TK's Emisar D4V2 review

I think the first builds with that were 2020-03-18, aside from the K1 which had it since launch. It applies to anything with a “voltage” mode for RGB aux LEDs… which currently means just Emisar and Noctigon products.

Excellent :slight_smile:

How about this idea, which includes “Aux mode” into the main operation mode:

- From On, you reach aux mode with 3C (currently this switches between smooth and stepped ramp)

- From Aux Mode, you reach regular mode with 3C

- In Aux mode, 1H is used to turn the aux LEDs brighter or dimmer

- In Aux mode, 2C will reach max aux brightness (aux turbo), another 2C will go back to previous aux brightness

- In Aux mode, 2H will cycle through the various aux operation modes (R, Y, G, C, B, V, W, Rainbow, Volts)

  • In Aux mode, 1C will turn the light off

friends,is possible to dim the aux leds? thanks a lot…stock aux led lumens are a lotttt…

7/3 clicks with a “hold” on the last one cycles colors too.

I just updated the firmware on my D4V2 to 2020-03-18. Is there any documentation that has all of the changes that have been made recently (like the 15 click to get the firmware level) ?

Thanks,
Mike

The most accurate place to get that info is in the change log, a.k.a. “Recent revisions” on this page:

https://code.launchpad.net/~toykeeper/flashlight-firmware/fsm

Are the D4V2’s Aux emitters wired/coded in such a way that they will not vampire a cell past typical LVP? I remember TK confirming way back when that the original D4S’s aux had no intelligence with battery voltage, but I was hoping that the D4V2 did with the voltage aux setting.

When the RGB aux LEDs were added, LVP for those was added too. It checks voltage every few seconds while off, both to make the “voltage” mode work and to turn off the aux LEDs when the battery is low.

I was thinking today that I’d love to see a 21700 sized D4. However, the “twist” would be that there should be two different battery tubes for it. One with a relatively thick wall for 18650s and the other with the “normal” wall for 21700 cells. The idea being that the “thick” tube would give the light more thermal mass and make it tougher. Maybe use the same optics/electronics as the current D4v2 but enlarge the OD of the head to match the tube?

Anecdotally I don’t think battery tube mass affects thermal performance much except maybe in hosts that are have a single piece head/tube like Zebralight.

Indeed.

Am I correct that the side switch is on a separate pcb? Thinking about a custom switch already, depending on how tight everything is inside. Doesn’t look like there is much room to do what I am thinking, though.

Also, is the FET used just for turbo mode or is it being pwm’d in other modes like ramping, strobe, etc?

For the aluminum D4 and D4V2, I don’t think the switch is on a PCB at all. It fits very tightly into a pocket machined in the head right below the switch boot. Below that pocket there’s room for 2 tiny wires to connect from the switch to the driver. There’s no room for a special pcp for the switch unless you start dremeling or glue on a custom raised switchboot.

I haven’t tried opening the head in the Ti and Brass D4v2. Presumably those have a larger pocket with room for a custom illuminated PCB.

I think FET is used for the final 2 modes (turbo). I think the power right below full power uses PWM.

the FET is PWM’d in every output level over 350mA up until the final ramp step of 150 where it is wide open.

Alright, thank you! That means I will have to basically treat this thing like a loaded gun while in my pocket or if a “muggle” has access to it. I have a few muggles to worry about.

Why I was asking:
I was thinking maybe I could add a dead-man style enable switch for turbo mode so that it would physically disconnect the FET gate from the uC when you release the switch. Doing that would obviously not be compatible without completely new firmware and a minor pcb hack, and loss of the functionality that TK spent a lot of time creating. This gives me an excuse to finally do some board design again! The size of this thing will make that a real challenge. I have been getting rusty, it’s been a decade since I last designed a board and I am itching for something fun and functional. I was going to get an FW3A but the price and build quality of the D4/D4V2 is very attractive. To get proper optics in the FW3A you are right back at the D4 pricing!

Maybe that’s because manufacturers just don’t engineer them properly to sink heat from the head.

If you mean accidentally having the button depressed for a long time, resulting in high output (heat) while the flashlight is in your pocket, it’s nice to know that the later versions of Anduril now have “stuck-button” precaution. Basically if the button gets depressed continuously for about 10 or so seconds, the light will ramp to the top and then ramp down to the lowest brightness… and if the button is still stuck, the flaslight will go to Lockout mode automatically to prevent it from turning on accidentally again.

When exactly are later versions with this feature? Interesting just tried it:

  • D4SV2 ordered Feb 20th 2020 from IOS -- doesn't have the feature
  • Dec 2019 Anduril I customized in a SC31B mod -- yes, has the feature!

Ouch! Guess Hank is not keeping up, but I can understand - it's a lot of extra work if they were already manufactured earlier. At least I can now re-program the D4SV2 without taking it apart.

I had read through a lot of the later posts and did not see that was added, thanks! I ended up buying a gray KR4 last night since the traditional tail switch seems easier to modify and I can give it a quarter turn to disconnect the battery.

In the end, I am hoping I can cram in a design that at least performs to SIL2. in the meantime I still have my eye looking out for a used D4. I have a feeling there will be some soon. It’s scary that people on Reddit are saying they bought one and are asking how to charge the battery. Holy crap! I expect those will show up on eBay soon.

Hank told me that the switch is glued into the tailcap.