If you have the V2 it looks like you can just ramp to ~500 lumens and use Fast Click to turn it on and off thereafter, memory will retain your level.
Perhaps a level lock or single mode use is a feature that could be built in to future versions if there is a demand.
The reasons for this have been discussed in this thread in a fair amount of detail. The 219C is generally pretty similar in efficiency to the XP-G2, but in a direct-drive light, the emitter that draws the most current tends to be least efficient because all LEDs lose efficiency at high power. Using medium modes doesn’t help with this because they’re produced using PWM; the LED is rapidly turned on and off, and the percentage of time it’s on determines the brightness, so it’s always operating at the worst efficiency the battery has power for.
Forward voltage determines how much power the emitters can pull from a given battery. Lower voltage = more power. The 219C has much lower forward voltage than the XP-G2 and a fair bit lower than the XP-L HI.
On low modes, the D4 uses a linear regulator. This runs the emitters at the same current regardless of voltage, where the 219C should actually be more efficient in terms of lumens per watt. The linear regulator turns any voltage the LED doesn’t need into heat so the whole system is still a little less efficient with the 219C.
With a current-controlled buck, boost or buck/boost driver, the 219C has much better efficiency than in a linear or direct-drive configuration. Compare the two emitters in a Fenix TK25 - XP-G2 left, 219C right:
Firelight2 did you get your different samples eventually? :lookstohimwithbigroundedeyes
Or just buy a weaker, less dangerous to children and newbies, flashlight to be safe. Don’t put a 55mph governor on a corvette
I agree with this. My D4 and Convoy S2+ triple and the FW3A (when I get it) are in a different category of lights and they are not for anyone but me to tell the truth. The D4 got amazingly hot and frankly surprised me, but after using it for a couple weeks I love the power (although I mostly use it for reading at night). It’s just a light to use with plenty of caution.
I’ll hand another, less-powerful light to the kids if they were to need one.
The more I tinker with this flashlights operating system the more I like it. This is by far the best I have used. TK I see your little tweaks all through it and I am amazed how many fine details. A small example would be that I just found out that if, for some weird reason, I needed to use the beacon mode, I can now set what exact brightness level I want. Originally it was arbitrarily set at 30% if memory serves. I love the thermal control, it is nearly seamless, in practice I have to pay close attention to see each small decrease. I also find the fact that I can disconnect the power and then reconnect it and when I power the light on it goes straight to maximum regulated (7135 chip) power level for battery conservation. Great Job putting your expertise into this flashlight. I have zero idea how you do it, and I have zero idea how you even learned how to do what you do. I get a headache thinking about it 
I second that, all the thanks to TK, the Goddess of Light.
The D4 v1 was a nice little toy, but v2 is a masterpiece. I never imagined to see model predictive control being used in flashlights.
And by making the code open source, I imagine the UI and temperature regulation code will be copied and used in many future flashlights.
I believe we are witnessing a major milestone in the advancement of flashlights.
The XP-G2 is made for a max current of 1.5A, the Nichia 219C is made for a max current of 2.4A (1.8A nominal) and the XP-L is made for a max current of 3A. Easy to see that the smaller emitters are being overdriven and as such are making more heat for the lumens they emit.
We have learned, of course, that these Can be overdriven, they all make excess heat when pushed though. LED’s used to be known for low heat and high efficiency, and they still do that if you keep them to the current level of a single 7135 regulation chip. None of these were designed to run direct drive through an FET off a 4.2V high current Li-ion cell.
I almost bought one of these when Richard got the last batch in, but I really don’t have the available funds to spend on a light that mimmicks so many of those I built over the past few years. New to a great many of folks here, very old news to me. The driver of course has new tweaks and represents a milestone in refinement, other than that, this light is what I’ve been doing for a while now… small form factor in hand, large form factor in output. That always put a smile on my face…
And I agree, the driver in that one light must be sitting on the switch wires. Y’all think about it… Richard got how many of these in and pulled the driver, reflashed the firmware, replaced the driver and sent them out post haste… at no charge? Yeah, bound to happen sooner or later as it is so easy to have the small switch wires coil wrong when you put the driver back in. C’est la vie.
Dale there is a good chance that it wasn’t anything Richard did at all. It looked like the battery tube had some kind of protrusion that caused it to dig into the driver base as it was screwed in fully. When the guy realized it he filed or sanded the edge until it was even , but when it was screwed fully in again the FL was still having the same issues. I think by that time the damage had already been done and one of the wires on the opposite side was compromised. He even said it would work a little if he only tightened it partway. Either way I am sure Richard can and will fix it for him. As far as I can see that is the only problem I have heard about with the D4. Pretty darn good considering the high amount of heat and light generated by such a small form factor.
I don’t believe he re-flashed this light as it was purchased this week and was part of the latest batch which I think was shipped to him with the new firmware already. The box says v2 on it.
I’m going to take it to a friend with a soldering setup today (I’ll post pictures) and look to see if it’s an easy fix. If not I’ll send it back.
This is the sort of thing that would be worthwhile warning the production guys about, in case someone on the assembly line has been twisting the wires the wrong way during assembly, eh?
And add “tighten the thing down” to the “does it light up” quality assurance test step?
Makes more sense ledalex, Richard is very conscientious about his work and it would be a rare day indeed for something like that to slip by him…
The source code is available, and info about flashing firmware is at the Link in my signature. You can make it do whatever you want.
I agree. TK I have to say I absolutely love the 3D tint on the xpg2 in my newest D4. It’s my favorite so far, at least in floodier lights like this one. The slight rose tint is very pleasing to the eye and doesn’t wash the color out of everything you look at. Thanks for the recommendation
Here’s the switch wire that was.m being compressed:
I added some electrical tape and put it back together:
It still doesn’t seem to be drawing enough current though because the output is much lower than it should be and it’s stepping down even lower after a few seconds. EDIT: battery was not fully charged (4.0v) charged it a little more and it does seem a bit brighter. I’m waiting for it to fully charge now.
I miss the trick to get the default 100% of the 7135. Could you please add in the OP as a separate point under Interface / Other actions:
If you want a dedicated brightness you can untwist and twist the tailcap to get 100% of the 7135. Which gets you 138 lm (319c), 161 lm (XP-G2) or 145 lm (XP-L Hi).
Thanks
Joe
Were strands broken on that damaged wire? If so it’s effectively a much smaller conductor.
Or was it just the insulation broken so it grounded out?
I’m not sure that would make a difference here. The broken wire is only a switch wire. If it works at all the switch should work.
Since main current doesn’t flow through switch wires, if output is still low I don’t think reduced conduction through the switch wires would be the problem.
A couple other notes:
- Even a 4.0v the head of the light should get too hot to touch within 20-30 seconds. If it’s not something else may be wrong.
- The head of this light can get VERY hot. Standard electrical tape can melt and might not be the best choice to reinsulate your broken switch wire. I suggest using Kapton tape or Arctic Alumina thermal epoxy (both available from Mountain Electronics).
Added. And another note that the tailcap is a good way to lock out power.
Yeah, I think something else is going on. Unscrewing the lens and moving the board with the leds results in some weird stuff going on. Maybe the driver is bad.
I haven’t heard back from Mountain yet. Hopefully they respond. You would think he’d check them before sending them out.
Hank, I don’t see why it would matter how good of a conductor that wire is. If the switch is working, it’s allowing enough current through.