Iām drooling all over my keyboard with anticipation :disappointed:
ā¦. when are the new D4s with Toykeeperās driver tweeks going to be available?
Will MtnElectronics also sell the version with the tweaked driver?
Well, I havenāt finished the code yetā¦ so not until thatās done. But hopefully within a few days?
I think RMM is planning to update his stock too, because, as T18 discovered, the thermal regulation really needs some updates for safety.
ā¦ For the sake of avoiding posts aboutā¦ roastedā¦ well you know. :person_facepalming: 
Accidental activation almost caused roasted nuts? ā¦.Yikes! 
Lights with side e-switches come in 3 flavors:
- Raised button - the button sticks up above the casing of the light. Easy to find the button, but also very easy to accidentally activate. Takes up very little space inside the light.
- Flush button - Button is at the same level as the casing. Accidental activation is less likely than with a raised button, but the button can sometimes be hard to quickly find.
- Recessed button - Button is recessed below the level of the casing. least likely to suffer accidental activation as the raised sides of the light tend to keep things like the sides of your pants away from the button. A deeply recessed button can be easier to find than a flush button as it acts like a funnel for the thumb. This is probably the best option for a pocket EDC light as it offers the most protection from accidental activation, while still having the convenience of instant access with no lockout. Downside of this type is it may take more space inside the light.
This is one lesson Zebralight learned.
The Zebralight SC51 had a large soft-touch raised button. Accidental activation was very common and tailcap lockout was a necessity. There were lots of complaints about it. Zebralight fixed this problem in their new switch design, which debuted in the SC600 Mark 1 and has been used in all their subsequent lights. Their new switch features:
- a smaller button
- stiffer switch. The button is still easy to press, but requires more pressure than the original SC51 switch.
- deeply recessed switch. The high sides of the casing surrounding the button makes accidental activation very unlikely. Tailcap lockout is no longer necessary.
For the D4: From looking at the disassembly of the light, the simplest hardware solution to solve the problem of accidental activation would be to replace the bare-metal retaining ring over the rubber switchboot with one that has high sides that project above the level of the light by several mm.
With this proposed new retaining ring, the sides of the retaining ring would tend to come into contact with your pants before the button making accidental activation much less likely. Replacing the Emisar D4ās retaining ring with a raised one is a good solution as it would only require replacement of 1 part and would effective convert the light from a āraised buttonā to a ārecessed buttonā.
I measured the XP-G2 (5700K) and XP-L HI (6500K) versions. Here are the numbers I got for all three emitter families on the same cell:
- Nichia 219c 5000K: 3826 lm (Efest 18650 @ 4.18V)
- XP-G2 (4000K or 5700K): 3343 lm (Efest 18650 @ 4.16V)
- XP-L HI (5000K or 6500K): 4322 lm (Efest 18650 @ 4.18V)
Also, the XP-L HI melted my ND4 filter. Dangit. Thatās okay though; it still works for my purposes.
Will XPL2 emitters fit under an unmodified Carclo quad optic? If so, what would be the output with them Toykeeper? 
Nope (IIRC), and no clue. But there are reasons people arenāt using XP-L2 in flashlights. I hear the beam is pretty terrible.
Iām surprised the output was so high with the XP-L HI. I figured its higher forward voltage would limit it to about the same as the 219C.
I think XPL HI has generally always produced more lumens than 219C. Thatās been the case since both emitters came out in numerous different reviews. XPL HI simply has higher output. More throw too.
219C may pull more amps so the light heats up faster and battery runs down faster, while XPL HI tends to produce more lumens.
That was my impression too. But based on Texas_Aceās tests of the D320 219C and the XPL HI V2 1A, they produce right about the same output at a given current. So with the lower Vf of the nichia it seems like it should output more. Maybe this light has lower bin 219C.
I suspect Toykeeper also measured max output on a full cell.
It could be that XPL HI starts off with higher output on a completely full cell but due to the higher vF as the cell depletes the Nichia will overtake it in output.
Now Iām tempted to mod my D4. I have Richardās new quad board with XP-L HI V3 3C and 219c D320. The problem is that I love the light as isā¦
Well did edit that part but I guess my point was made except it would have been roasted peanuts, that situation sure was certainly NUTSā¦
Just the combination I think of those being new jeans and sort of stiff, along with seat belt snugged caused the pocket material to fold up and just press that switch all in under a minute blazing hot, couldnāt even hold the light.
I wonder what the people behind my truck were thinking when they see me bailing out of my now rolling truck, I didnāt even have time to put the E brake then at the same time doing a little jig around in the street holding my pocket, probably looked differently to them, and pulling out this massively bright light, man that hurt, must have looked insane I guessā¦ Still hurtingā¦
TK, not technically correct on one pointā¦ the XP-L2 will actually fit under the Carclo optic. Itās a tight fit and you can do damage if not careful but they do indeed fit. (I should say the ones I got early on from Cutter fit, havenāt done it again in a while due to, uh, circumstances beyond my control)
I have the user-configured thermal calibration working. The UI for this is a little more detailed thoughā¦
- Fast-click 10+ times, then click and hold.
- Light blinks out the current temperature limit in C.
- Let go of the button if you want to exit now with no changes. Otherwise keep holding.
- Light ābuzzesā for two seconds. Let go during this time to set limit to the maximum (~70 C). Otherwise keep holding.
- Light goes up to turbo.
- Hold until you feel itās too hot, then let go.
The light then blinks out the new temperature limit in C, and shuts off.
This way, you can check the current number, set it to maximum, or set a new calibration by feel. Also, setting the ceiling to a high temperature ā¦ will hurt. But thatās probably a good thing.
Does this seem like a good way to do it?
Sure does to me, great work TK, is this in the next releaseā¦?
I like that it flashes out what it is currently set at so you can check where you left it.
XP-L is pretty beastā¦ but I think Iāll go Nichia on this thing. Would be my first floody Nichia and highest amped light. My mouth is watering at the thought.
Iām getting pretty close to done.
Currently, there is no way to disable mode memory. Itās easy to bypass during regular use, by turning the light on with a long-press or a double-clickā¦ but it could be a potential issue when kept in a pocket or a bag if you left it in a high mode and it gets tapped by accident.
Making it optional would need some sort of config mode. Adding an actual config mode would require deeper changes and make the interface more complicated.
So, Iām trying to figure out if itās okay to always have memory or if it needs a config mode.
Anyway, hereās the UI so far:
Changes from the original version include:
- added full thermal regulation with user-calibrated ceiling
- added mode memory on click-from-off (default 100% 7135)
- made beacon use current ramp level
- made double-click toggle turbo (not just one-way any more)
- made LVP drop down in smaller steps
- calibrated moon to ~0.3 lm on Emisar D4-219c hardware
- blink when passing the 100% 7135 level, for reference
- fixed display of āzeroā digits
- fixed/cleaned misc internal things
DROOOOL
I NEED 
:laughing: :smiling_imp:
I LOVE that UI, thank you for your wonderful work TK !
I wish the thermal config was that easy in my TA driver with Narsil (I have it in a Nitecore HC30 and struggle to use the thermal config 
)