Custom 17mm 5Amp PWM-less linear driver-preliminary

Thanks Tom, that’s exactly the kind of independent validation I was looking for to show that “should be” measures up to “is”. Thumbs up.

I had the driver in two different HD2010 pills, but using the same HD2010 body. Not an ideal comparison and it's only one value. Hhmm - should go through the lower modes as well...

Thanks for doing that real life comparison!

(Im still waiting for the driver... )

Hello, led4power

Could you change mode option?
Simple mode option is good.
However I want many mode option like Zebralight’s lights.
Simple 3 mode by selecting how long time click it on and each mode I can adjust sub mode.

As you knew, this circuit’s efficiency is very good. I want to use it’s circuit in many power output other that moon, 1, 20, 100% and Turbo.
My wish mode group is below.

1. High group -> 100, 70, 50% and Turbo

2. Middle group -> 30, 20, 10%,

3. Low group -> 5, 3, 1, 0.1

4. Special -> Strobe, SOS, Beacon

Please tell your think.

Thanks
Kim.

s. sorry for my bad English.

I think perhaps the real question is if you can find anyone to write new firmware for this driver.

I’ve thought about writing a ZL-like UI for attiny13a chips with a momentary switch, but I don’t know what MCU the led4power driver is using. It looks like something else that I’m not familiar with. And I’m not sure if the output levels can be changed only via firmware or if it also requires hardware updates.

Why not ramping?

It's much more intuitive(at least to me),and you have practically unlimited numbers of modes(8-10 bit resolution to be precise).

Some people like ramping. I kind of prefer a fixed set of modes with known output and runtime. This way, I know better what to expect (will this mode last for 5 hours or 15?) and can get a better feel for how dark-adjusted my eyes are at the moment.

With my JETbeam RRT01, I can always get exactly the right amount of light… but I never have a clue how long it’ll last. And I always end up setting it as dim as possible for whatever I’m doing, and then thinking my eyes must be really tired or something… until I turn on a fixed-output light and find out I only had the RRT01 at like half a lumen (because suddenly 3 lumens looks bright).

I normally configure my lights for 5 modes — moon (0.1 lm to 0.5 lm), low (3 lm to 7 lm), medium (15 lm to 40 lm), high (100 lm to 200 lm), and higher (300+ lm). Depends on the light though. I really like the ZL UI, which allows for 10 levels with reasonably easy access to all. It’s not so great when lending the light to a muggle, but for my purposes I like it a lot more than ramping.

If I can find time to write a ZL-like firmware for attiny13, I’ll include the source in my firmware repository. I’d kind of like to have that UI on my Skyray King, but I’ve been really busy so I’m not sure when it’ll happen.

I haven’t tried ramping yet but it sure sounds good. Maybe I won’t like it as much after the novelty wears off though. I do like variety so either way I’ll probably end up with several different ui’s rather than just one for every light.

DrJones lumodrv (R86 edition)

Ramping and user programable modes in one UI. Best of both worlds..

No sources for lumodrv though… so no joy.

?

You buy them from DrJones. You can specify number of modes.

I may just be weird. I mostly live in a world ruled by source code, where anything without source has little if any relevance. Seeing people release only the binary, or even making code available only in hardware form, is a bit of a culture shock for me. After leaving proprietary software behind a couple decades ago, it’s a deeply ingrained reflex to grab the code and fix it any time something isn’t quite right. It takes a conscious effort to not ignore anything I can’t get source for.

Ramping is fun but not something I’ll use running on trail. (Planning to put the driver into a headlight). Taking attention away from foot placement could result in a stubbed toe. Like ToyKeeper, I like having solid modes with predictable runtimes. I also have a couple lumodrv s but asked Dr Jones for solid outputs of 5, 25& 80%. ( I love the battery monitor!)

@led4power

Congratulations. The test results show you did a remarkable job and after a little tweaking here and there this will be a helluva driver for high and low current applications. I’m truly impressed.

Eh? Weren’t you purchasing ZL and other proprietary flashlights up until the last month or so? Don’t you use tons of proprietary commodity items on a day to day basis (such as your TV, the BIOS on your computer, the computer in your car, the stereo in your car, almost all your network equipment, definitely your modem, etc)? I like open-source and free-as-in-freedom as much as the next cat, but I get the feeling that this “conscious effort not to ignore” claim is just a high-pressure way of [not] asking people to release sources. Maybe it’s more accurate to say that you make an effort to focus on free & open alternatives when they are available?

It’s very possible that I’m just really happy to have the ability to change the firmware now… but it had been bugging me ever since I first got into flashlights. Even the best stock items could be improved, and it bugs me when I can’t improve it myself. For example, I’ve tried repeatedly to get changes into the Zebralight firmware, but they generally just give me a pat on the head and tell me I’m cute.

I’m not Richard Stallman, so I’ll use proprietary stuff when there’s no other option, but it bothers me and I tend to avoid it. Like… my TV is running a customized version of Cyanogenmod, and my notebook is running modded BIOS, but I haven’t attempted to change the firmware on my microwave. I’m pretty sure though, if I was an active member of, er, Budget Microwave Forum, I’d probably have a dozen microwaves with firmware I wrote myself.

This new driver is pretty exciting, a major step forward for lights… but it’d be even more exciting with source code and a flashing kit.

Oh - btw, I'm trying to fathom what I'm seeing, and was leery of posting this. But, it's consistent and I checked my lightbox/meter against other lights (low and high output) and everything is in order.

So, I got the LD-1 driver jumpered at what should be non-turbo, mounted in a big fat VOB copper pill in a HD2010, 20 AWG wires, XM-L2 U2 1A/Noctigon screwed down using GC Extreme, wired up springs using 22 AWG, NO-OX-ID treatment on contacts, measured this:

LG HE2 4.21v, 5.13A tail, lumens: 1,666 @start - 1,581 @30secs, 105 kcd at 5 meters

It's consistent, and checking other lights, the meter seems fine. The beam pattern sucks because of the off centered LED (crappy centering piece that's too small), but it sure is putting out a lot of light! It's still, I'd say, maybe 100-200 lumens higher than comparable setups, but the difference is with this: screwed down into the copper pill with 20 AWG wire (I was using 22 AWG, not using screws), and of course the LD-1 driver.

Fair enough, I agree with a lot of what you’re getting at. I tend to stray towards more flexible solutions, and those solutions tend to be more open, and… so on and so forth. It’s interesting how stock firmwares become much more flexible nearly simultaneously with the release of aftermarket firmwares (FOSS or not).

You’re right. Ramping is good.
But I used to with fixed mode option.
I just want to know fixed mode firmware will be done.

Can you do that? How many circuits do I have to buy if I want my own modes option?

So are we any closer to a date when this will be released for sale?

I’m in for at least 6 if they are around $15 each.