TK's Emisar D4 review

Here is close to a “kit”:

I’ve read hundreds of posts and finally successfully flashed. Here is a condensed version of the hardware and software you would need if running a Windows computer (I use Windows 10)…enjoy!
I had some trouble figuring out exactly what to buy, but I figured it out after reading all the posts for months. Just get those three things I linked below for hardware.
I bought this:
https://www.amazon.com/gp/product/B00AX4WQ00/ref=oh_aui_detailpage_o02_s…
and these:

and this from ebay:
https://www.ebay.com/itm/Programmer-Testing-Clip-SOP8-Pin-SOP-SOIC-8-SOIC8-DIP8-DIP-8-IC-Test-Clamp-NEW/172420293749?hash=item28250ce475:g:sjYAAOSwK~RaTaoE
you have to bend the clip pins a bit, but works fine and it’s not 20$!

ya…cheaper from china…but not by much, and I prefer Amazon reliable quick ship. Not worth saving 3 or 4$ ordering from china and waiting a month.
Windows programs:

uspasp driver:
here is the driver and manual for the chip programmer…read page 7 for install
scroll down for the driver download and manual pdf

avrdude flashing program:

hit the green download button…installs an “install” program…
TK’s hex files library:
http://toykeeper.net/torches/fsm/

If you plan to do this for the first time, I would recommend adding a board like this to your purchase list. It allowed me to program a dummy part without tearing apart my light and risking damage the first time. The one I bought was local and cost less than $4. In the total cost of things it was worth it for sure.

Also DEFINITELY use a wooden tool to try to remove the board. I forgot this on my second attempt, broke a chip resistor using a metal object, and still have not recovered, even after replacing the broken part. I think my clip is just not connecting, but I have put this project down for a bit.

sbslider, that’s why he’s interested in learning on the D4S, it has programming vias on the driver board so you don’t even have to remove the driver. :wink:

:blush:

Hey, it’s a good tip though, what you said… for sure I’ve busted up some driver components trying to bust glue before just like you said. I’ve learned to eyeball down in the hole and maybe feel around a bit with the tip of a small jewelers screwdriver in an effort to get on the pcb before hitting it. :wink:

I remember one that was a true PITA, couldn’t get it to budge. Finally got it all apart and it was the strangest set-up, they had a small driver board up top of the contact driver board and I had knocked the screwdriver all the way through that first board! Totally annihilated it, what a mess! lol (a solarforce light, if I recall correctly. Had a top board soldered into a brass holder and the bottom board also soldered in then the assembly is press fit, sure a trick to get out)

I notice that the Ti version is out of stock on Hank’s store, so maybe it is nearing the time for a new model/revision?

I would very much like another D4 but not until it gets the newer RampingOS V3

I think black (alu) body with the copper head would look amazing and hopefully not as expensive as the Ti model.

I don’t know if that will happen, but I sent Hank updated firmware today.

http://toykeeper.net/torches/emisar/?C=M;O=D

Changes compared to the previous version of RampingIOS V3:

  • Added a moon timing hint. When doing a hold from off, this blinks when passing the threshold between “click” and “hold”, to indicate when to let go for moon.
  • Better timing consistency on hold-from-off. (before, the stepped ramp took an extra 0.4s to start ramping up)
  • Fixed a bug where changing ramp style could go to the wrong brightness level. (go to a med mode, turn off, hold for moon, click 3X to change ramp styles, and it’d go to the med mode instead of staying at moon)
  • Added a D4-219C build target. (FET ramp power reduced to 80%)

Very interesting, thanks for the new info.

Wasnt able to reset password. New problem. The light had problem with switching off after got warm, and had been turning on often, but last time it just couldnt turn off so i had to twist head off. I contacted hank but he only said to take bezel off and look there for some screws, but i dont even know what bits to buy, are they philips or waht. This being asked, hank evaporated disappear. Very pity that i actually sent him the whole amount too fast.

Said:
Looks like there is short somewhere, would you please unscrew the bezel, take
the optic out, then try to take those two screws out, and see if it will work fine again.
But when the screws are not there, please use the low mode only, not the turbo mode.

If the light is stuck at a bright mode and won’t turn off, it generally indicates one of two problems:

  • There is an unintended connection somewhere, allowing power to bypass the driver entirely. If this is true, then the light should act like a 1-mode twisty — tightening the tailcap would turn the light on at a very high level, without even pressing the button.
  • A button wire is broken or shorted or disconnected. If tightening the tailcap causes the light to blink then turn off, but it doesn’t respond to button presses, this is probably the issue.

Yeah, but what now. I am not tecnically teached to repair this pos, and hank doesnt give a f.

There was a week long holiday in China. Everyone is just getting back to work.

Most flashlights are easy to take apart. I’m sure someone here can walk you through opening up a D4 if you ask. It is probably somewhere in this thread…

Username DOESN’T check out :person_facepalming:

Yeah, not sure who that is, but it seems the account was created just to complain about the guy who runs Emisar.

What Hank said was probably correct, and is likely a useful answer. Sounds like something broke, and it’s one of two types of electrical shorts. It should be relatively easy to fix.

To add to the list of possible reasons a light would fail to respond to clicks and exit turbo mode, my friends D4 did the same thing after I accidentally popped off the R5 and R4 resistors while disassembling for an Anduril flash. We managed to get R5 back on, and we thought we got R4 back on as well, but it only looked like a good connection, actually it was open circuit.

With R4 essentially unpopulated, the light seemed to run fine, except it would not enter any strobe modes, and once you entered turbo, the only way to exit was to disconnect power. Very strange. According to zeroflow, “R4: MOSFET current-limit”. So this would seem to make sense?
Who knows. I fixed the connection properly and it’s back to normal now. Just thought I’d chime in with my experience.

Well, i couldnt find. But were there any torx screw? U sure remember!

d4buyer. Ok? Cant remember password.
unintended connection somewhere, allowing power to bypass the driver entirely. - how is it easy to fix? I am just an office worker, so i am trying to get answer, as hank is silent for over two months. Just got tired of nokia as a light.

“just an office worker”, we all started somewhere, the first step is to realize you aren’t “just” anything… your potential is unlimited. :wink:

Its good but no time