Difficulty removing drivers from Emisar D4, D1, D1S

I definitely agree that Andúril is worth the effort. I put it on my Q8, and I liked it enough that I am performing (well, attempting) surgery on the Emisars shortly after getting them.

I’ve tried using a thin pair of steel hemostats with rounded tips to push on the driver, using a lot more force than I could exert with the wood skewer. None of the three will budge.

Looking at that photo of a D1S, it looks like there’s a film around the outer edge of the driver (where the vias are). You don’t think this was glue? If there is glue, I wonder if putting them back together and running them on turbo for a while might weaken the glue?

Hmm, I think you’re right. I didn’t notice that before.

I think your heat-up idea is a good one. Do you have a heat gun?

The drivers are glued. To get them out I used a very small metall stick with flat ends, passed it through the hole at emitter side onto free space on the PCB (NOT on top of components) and hammered it very carefully.

A plastic rod or toothpick usually works, with some careful tapping from a hammer or something. I’ve avoided using metal for this though, to reduce the chance of damaging the driver.

Just had the same issue.

My “older” D4 that has the first Noctigon 4XP V1 MCPCB, the driver came right out. Had to use minimal force with my plastic poking device.
The newer D4 that came with Noctigon 4XP V2, had the driver glued. I had to lightly hammer it out with my poking device.

As everybody else says, mind the thin white wires to the e-switch, they are short.

But now i have two “new” D4’s. One with NarsilM 1.2, and one with Andúril :heart_eyes:


I was able to get the drivers out of all three Emisars. I made all three of the changes below at the same time, so I don’t know for sure how much each of them contributed to the success:

  • I used a blow dryer to heat up the driver.
  • I carved a chopstick down so it just fit through the hole in the head - this was sturdier than the skewer I tried previously.
  • I used a pair of pliers to tap on the end of the chopstick.

Unfortunately, the D4 driver was damaged in the process - I broke off the R5 resistor. Having no SMD soldering experience, I didn’t have any luck reattaching it. After a few tries, I noticed that both the end contacts had came off the resistor, which means there was nothing for solder to stick to. You can see one of them still on the solder pad in the pic below, on the right pad.

I contacted Richard at Mountain Electronics, and he had a single spare driver on hand he was willing to sell me, so I will be getting the light going again soon. Richard mentioned that his MTN-22DDm driver works in the Emisars as well - good info in case anyone else breaks something removing their driver. Richard mentioned he had some difficulties getting the drivers out as well; he had to reflash about 150 lights from the first run of the D4, and he ended up with 4 non-working drivers.

The D1 and D1s are both running (a slightly tweaked) Andúril now, and hopefully the D4 will soon join them.

I’m curious. What did you change?

Very minor changes - nothing significant:

  • I wanted the ramping up from moon to start sooner, so I decreased HOLD_TIMEOUT in fsm-events.h from 24 to 16. This was based off the discussion starting here between you (TK) and goshdogit.
  • I wanted the max non-turbo level to be a little lower, so I changed both ramp_smooth_ceil and ramp_discrete_ceil in from [MAX_LEVEL - 30] to [MAX_LEVEL - 40]. (These are in the Anduril firmware file, roughly around line 165 or so, in case anyone else wants to adjust them.)

Congrats! I hope your D4 driver swap goes smoothly.

I’m still using 15 on several different lights without issue.

I set my D4 to –50, but I did it through the UI after flashing. Yes, I clicked the button fifty times. :smiley:

After using them for several days, I actually plan on lowering MAX_LEVEL down to –50 as well. I may re-flash the driver instead of getting the thumb workout. Even though I know that I can’t visually tell the difference between –49, –50, or 51, I’d worry I was off by a click and I want it to be right, dang it.

Yes, I realize that’s a bit ridiculous :stuck_out_tongue:

Not considering your audience! :wink:

So I've had the same problem here, gentle add I could be, and actually reading this post and a couple others several times, I popped off that same resistor.

It's only a 4.7ohm, though. Does anyone know what it does? I bridged where it was too see if it still works, and from very limited initial testing of "it blinked on battery connect!" And "double tap to turbo works" and then shutting it down because it's all tore apart with no heat sink.. it seems to work. Thoughts guys? Do i try getting a new smd resistor from mouser? What package size even is this stupid thing? Or do I just put it back together (flashed with anduril of course) and see what happens?

