If 6usd or thereabouts happen, I’m in for at least 2-3.
They can not bend in this design, please look at the drawing, the plastic part of the led is surrounded by the aluminium shaft for half its length, the part sticking out is in a cavity. This is not a weak point for the robustness of this light.
Count me in.
The parabolic reflector design does look reasonably robust to me; the bottom third of the LED encapsulation is still well socketed into the reflector. If the light is potted, it won’t be possible to push the LED backward into the driver cavity or bend it sideways, either.
It might not be quite as indestructible as the deeper-set arrangement of the Fenix E01, but it should still be pretty solid.
However, a standard parabolic reflector with the LED chip at the focus may not behave as expected with a 5mm LED, because the LED encapsulation is not curved for the light to arrive at right angles to the boundary between the encapsulation material and the air, which is what you need to avoid unwanted refraction.
Instead of having a hemispherical dome centred on the LED chip like a power LED does, a 5mm LED has a cylinder with a hemispherical dome on top. This is liable to bend the light path as it moves from the LED encapsulation to the air, so different parts of the parabolic reflector will see the light appearing to arrive from different points with a range of apparent offsets from the reflector focus. This means it will never form a collimated beam, no matter what you do.
In short, I suspect that the parabolic reflector will produce a fairly horrible beam pattern. I’d still be interested in seeing a picture of the actual result, though, just to make sure 
I also suspect that the deep-set position of the E01’s LED in its conical reflector sidesteps the effect of varying light paths by simply discarding all light exiting the sides of the LED encapsulation.
I presume that it would be possible to create a reflector curvature that would compensate for the refractive effect of a 5mm LED’s encapsulation, but making such a bespoke reflector would be astronomically expensive by flashlight standards. It would also be thrown off by variations in the 5mm LED shape from one LED manufacturer to the next.
Personally, I’d be inclined to stick with an E01-style conical reflector, on the basis that we already know it works.
^ I have had those thoughts, in fact the curved top of the led is a lens that projects the hotspot and that light is not hitting the reflector. The reflector is there for light that comes out the side flat part of the led, in the hope that it can be collected and added to the hotspot. I can see that happen with some experimenting of where the correct focus is. I’m not sure either if it really works, and if the amount of light going sideways is at all significant. I have no 60 degrees Yuji leds or I would do some experimenting myself.
If the Sofirn folks find that it does not work, they will automatically return to the conical reflector, they already reported that is easier and cheaper to do the cone.
In for a 3200k
Here is a survey for the tailcap. It’s been a while since I’ve done one of these, so if there are any issues please let me know!
Seems to work well. But can you add the Manker E01 as a possible design, so with a lanyard hole at the side and a magnet in the middle. I do not think we will get the magnet but a suitable hole (I think of 8mm diameter, 2mm deep, a magnet that size will easily hold the light and is 17 cents at Fasttech) should be no problem.
I took the survey, but I wasn’t sure about the end cap shapes. It would be great if there was a page that showed and named each tail-cap shape individually. The link back to jon_slider’s post didn’t show all the shapes that were in the poll and didn’t have them individually labeled either. Thanks!
I’ve added that in along with a Yes/No on magnet.
Jon’s post had them labelled, and other pictures followed in that section of the thread. I’ll see if I can get better pics for each individual one.
Edit: I’ve added pics to each. Unfortunately it contorts the image substantially on phone, apparently. I’d have to get home and do some basic image editing to make them look a bit nicer.
I’m in for 2
Please add 2x5600k Sofirn $6.
After thinking about this some more I came up with a new tail design.
My design uses a standard split ring, hangs centrally, tailstands well, includes a magnet and shouldn't add any manufacturing complexity or cost.
Here is a visualization of my idea (please excuse my terrible photo editing skills):
The area enclosed in black should all be the same height as the central lug. This should make it tailstand fairly well (not as good as a completely flat bottom, but definitely better than the three spoke design of the E01) and the split ring can freely move from a central position to one side. The tailstand area can be increased if a slightly larger split ring is used. The Olight i3E in the picture above uses a 10 mm split ring, a larger 12 mm split ring works better with the design (12 mm is also the size you'll find on medium sized Swiss Army knives).
