and significantly lighter than an Emisar D4. Should be easier to carry.
  • The clip feels good and seems to work well.
  • The clip does not rotate, despite appearing loose in an earlier pic. It seems to be sandwiched between the ledge and an O-ring.
  • Physical design overall is very nice.
  • The driver and firmware work together, despite having never been tested on hardware before.

The bad:

  • Sometimes when I put a battery in and tighten the host, it acts like the button is held down. May require small adjustments to the switch tube or contact ring to make sure this doesn’t happen.
  • The button is very jittery. It’s hard to get the intended number of presses. This can be fixed in firmware.
  • Moon mode is visibly unstable, which is odd and unexpected. Might be fixable in firmware, but might just be a side effect of the prototype having the wrong brand of 7135 chips.
  • The beam has corona artifacts, normal for the 10507 optic. Can be fixed with 10511 optic.
  • The XP-G2 beam has a bit of a green ring. Can be fixed with XP-L HI or Nichia 219c emitters.
  • The button is very slightly loose and its click is a bit louder than most e-switches.
  • Tail section ano doesn’t match the rest of the host. Lumintop was experimenting with different methods.

Other notes:

  • Due to the limited thermal mass, the FW3A isn’t going to be able to sustain high output for very long. The D4 has difficulty and it has significantly more mass. The FW3A is comfortable and practical due to it’s light weight and smooth shape, but it’s not made for extended use on turbo. It should be able to do ~3000 lumens, but I expect the sustainable level will be a few hundred lumens.
  • The entire tail section seems like it’s glued together.
  • My clip appears to be slightly bent.
  • The clip could be a bit stiffer.
  • Emitter wires may be a little thin. Might be a good thing though, to reduce maximum power (and heat) a little.

Firmware on my prototype is a beta version of Anduril. Basic use is mostly the same as NarsilM, but it has a few extras.

I have a few things to fix right away. Development may be a little awkward because it looks like I may have to solder every time I reflash the driver, but that’s not a huge obstacle. Anyway, some things for me to fix…

  • More tolerant button de-bouncing.
  • Unstable moon mode (if it’s a firmware issue).
  • Thermal regulation. (I didn’t test it yet; I’m just assuming it needs fixing)
  • Discrete ramp mode acting a little weird, not sure why.
  • Misc small firmware details, like perhaps using 8x7135 for strobe instead of the FET. Etc.

That’s all for now. The FW3A is real, and is making good progress, but there are still things left to do.

Thanks for the update on the “work so far” TK! It really seems to need some improvements and changes, but I must say I’m very pleased to see this “project” getting real shape :wink:

From what you said some things are of more “concern” to me, like the button and the noise, and the capacity of heat absorption. But, due to the design and to keep it “small”, this has to be a light for really short turbo boost!

I won’t enter the LED choice discussion as I’m not expert on that, but for what has been said, the LEDs can be easily replaced as well as the optic? Maybe this is not a major concern for me but for other members it may be…

Thanks for the update on this and… please get that Anduril ready to rock on this beauty :smiling_imp:

BTW, I hope The Miller is ok! :+1:

Nice report! I’m still very much looking forward to having a few of these lights. The size is wonderful and wish more manufacturers would make them this short. I think that’s why I like my Zebralights so much. A single emitter one would be a hit with me as well due to the small size for an 18650-powered light.

As for the power, it’s nice to have lots on tap, but I’m very happy with the middle settings of my Convoy S2+ and Astrolux S1. It’s fun to crank up my D4 to impress others, but I think that this light will still be popular for having a great size, extra light when you need it and ramping to allow anyone to adjust to their heart’s content. Oh, and a pocket clip!

Looking forward to seeing changes and updates as they come.

This thing really is small for an 18650 light. It’s shorter than some 1xAA lights I have, and even shorter (and lighter) than the Brass Beauty (though not as narrow). It brings good runtime and hotrod burst mode to a small form factor, with plenty of flexibility and extras. Once we get it finished, I think it’ll be a huge hit.

Like! Thanks for the update.

Just so I understand, is the inner tube around the battery electrically isolated from the body? Is this the circuit path that the E switch uses?


It looks like it. The klarus xt11s has an identical inner tube/setup

Pictures of how Klarus does it

Just wondering, where was it decided to change the plan? I signed up for…

What do you mean? What has changed?

Was that question about the tube? It seems to be there…

The long ramp from the front section, down the tube, makes it look like a Nitecore Concept 1 or HDS Systems EDC Rotary. Not saying I don’t like the style, I do, it made me buy the Concept 1, but Fritz’s design is cleaner, neater, more business art and unique. The custom straight off the lathe look is what drew me to it in the first place.

Oh, now I see what you mean! Yup, indeed this prototype seems a bit different from the original drawing! Maybe it’s just due to being a prototype? Or will it be like this in final version?

I like the neat clean look that Fritz designed…

Minimalistic, elegant in it’s simplicity.

Oh, you mean the shape under the clip? The hard edge would grab your clothes and make it difficult to unclip. The tapered edge allows for it to slide cleanly.

This change was posted a week ago so a didn’t realize that was what you were talking about.

A simple 45º bevel at the hard edge in question would solve that without adversely affecting the simple elegance.

Sure, I could just chuck it up and recut the tube, no problem, but figured I’d ask why it was changed.

Just like Fritz’s clean lines as compared to mimicking others…