Personally, I don’t mind the weight because the light is so small anyways. I vote copper.
You have a point with the wear Djozz, but soft copper is certainly not a good option than.
Maybe an alu pill and a brass ring for the ball slot.
Or maybe a more wear proof alu like 6082 or something.
Although I think it would take quite some time to wear out soft alu.
Brass = the golden standard for metal machining. Any hard/easy to machine standards always compared to brass.
Brass has the best dry lubricity for metal to metal contact, superior galling resistant. Even against it own
Aluminum can be anodized or greased for better wear resistance. But seriously, a true 20xx aluminum is very strong. I don’t think all those manufacturers with “aerospace aluminum material” really use 20xx or 70xx aluminum. These aero grade aluminum can also heat treated for even better strength. I will check local supply for used 7068 scraps. IMO, 7068 Al is the ultimate flashlight material.
[Clemence]
/\ 
I think Alu is fine
I have a box full of damaged Vertex dirt bike cast racing pistons, they have very high silicone content (some as high as 20% for 2000 model). The Si content is so high they are brittle yet very hard, perfect for high revving low torque 2 strokes. Tried to cast them but unusable other than for their specific purpose. Tapping a thread on it was useless.
I heard Oveready use superior ultra strong aluminum billet alloy for their lights. It’s doesn’t really requires anodization.
[Clemence]
ALU 7068 voted.
I say aluminum. The light is too head heavy. I have had other lights with ball bearing and magnetic ring in aluminum without issue.
I also well grease all my ring lights.
I like copper mcpcb and aluminum pills. Uses the superior thermal properties of copper where its most important. Moves the heat quickly to the aluminum and keeps the flashlight nice and light.
I predict synthetic diamond is the future for extreme high power. DCPCB has been used in military and space equipment, mainly for laser and CPU. The cost of synthetic diamond keeps dropping as the technology advanced.
This far, the nearest concept to DCPCB at mass production rate is AlN direct metallisation copper board.
Soon there will be copper based nano ceramic insulated MCPCB very similar to my previous board. The factory introduced it to me but the whooping $10.000 commitment cost held me back from even read the spec any further.
[Clemence]
More pics added in OP
testing Clemence!
Apart from the nice graphs, I think it is nice to know that the 18350 Vapcell has a functional thermal fuse.
Thanks for the tests Clemence! Very useful.
Are you planning to change the driver too to add LVP? That seems like a tough job.
I would do aluminum for sure on mine. These are heavy lights, I agree. To lessen that weight would be great
I really want to if I have the skill. Need more time to learn building my own driver. I have long overdue programming lesson to begin with. As for now, simple mechanical fix to make it accept protected 18350 is the easiest solution
I won’t worry about the overheating problem since I will use heavier resistor to limit the output to about 300lm ish
[Clemence]
Definitely let me know on the Aluminum version. May need a few.
Far better would be for Jetbeam to update their circa-2012 driver design. They should be able to not only add at least a thermal stepdown and low voltage protection, but also achieve better efficiency and better regulation.
Although the new design looks a little bit nicer aesthetically than the old one, I’m unclear what functional improvements were made relative to the old version.
However, if you want a rotary and an HDS is out of the budget, this is apparently the best game in town despite the shortcomings. I’m looking forward to seeing what Clemence is able to come up with for a mod kit.
I haven’t seen any good protected 18350. I bought a bunch of the Aspire 18350 that is known to be the highest capacity 18350 in existence and some of the older Vapcell 18350 before the supplier changed their formula to a lesser capacity. Unfortunately there are no protected versions of those cells.
Hmmm limiting it to 300 lumens is kinda low. Your graph shows it starts at about 780 lumens. But not sure what else can be done.
I can’t believe Jetbeam didn’t implement any thermal step down or LVP for a light of this price.
300 lumens might be a bit low but the 400-500? lumens of my triple old style rrt-01 is more than sufficient for most tasks.
Keeppower and Eagtac have high capacity protected 18350s, but I haven’t tried them yet.
My experience is that 600lm for a S2+ is about the max it can sustain but it gets bloody hot already. The RRT-01 is a bit smaller and I would not want it bleeding hot, so 300lm sounds pretty sane. Maybe 400 because we are BLF? 