I think the delay is getting the tube-in-tube durability figured out. Is there a reason they donāt just epoxy the tube inside the other tube so there is no movement or chance for wear between the two causing shorts? Do we need the inner tube to come out for any reason other than the cool factor?
The tube-in-tube design with just anodizing to insulate them never seemed wise.
A better way of doing it is how Liteflux did it in the LF2XT: Have 2 tubes - an inner aluminum tube nesting inside an insulating plastic tube. With 2 tubes, there is zero chance of wear causing a short. And the tubes do not need to be thickā¦ very thin aluminum and plastic tubes should work fine.
If I understand correctly, yes, that is why proto2 wasnāt the final version.
Itās not a problem with the inner tube rubbing against the outer tube. The issue is the inner tube rubbing against the driver retaining ring. And to fix that, I think the gap between them has been widened.
The proto2 lights work fine after putting a strip of Kapton tape around the front end of the inner tube, but it shouldnāt need that. So a revised CAD model was sent, and is being produced.
FWIW, the prototypes so far are:
Proto0: Made by Fritz on his lathe.
Proto1: Made by Lumintop, had several issues to fix.
Proto2: Made by Lumintop, fixed all significant issues except one.
Proto3: Being made now.
Firmware is pretty stable, but I think Lumintop opted for the safer of two thermal options ā the method which drops to the 8x7135 level at the slightest hint of overheating while the FET is active. Normal regulation is used below that level. Specifically:
Ramp level 1 to 65: No thermal regulation, not bright enough to need it.
Ramp level 66 to 130: Full thermal regulation.
Ramp level 131 to 150: Stays at requested level until overheating is detected, then smoothly drops to level 130. Will not attempt to regulate back up above level 130. This is basically a temperature-based step-down with a smooth edge.
With default settings, level 130 is the ramp ceiling. So every default level except turbo has thermal regulation, and turbo has a temperature-based step-down.
Additionally, there is some extremely paranoid thermal code in muggle mode:
Overheating shouldnāt be possible at muggle-mode levels, but if overheating is detected somehow, assume something has gone seriously wrong. Drop to the lowest muggle brightness.
I think the plan is to test proto3 whenever it arrives, confirm everything works, and then start production.
The user can calibrate the sensor and set a different temperature limit. The user can also change the ramp floor and ceiling levels. However, the level at which the thermal algorithm changes is hard-coded. Changing that requires reflashing the firmware.
ToyKeeper, is it possible to buy the driver and tail switch for this light by themselves? Iām interested in playing around with the electronics without waiting for the production lights to ship.
I will just hook it up at my bench. Iāve never played with a light with an electronic tail switch and a ramping driver, Iām curious to see how it works.
It will work like any other combination of driver with e-switch and ramping firmware. Having the e-switch in the tail is only special inside the flashlight. On the bench, it wonāt be in the tail anymore, just a standard e-switch setup.