I’m not sure, but I think R5 helps to stabilize power and prevent spikes. With a full battery, try running at just below turbo (or try a tactical strobe mode), to see if there are any spikes big enough to cause problems.

On older drivers, pushing too many amps through the FET would cause so much electrical noise that it would make the MCU reboot. Part of the fix for that involved adding a resistor which is traditionally called R5. I’m not sure if this is the same thing or not though.

Thanks tk for the insight! I’m gonna flash it and put it back together this evening, I’ll report back with how it acts or responds. I know it’s not a guarantee that it’ll not fry itself, but I feel a little more confident now. Nothing left but to try it, if it burns? Guess I’ll get a TA driver lol.

So I’m late here, but I did get it back together, and I ran the D4 through it’s paces with a fully charged battery, and it seemed to do just fine. I didn’t notice any issues.

I’m sure like you said, it was there for a reason. But maybe the rest of my components were slightly higher in the 5% tolerance range or something.

At any rate, if I end up with trouble I’ll post back here. But for anyone who finds this in the future, who has the same problem as me… just solder bridge it, I guess? Mine seems to have ended up fine, maybe yours will too, is worth a shot.

You are right, R5 helps with voltage stabilisation. Probalby inrush current limiting and such stuff.

So this leaves you with 3 total options:

  • Solder bridge and keep your fingers crossed it won’t be a problem
  • Replace with a 4R7 0805 resistor
  • Contact Hank of intl-outdoor and buy a new driver for $6

As for other components which may be defect, each of these workarounds may cause problems with some being more dangerous.

  • D: reverse-polarity protection, DANGEROUS may destroy driver and battery, bridge as a workaround
  • C1: driver capacitor, ignore
  • C2: MCU capacitor, ignore
  • R3: MOSFET pull-down, DANGEROUS may keep the LED at full turbo at all times
  • R4: MOSFET current-limit, bridge
  • R5: Input current limiting, bridge
  • Mosfet: Unsolder, set upper ramping limit to end of regulated mode
  • 7135: Unsolder, set lower ramping limit to end of regulated mode
  • Attiny: Get a new driver

Informative! Thanks!
I think if I lost the attiny I would try to solder it back, even with the eight legs I feel like it’d be easier than that stupid tiny resistor, gah I spent so much time on that! I need a hot air station haha.

And I didn’t realize you could source whole new stock drivers, that’s good to know. I’ve got some TA drivers on the way from Lexel for my H03, looking forward to that.

Thanks for all the help guys!

Yeah, you have to know or ask, but Hank sells nearly anything if you ask.

For example someone else git a D4S Aux emitter board with red LEDs from Hank.

The same with Simon from Convoy on AliExpress. You can get any combination of host and driver from him.

This thread was helpful to me. I also removed the R5 resistor while pushing out the D4 driver.

Then it got worse, I removed the wire from the switch side.

Anyone got info on getting the switch out?

It’s super easy to rip that thing off haha. Glad it helped you! Thanks for the reminder that this thread is here, I meant to update it with my latest experience.

TL;DR: I’ve confirmed you can successfully bridge the R4 resistor and the light should be fine.

While disassembling my buddies D4, I managed to pop off the R5 AND R4 resistors, oy… we got them both back on and the light was acting extremely odd, no strobe modes would work at all, and if you entered turbo, you could not exit it without disconnecting the battery. Also occasionally it would fail to ramp out of regulated mode, just stopping at the end of regulated mode and staying at that brightness. Very weird.

Turns out the connection we made when resoldering R4 was no good. A little more fiddling with the iron and it was right as rain. Mostly…

I then ran into the problem wherein the light would act up in the same manner as it was when be R4 was disconnected, but only when it was super hot. I figure the connection on R4 would only fail if the board got hot enough to expand a small fissure in the solder joint.

By this time I had gotten my hands on some good solder wick and a tub of solder paste so I planned to fix it properly, but when I got it apart and got the resistor off, I dropped it into a large thick shaggy rug… never to be seen again…

So I just bridged it and prayed. Lo and behold it works just fine, ran it through all modes. So it’s seems I’ve confirmed you can bridge R4 as well with no issues.

I need to be more careful with these stupid things haha.

Edit: I didn’t see your remark about ripping out the switch wire, that sucks man. I’ve seen advice somewhere on how to get the switch out, I can’t remember where, look for D4 switch mods, that’ll get you on the right track.