That copper bowl seemed pretty solid when he chunked it down there at the end. Pretty cool! I want one! 
I bet he could make a mean stillā¦.
Ooooh, too much work. Try explosively formed metal workingā¦ just blast that puppy onto a mold.
I would think presses or extrusions would introduce micro stress cracks that would be difficult to polish out. Donāt mind me though, what I know about lathes wouldnāt fill that reflector. This has been a very interesting thread to follow. Thank you.
Try it. Itās easier than you think. Make a solid paraboloid from steel and press a sheet of aluminum over it using a tube. If you donāt have a press, use a hammer. Because of the shape, you may need to experiment a little and use progressively larger tubes with the right taper but pressure wonāt be a problem with a small reflector. Maybe even use a male and female with just enough space in between for the aluminum. If the aluminum and steel are polished and oiled, it shouldnāt need much finishing.
I got the reflector today and have been busy! Nicely done, and the adapter sleeve is a great idea. As expected, my measurements translated into Buckese gives a not-quite-perfect fit, but that's ok, nothing a little work with a file didn't correct.
As can be seen above, I had to file an angle on the outer portion at the bottom to clear my connecting wires. The hole for the LED was a bit small, the reflector wanted to sit on the corners of the die substrate and actually shorted at one point, so I'm going to have to replace my driver spring, it's a bit worse for that little experience and is now a bit lopsided but now I'm using an insulator pad and all is well. I used Mother's Billet Polish to give it the shine, by hand, as I had no way to mount it to turn it. Literally an old T-Shirt and my finger pressed hard into it and turned by hand. I took a few turns at polishing top to bottom as well (an in-and-out maneuver, instead of round and round), in an effort to smooth some light ripples. The most difficult part is right at the opening for the LED, but I don't think it's affecting the beam much. There are a few rings, and it isn't quite as tight a beam profile as the OEM reflector, but there's a side effect that is rather pleasant...the color of the light is cooler from the Bucket reflector, while the OEM by comparison almost makes it look like the LED is de-domed. I assure you, it's not. I shot the following mouse-over's with the Bucket reflector first, went in the house and put the OEM back in, came back and took duplicate shots. The settings are all the same, I didn't even move the tripods. Canon G1X @ ISO1600, f/5.8 1/2 second exposure. (wide angle shots are f/5.6, cause that's how the camera rolls)
In all mouseover's Bucks reflector is first, mouseover for the OEM Raysoon F13 reflector.
First, the red oil drum shot. Wide first at 28mm followed by a second mouseover at 112mm. The drum is 97yds away.
For the next 2 wide/tele mouseover's, the garage door on the trailer shed is 58 yds away. Again, Bucks first, then OEM at mouseover.
I made a mistake earlier quoting lumens on this light, transposed a different light's lumens output from memory. This Raysoon F13 has a Qlite with 8 chips stacked on the inside, for a total of 16 chips and is making 5A at the tailcap from an Powerizer LiNiMnCo 26650 cell. Earlier tests when I first built the light showed 1124 OTF lumens at start-up and 1080 OTF at 30 seconds. I did not do full 30 second tests as I was doing quite a few and didn't take the time to charge the cell between em all for dead even results.
That being said, when I first took the reflector out of the box it needed some polish work. I tested it anyway and got 721 OTF lumens within 5 seconds. After a brief polish session, I got 1061.5 OTF with the stock set-up showing 1090.2 in between the two Bucket tests. After getting some file work and taking .040" off the overall length from the bottom, I have the reflector sitting solidly down onto the pill and making 1056 OTF lumens on the weakening cell. For the beamshots, I charged the cell back up.
So, while it may look brighter with the OEM reflector in place, it's actually misleading. There is more spill coming from Bucks reflector so a very similar lumens output is being dispersed more evenly. The color difference is quite apparent, and I like it! So yeah, I'll be keeping this reflector in the Raysoon f13! :)
Thank You Buck, for all the hard work and dogged determination.
Dale, I'm actually still a bit surprised that it works at all. Thanks for doing the beam shots and the rest. Did you try adjusting the reflector up in the sleeve to see how that changes the focus? I'm curious what raising the reflector up a little would do. The focus point is just a bit below where the parabolic curve of the reflector ends. Adjusting it up in the sleeve would also eliminate the need for the insulator pad.
I wish I had the right felt bobs to get down in the bottom and remove those grooves. The grooves that I was able to polish away higher in the reflector did really tighten the beam up. I'm fairly sure that the remaining grooves are what's keeping the hot spot wider than it could be. Amazon has some 6mm felt bobs that might be just right.
I just raised the reflector about 2mm above the adapter, instead of about that much below it. Lux when from 11.75Kcd to 12Kcd. So yes, it did indeed increase the hospot to raise it . Seems counter-intuitive as it has the bottom of the reflector above the emitter. (FWIW, the OEM emitter was doing 35Kcd)
If I feel up to it later today I might try a few things. Drugged at the moment, right side up is kinda sidways and crooked. Bad day.
