Yes I know and for this reason we completely re-invent the way power is managed in the guts… In fact it’s super-easy; like I said it’s a different approach. What we do here has been proven to be the right way; we did the “something” which originated the major brand grilling monitoring (from your phone), smart home implementations, also from your phone… and a few more things which did sell to major brands under NDA, so please don’t ask… Now… Torches are a new hot thing for us, but it’s just another kids (nerd kids - that is) task… In real life we do work on serious shit; this is just us playing with code and some hardware like we used to do when we were kids… Just wait and watch! I checked-out your link; pretty cool but very different from our project. We are going to offer full custom control of a light producing device. Only the sky is the limit. There will be remote and scenario triggered sequences, home protection, alarms and so much more…
I still hope to pursue convenient methods of transferring data to a light. It can enable some pretty useful things. Specifically:
- For regular users, it’s kind of a fun novelty.
- For regular users, it becomes relatively easy to have one’s favorite UI on every item.
- For regular users, it can enable post-production firmware updates, to fix bugs and add features.
- For regular users, the possibility of updates improves the feedback loop between users and developers, so the features can be optimized better for what people want.
- For vendors, it becomes possible to calibrate and customize products without the usual hassle of resoldering parts between measurements.
- In general, if the code is open, it helps facilitate a collaborative community… which is good for everyone involved. As one person put it: You get what you pay for, everyone gets what you pay for, and you get what everyone pays for.
These things should ideally require as little change as possible to the host designs and performance. If it needs a USB port, for example, the host must be bigger and take extra care for waterproofing. If it uses a radio, it may require part of the host to be non-metal to allow transmission, and could impact battery life. It may be feasible to put pogo pads or something on the visible side of the driver, but that still requires a few dollars of extra hardware for the update process.
So I found a way to do it with no extra hardware, without changing anything except the driver — use the LED itself to receive data. The main downside is that the data would be transmitted from a phone screen or similar, which has a pretty low and highly variable refresh rate. So the transfers are slooooow. Sending a complete ROM would probably take hours. It could be done from almost any device though, by simply going to a web page and clicking a button while the light is held against the screen.
It may be preferable to use bluetooth instead, for ROM updates. But then there are a bunch of complications related to the radio, and related to the phone/PC side of the transaction. Some people have opted to put the radio under a plastic or rubber portion of the host (Manker Godmes, Quark Smart), some have tried to expose it through the reflector or optic (Flex Asgard). I haven’t seen one which did actual firmware updates that way though; so far it has mostly just been gimmicks.
This version, which works perfectly fine for someone like you and me, might be seen as rather complicated for the rest of the world… The concept you are talking about appeals best to a “click-happy” person, but in today’s world we also use displays which make more sense because YOU CAN ACTUALLY SEE WHAT YOU ARE DOING. Don’t get me wrong, I applaud your results using conventional options, but there is more; don’t you want to see the potential of MORE being unleashed?
hmmm ?
Yep…more words than action
a lot more words
Oh, um, Tterev3 made one with a display:
Some other people, and some companies, have also added displays to their lights. The Nitecore TM26, for example. I don’t recall offhand what the others were though. The idea seems to come and go periodically.
Thanks for the review, i just got mine in cyan with xpl hi.
When the tailcap is tighten, there are two blinks, I’m not sure if this have any other meaning or just a notification that it is ready to turn on. If it is just a notification, it would be much better if the blinks indicate the battery level, like 4 blinks if battery greater than 75, 3 blinks if greater than 50. So you know the battery level everytime you unlock the flashlight, without having go to battcheck mode and count 12 blinks (for example if battery is 3.9 volt, and don’t know 3.9 volt mean how many % left 
).
Also two method of showing battery level is also great. Please let me know if you think this is a good idea
On my firmware, the “power is connected” blink is much faster so it doesn’t really take any time. The slow blink was there because that’s how Tom likes it.
It could display remaining charge then, but it’d greatly slow down the process of physically unlocking the light for regular use. It also would need to wait a bit for voltage to stabilize before doing the measurement.
If this is a thing you want though, the firmware is available for modification. The simplest way would probably be to use Anduril, edit the setup() function to go into battcheck state instead of the “off” state, and edit the battcheck code so it’ll go to “off” state after one loop. And probably change the battcheck style to 4 blinks instead of volts+tenths, and perhaps speed up the blinking.
Another option is to make it do this only if the button is held at boot time. There is already a clause to handle this differently, though it isn’t compiled in by default. Usually I use “hold button at boot” for a different purpose though — factory reset.
I think 4 blinks can be made short enough that not to cause any discomfort when we want to turn on the light after tightening the tailcap and we are fast enough to count.
I agree with you that on the d4 the default two-blinks is quite long.
You maybe right that the light need some long time to measure the battery level right after tailcap is tighten
this might sound strange but i want to limit the maximum current when using fet, is it possible to add an resistor to FET modes only?
D4Ti, C type (purple), and F type (golden-blue) are back in stock, with very limited numbers.
The D type (purple-blue) is out of stock and discontinued.
The colorful version D4Ti will not be produced again when they are sold out.
It would probably not be easy to physically modify the driver to reduce FET current, but maybe one could reduce contact to just one leg of the FET’s ground pads? I’d suggest using thinner wires from driver to MCPCB instead.
Or you could modify the firmware.
Holy cannoli those new D4’s are stunning! I put one in my shopping cart just to pretend I could afford one of those raw titanium ones. $100 for one of those really is a great deal for what you’d be getting. I’d love to see how much of a difference the copper head makes on those hot rockets. I just got a D1 (thanks bansuri) and it gets super hot. Makes sense though because it shines farther than any other light I’ve had the pleasure of using.
It rocks, but because of titanium being a worse heat conductor and copper better than aluminium, I think it will level out.
Got my 219c with 18350 tube. Running 30Q and vapcell 18360.
I both love it and feel frustrated at the same time. It repeatedly steps down at 100% 7135 level at room temperature of 22C / 72F. Even at lower output levels. This is starting with a cold light and the thermal reset to 65 degrees. Is this typical?
No, that’s not typical. It sounds like it either thinks the battery is low or its thermal sensor is way off. I haven’t heard of one being that far off before though. Being off by as much as 15 degrees would be normal, but 40+ is way out of spec. I’d double-check the batteries and make sure the parts are screwed together firmly.
In some newer firmware, the thermal floor is higher (about 150 lm), and the user can calibrate room temperature to avoid issues with hardware variations. So maybe that could be a solution too.
Thanks. I recall it reading room temp about 14 to 15 degrees out from actual. I’ll check to make sure the board is tight.
Have anyone measure the current go through the led board?
Thanks a lot Hank! Is there any chance for a price drop on the aluminum version of the D4, too?