I want to build a “wow” light just for the sake of it and because I haven’t done any builds for a little while.
I am wanting to run it off of a single 18650 (or 26650) and will be direct driving it or possibly using one of those BLF17DD drivers if I can get a hold of one.
Will be braiding everything, using 20ga silicone wire, an XM-L2 on a Noctigon and I currently have a Sony VTC5 spare to use or I will buy a high drain 26650.
So hit me with some good host options. Money is not an issue but bear in mind I don’t want to buy an expensive light with high quality internals only to rip them all out.
Current ones I have considered are XinTD C8 V5, Trustfire X9 (which I would chuck a big chunk of copper in because of the hollow pill) or a Ultrafire HD2010.
I know the bigger reflectors of the TF and UF will be better but the quality of the XinTD is much better and some of the HD2010 reviews I have seen show some questionable QA issues and it only has an aluminium pill.
I'm not crazy about the charging gimmick, but that's more or less a 26650 C8 with a momentary side switch. And no pill, I believe, supposed to be a solid shelf one-piece with the head.
Using the BLF-17/20DD drivers with the momentary firmware is as simple as connecting one wire from the switch to pin #2 on the MCU and the other wire to ground.
17DD here with momentary firmware in a little UF-T50 clone:
edit: X9 head is 59mm OD, H33 is 55mm, that's not much difference even if the reflectors scale directly with the head OD, which there's no guarantee of. Difference would be minor at best. X9 is also $10 more for a 18650 light with a 19th century-tech tail clicky.
I am actually keen on the rear clicky because it is what I am used to. I had forgot about the X3. That could be another option. Protected cells not fitting is no issue, I don’t think a protected cell will be able to supply enough current anyway? I assume a direct driven XM-L2 with lots of braid and good connections will draw ~6A?
The momentary switch UI is amazing. From off, a short press gives you the lowest mode, or a long press gives you the highest mode. At any time, a short press makes it one level brighter, and a long press makes it one level dimmer.
You may want to stay far away from it, though. It will make you hate all your other clicky switch lights.
Doh, I think the way you explained it confused me. When you said short press to lowest mode, I was thinking of that got it to moonlight, or the closest thing to it. It's really just a UI that starts in low without memory. That firmware just got me started learning C. I've worked with a lot of different languages over the years, but somehow never worked with C. It's a lot of new ground for me since I've never done firmware programming or designed PCB's either.
It starts in low OR high, depending on what you want it to do. It doesn't have an instant-off function and memory, but to get that you'd have to sacrifice the ability to choose which mode you're going to get before the light is turned on.
The mode order is circular, with 'off' being just another mode but with 0% output. Short presses go clockwise (brighter), long presses go counterclockwise (dimmer).
You're right, 0% is effectively a mode, and if I read the mode code correctly, there is mode memory in that it's always loaded in the array, but it isn't set up to work like mode memory usually works. Locking out loses it though.
No, you cannot turn the light off unless you cycle back around to the 0% mode. There is no memory. If you plug in 5 or 6 levels when building the firmware you're never more than 2 or 3 clicks away from the 'off' mode. More than 6 or so and it can get a bit awkward.
I didn't say different. I'm talking at a code level. There's memory, but no way for a user to access it. With direct to off added, bringing memory should be very simple as long as direct to off doesn't make mode_idx=0. Memory is used to get turbo. It just reverses from 0% into the previous mode, which happens to be turbo.
Oh, I think I see what happened. I made a typo. I typed "mode" when I meant "code".