I had taken a half mmm at a time off the protrusion at the rear end, getting the emitter up into the reflector where the curve started. With 1mm off, there started to be a silver ring showing at the base of the curve, which was catching the substrate under the die. Then the insulator gasket raised it back up a bit and lost that.
As far as the remaining grooves go, I think I can sand with 600 ā1000 grit paper with an opposing stroke, top to bottom or vertical instead of lateral, this should knock the top off those rings and give a smoother appearance, polishing it back up to mirror finish of course. Wish I had a way to turn it, Iād mix some hand polishing with some machine rpms. 
Elbow grease and a bit of old shoe leather really works wonders.
Iāll be back as results make themselves known.
I like it! Itās not as well focused but it looks much nicer than the F13 with those awful transitions from center to edge. I hope it doesnāt lose that quality in focus.
I like it too, Iāve played with it some at Buckās suggestion and raising the reflector actually does tighten up the beam a bit, I got another 1Kcd but itās still considerably lower than OEM. 12Kcd vs 35Kcd. Lumens output is very similar, so thereās no big losses there, itās just radiating the light in a broader pattern with less intense a hot spot. Works well, but Iāll seeing if I can refine it, and then I might just put a de-domed XM-L2 U2 1A in and see how it likes a de-domed emitter. Then Iāll compare both all over again.
Made some changes to the light, and reflector. I dropped a 20mm Noction with XM-L2 U2 1A emitter on it in the gas, then proceeded to stack 4 more chips on the Qlite in the Raysoon F13. So now, with an added 1/4ā thick piece of copper under the emitter shelf and thermal pasted into place, the F13 has 20 chips and easily pulls over 5A. A rested Moli showed 5.56A before it started dropping. The Powerizers are in the charger. I spent some time hand sanding the reflector to get rid of some small ripples. Itās looking pretty dang nice now, but Iām having some difficulty getting the de-domed XM-L2 up into the reflector. Might have to take a bit more off the bottom, but Iām seeing the silver substrate in the edge of the curve so I think it should be far enough up. Tighter hot spot now than before.
When the cells get topped off Iāll run some more tests.
The newly polished reflector, showing some TLC
Looking damn good! This has been a fun, yet aggravating project. My thanks for taking the time to experiment.
While the lumens remains a close race with the OEM reflector, the lux is a different story. With the Powerizer fully charged, 20 chips on the Qlite, Iām seeing 6.34A at the tail. This gives a start Out The Front lumens of 1390.35 for the OEM, 1373.10 for the Bucket.
But, with the OEM reflector in the newly modded light, Iām getting 96.5Kcd for 621.29M throw. The Bucket is up to 35Kcd with the OEM centering ring, for 374.17M throw. Much more spill getting out, albeit the hot spot is better now with the centering ring on the de-domed XM-L2 U2 1A.
I would get beamshots of this, but Iām a wuss. A cold front has come through and the North wind blowing at 33Āŗ makes me stay inside. 
Tomorrow.
There are some rings on the periphery of the beam, but the OEM has much worse stuff in the beam, again trade-offs.
Personally, I really like the chunky shiny solid aluminum in there. So the Bucket is staying in place.
More beamshots when I can.
Cool. I'm glad you like it.
Iām sitting here with my wife, on our 11th Anniversary, toasting to the new year with some sparkling apple-peach cider in our Waterford Crystal toasting flutes and guess what I see?
The inside of the Waterford Crystal glass appears to be a perfect parabola! Crimey! Iām trying to figure out how to pour molten metal into a crystal glass to make a reflector! Itās an addiction I tell ya, pure and simple! And yāall started that reflector craze right here, in this very thread!
Happy New Year!
No need for thatā¦ you can silver it like they do (did) mirrorsā¦ formula involves silver nitrate and sugar. Something like:
hint: use distilled water (chlorine in tap water will cause silver chloride to form), also clean the glass firstā¦ like with nitric acidā¦
You will need to do-do it regularly since the silver tarnishesā¦
Cool, then the only real problem would be cutting the crystal glass and boring a hole for the emitter to sit at the focal point of the curve.
The problem is the thermal expansion of the glass. If you had a fused quartz glass, it would work fine, but with any other glass you would have to heat it and cool it slowly to keep the temperature even or else the internal stress will break it.
Remember that, if you rotate a liquid as it hardens, it will form a parabola. We did that with polyester, but it would also work with metal. Best to have a furnace with a rotating element inside, but it might also work well to pour the metal into a rotating mold.
The problem with the Waterford Crystal toasting flute is a moot point. Mess with it and my wife would kill me, so itād be pointless. No?
I suggested waxing the inside of the glass and pouring JB Weld in it to get a male mold. She was almost into that, then realized the possibility of the crystalware being ruined and shut me down.
No, not if you write me into your will firstā¦ J) Iāll take the cameras and flashlights.
Sheād keep the cameras and give the flashlights away to disaster relief.
Edit: I feel like thereās some kind of conspiracy going on or something, geesh!