This is some of the most useful data I’ve ever seen on BLF. I feel like this should be stickied or something, placed at an easy-to-find spot to answer common questions. It’s certainly a lot better than my lame attempts to explain the effects of various optics.
I wonder if there’s a way to produce a “beam ugliness factor” similar to the “PWM snob factor” used in other measurements. Like, maybe a sum of the total deltaxy tint deviation distance travelled, or perhaps difference from a straight line, or something like that. Adding up the total vertical travel in that last graph to produce a single number. It’d be nice to have a number to quantify how good or how bad a beam looks.
The deviation should also be weighted by the relative output compared to the hotspot. It get’s kinda subjective when trying to come up with a formula. For me it’s enough to look at the graphs. Any sudden large movements back and forth especially near the hotspot are bad
All of those are still miles better than an XP-G3 though:
Could you possibly get a 10510 elliptical and test that too? I’m extremely curious about that one and its one of the main reasons I’ve been looking at this light…
The elliptical one doesn’t get much attention, mostly because it doesn’t make a circular beam. But I have one in the mail because I was curious if it might work as a bike headlight… wide but not tall.
Here’s a post from 2014 which shows how the beam looks with XP-E emitters:
It’s a semi-nice bike light in urban traffic. The beam of a headlight meets the street under an angle near to 0° and must not glare drivers or pedestrians.
Symmetrical (to horizontal) optics are less suitable. Too bright near the front wheel, too dark in 10 to 20m distance.
I’m not sure the FW3A would be a great choice for biking though, because bikes have a lot of vibration and it could potentially make the inner tube shift sideways enough to register as a click. I haven’t actually tried it, because it has been cold… but it wouldn’t really be my first choice for a bike light.
I have found this project recently and I found it very interesting.
It look like in other thread it is very popular.
I was thinking about making similar flashlight from my blf a6.
I was trying to find some information about the driver there but I did not manage to get through all messages there. Can you please post some more pictures of it from both sides here? Is it 20 mm driver?
I measured the 219b sw40 triple at 1200 lumens with a 30Q and 950 lumens with a GA. No bluish tint or other weird effects so using a medium current battery is likely fine on turbo. The 30Q I used isn’t new so it has a higher internal resistance than a brand new cell.
1x7135 level is 83 lumens with the sw40 so efficiency on low regulated modes is about 30% worse compared to the cool white XP-L HI.
I don’t want to convert A6 into FW3A.
I just pointed out that I found this flashlight interesting.
I have found flooder more useful as EDC light than something between flood ant throw like A6 or S2+. I also prefer triple than quad due to size difference.
Firstly I was thinking of using S2+ host, but A6 bezel will cover a bit less of carclo TIR lens.
Heat distribution is also better in A6 than S2+.
This is why I was planing to make an A6 with triple LH351D.
I have even started to prototyping my own driver for it with FET +1 +5 (or 6) 7135 with few extra features as well as tailcap switch with battery indicator.
When I have started this project I had no idea that there is FW3A on horizon.
FW3A look like exactly what my perfect EDC should look like, thats why I’m asking about some detail on driver as well.
I’m also wondering how it is handling heat that LH351D producing.
Thanks maukka, that’s good to know. I guess you’d lose a similar level of output correcting ugly tints with filters so swapping in 219b would probably make more sense? If only I had some spare 219b emitters
A standard MTN Electronics 20mm MCPCB should fit in this, correct? Going to assemble some tint options in SST20 so that I can interchange them when the FW3A arrives.