The red dot represents the magnet. On the i3E, which has a tail diameter of 12,7 mm, a 3 x 3 mm magnet fits no problem. The tail of the jet to be named Sofirn light should have a slightly larger diameter which means a 4 x 3 mm magnet should probably be possible.
Just for comparison: a 4 x 3 mm magnet has about the same retention force as a 5 x 2 mm magnet (according to a couple of magnet shops). The stainless steel DQG Tiny AAA uses a 5 x 2 mm magnet on the inside of the light (i.e. there is a layer of steel between the magnet and the surface it is attached to), yet the magnet is strong enough to hold the light, even horizontally.
What do you all think?
Perhaps TheShadowGuy could include my design in the survey?
The magnet does look a bit small to me, I really like the robust magnet of the Manker E02. But if it works, it works.
Derived from your design: you can do the same with a trit hole instead of a magnet hole, a trit hole fits easier even.
So stealing your idea :innocent: I wil make a drawing in a moment (after I put my son into bed)
I checked the retention forces with the help of a couple of (German) online shops which sell high quality magnets. It could of course be that some random cheap magnet of the same size is weaker. And of course this is all just "in theory" one would need to actually try it out to see if such a small (and off center) magnet is really strong enough.
A trit slot is of course also possible, I hadn't even thought of that, but it gives me another idea 
.
If we move the magnet to the other side we could have both a magnet and a trit slot, win-win 
.
It is a bit of a tight squeeze (4 x 3 mm magnet and 6 x 1,5 mm trit), but should be doable. I have measured everything with the help of my i3E (which has a 4000K high CRI Luxeon TX in it BTW ).
my 5 cents
- No to magnets. If you want to stick your light to a magnet, just put the magnet on the surface you want the light to stick to… the battery in AAA lights is very attracted to magnets and will hold the light easily, without having a magnet in your pocket, which will stick to all sorts of things, including keyrings… Just say No to magnets permanently attached to a cheap light.
2. Yes to center hanging a light that is going to live on a keychain, even if that means no tailstanding. imo tailstanding is overrated. I used to prioritize lights that can do tailstands, but I usually prefer to lay a light on its side, because when I tailstand it usually means the glare of the LED will be in my line of sight, which I hate
sooo
3. Tailstanding… nah
build a light with a center hanger, and no magnet, even if it has no tailstanding ability… thats my opinion… otoh, I know people consider tailstanding important, so if you have to have it, then put the keyring on the side of the tail.
4. regarding trits, NO!
definitely not on a 6 dollar light
I dislike trits, they are not stealthy, there is no way to turn them off, except that they die with time
5. Reversible pocket clip YES YES YES!
the ability to stop a light from rolling off my nightstand is a high priority (I do not keep lights on my keys).
I OFTEN clip a light to my hat for hands free use… the Fenix has NO pocket clip… if the Sofirn has a reversible clip option that will make it Significantly Superior, for the way I use my lights.
now back to your regularly scheduled dreaming
and btw, I applaud the progress Sofirn has made
regarding the “reflector”, KISS, there is no need for fancy angles on a 5mmLED light, it wont make any difference, and takes extra work
remember, this is a Cheap Light… fancy trits and fancy reflectors are completely out of character, and imo add nothing except extra machine work
regardless, I look forward to everyone getting their Yuji single mode AAA light
these are just my opinions, please dont argue with me about them… Im sure you each have opinions of your own… post them, and good luck
In for a couple. These leds are awesome. I mod my Photon Freedom lights with these.
I won’t argue Jon but you do have rigid ideas 
I made my drawing, I took the maximum possible 14.3mm tail diameter 2mm hole right in the middle, for fitting I took an existing and appropiate looking ring that I had, it was 10mm inner diameter. The grid is 1.25mm/square.
project looking good.
Still like the trit slot idea, personally if the light is nice enough I would put one in and it still looks nice without a trit, and the front end will collect debris just as well.
But the poll may get us a ‘BLF-preference’, and after all that, I would not be surprised if Sofirn has ideas of their own as well.