The bezel O-ring is very thin, so the seal can be easily broken. My white D4 didn’t even keep water out when it was new; it leaked through the bezel (not the lens). I fixed it by swapping the thin O-ring for one of the spares included that are designed for the body. It’s much thicker and a tight fit, but you might want to give it a go; it could be wide enough to seal the uneven surface of a damaged bezel.
I diassembled my D4 to flash new firmware, now it dims when i turn it off. I think the circuit is short somewhere, what is the best possible cause?
Update: after some test with multimeter, i found that the flashlight aluminium body was short circuit with the negative of mcpcb. I also found that the noctigon mcpcb is made of a thin top layer of copper to conduct electricity, and the thick copper bottom layer. My screws somehow connect the top and bottom layers. But the screws are also used in original D4, why does it cause problems now?
The flashlight Mod God’s yankin your chain… they’re funny that way.
I agree with everyone that suggested talking to Hank. He’s astoundingly accommodating to everything and anything.
Is there any option to configure the temperature sensor as in Anduril? My D4 marks about 15º more than the ambient temperature
Nope, not without changing the firmware. RampingIOS V2 doesn’t have a sensor calibration function. RampingIOS V3 does though, and so does Anduril.
Did you lose the insulating washers? There should be two washers per screw, a metal one and an insulating one. The insulating one goes against the mcpcb, and the metal one on top against the screw.
Thanks TK. Anduril works good on D4? time ago I read a thread where it was said that the temperature sensor of Anduril didn’t work well in D4.
Works very well on the D4. I have Anduril running on three D4’s. As long as you set temp limit at a fairly conservative level (mine are at 45C-50C). Then it never gets too hot.
Conversely, I have Anduril on all my D4’s, and I set the temp to maximum and have no trouble. TK fixed the old thermal problems, or at least put in a workaround. It’s called something like “THERM HARD DROP” I can’t remember exactly. But it drops the output down to a sane level after the temp starts climbing too high. Works real well, if maybe a little overzealous. So if you want extended turbo times, you’ll need to set the maximum temp higher. In my experience anyway.
Yeah, Anduril works fine on the D4 now. It’s just a bit more cautious than necessary when it’s in turbo mode. It’s treated as a burst mode, and as soon as it looks like it might be getting hot, it ramps quickly down to a level which doesn’t heat up so fast.
Thanks to all, I’ll buy an USB ASP Programmer and I try to install Anduril to see how it works.
Right you are.
- I don’t have a use for extended turbo bursts.
- I use a D4 for work. And regularly use it at “medium/high” levels, where the temp control keeps regulating. And when I’m done working I like being able to touch my light, not burn my fingers
Yes i did lose the insulating washers, the washers insulate the top layer of mcpcb from the head of the screws, but, the threads of the screws can still connect the top and bottom layers of the mcpcb??
Maukka…
In another thread I mentioned how the driver seems to be acting as a current limiting device. Though in theory a FET driver shouldn’t behave like that. Kiriba-ru questioned that.
What do you think about that?
Sounds like your screws may have eroded and worn through the top layer of the MPCB causing a short. Some options:
- Option 1: Try rotating the MPCB 90 degrees. The screws would then go in the holes previously used by the optic. Those screw holes are presumably not eroded so hopefully you should not get a short.
- Option 2: Omit the screws. The first run Emisar D4 actually did not have any screws holding the MPCB on. All you had was pressure from the optic and bezel holding the star down. The main purpose of the screws is so that when you tighten the bezel the star doesn’t rotate and possible break the solder on the driver wires. That said, many triples, quads, and single emitter lights work just fine without any screws at all, and the D4 is one of them. Since the screw are causing your light to malfunction, I suggest omitting them.
- Option 3: get a replacement star. Any 24mm standard quad star will fit. You can buy one from Moutain Electronics and reflow the emitters onto it.
Thanks for your answer. But I don’t understand what do you mean by “eroded”, because the screws holes are all metal so the screws can connect the top and bottom layers of mcpcb easily and no need anything to be eroded?
The copper substrate of the MCPCB is electrically separate from the PCB traces that power the led. When the screw wears through the red layer it may connect the traces to the body causing a short
I got a PM telling me that a resistor is a standard way of slowing down a FET. Quick search:
Yeah, it is. So it seems that a FET used this way is actually a high-ripple regulator.
But…I tried to use your scope results to actually measure the ripple. Now that I think about it, not knowing your sensor’s response curve it was a broken idea. But I got something I don’t understand - and I’d like to.
I see that at fairly low duty the peaks reach about 1.5V. But then they get lower and Turbo voltage is below 0.5V. So…I would interpret it as D4 being much brighter in peaks than with a FET wide open. I don’t believe that’s the case.
Any explanation?