D10 easy emitter swap.

If anyone is looking for an easy emitter swap, here it is. I ordered an XP16 with an XP-L HI (part# L024) when I got my KR4. I also got some 95 CRI unmounted SST-20 but I have to get myself some solder paste before I can try those on my 12mm mpcb’s. The XP16 fit nice and tight after some sanding since the edges were tapered. So I think that is the better solution. Otherwise I would need some way to hold the mpcb tight against the pill, so would need a gasket or a thermal epoxy. I forgot to measure the reflector ID. The height of the XPL when everything was tightened up was perfect already with no extra parts anyways. I think the wires are what is touching the plastic washer at the reflector when everything it tight, so I may try some kapton or a thinner washer to experiment. You still need something, or else risk the wires shorting to the reflector. Probably the last time the driver would ever work, the probably didn’t design it with a dead short failure in mind. The Piston has plenty of travel still. The light still ramps properly. Which is annoyingly slow… I really want to make my own driver that can run Anduril. I need too so some board design again, been doing industrial automation too long.

OEM XP-G, Efest 14500. Wasn’t the D10 R2 supposed to have an XP-E? Did I get an old version relabeled as an R2?

And the 5000K XP-L HI, same Efest 14500. Gotta fix that gutter. Shitty pine tree came down over the winter. I would rather take leaves than have these stupid house killers.

congrats on the mod
I agree it is the wires that are pressing the mcpcb to the shelf

here is how I solved that on one of my Nitecore PD:

(I sanded the kaidomain spacer down a bit, and opened up the hole to clear the LED)

however, since your D10 (is it actually a D11.2?) came with an XP LED, I would have just reflowed a 219b onto the stock mcpcb, and reused the stock centering ring, since the mcpcb is XP compatible:

(this is a D11.2 driver that came with an XPG, in a Hanko D10 host)

you can tell if you have a D11.2 driver by double clicking from off, which will produce SOS… there are no blinkies on the D10 driver.

Thanks jon. It is a D10, supposed to be an R2 but it obviously has an XR-E emitter just like what you have pictured. Cree ring and all. It says Smart PD D10 R2 right on the main body. Maybe they mixed up the heads at the factory and I got a non-R2 head? It was sold to me by a reputable vendor as an R2. They may have screwed up the body revision, too. The head is where the change happened, but the model is of course silk-screened on the body not the head. So pretty easy to screw up. Some companies have absolutely no revision control. Don’t even know such a thing exists, sounds to hard, etc.

Anyways, that spacer looks similar to the ones that Convoy sells, too. I have been thinking about that as a solution but I forgot to measure the reflector opening. I will have to take it apart tonight. The Convoy one is for a 7mm opening iirc.

I did not rely on the centering function, only the shim between the mcpcb and the white plastic ring at the reflector, to take pressure off the wires

I actually sanded both sides of the spacer, plus I enlarged the opening to clear the base of the led…

so the convoy spacer w 7mm opening will work as a spacer too.
lots of choices… any shim will do, we just need the reflector to press on mcpcb, not just through the wires :slight_smile:

respect for taking it to the next level
even though it works as is, I agree it is better not to rely on the wires for pressure.

Ha yes not crushing wires is important. Electrons love it when you divide by zero. In this case it’s low risk but it just doesn’t sit right with me:). What I did temporarily was just snug it up, it’s not very tight. The mpcb is tight enough in the pill that I don’t need the pressure from the reflector pushing the mpcb against the pill for heatsinking. But I do have some other mpcb’s for some SST20 I want to try. I am thinking that the driver in the ex11.2 might be best with the XP-L as a pocket rocket, and use a 4000k 95cri SST20 in the D10 with the weaker stock driver. To many options I can’t decide!

I like exploring options

fwiw,
the EX11.2 driver, on LiIon is completely unregulated and drifts upwards by over 100%… for me completely unacceptable. Works fine on CR123

what I like about the 11.2 driver, is that it has a 0.01 lumen minimum

the EX10 driver, with a 4500k 219b makes over 200 lumens on 16340
and it IS regulated output!

question… have you ever used 14500 in a D10?
I have only used AA and Eneloop.

I am using an Efest 14500 in the light for those shots I posted. It gets nice and warm on the 14500 but not hot. I am waiting on my Ikea LADDA cells still, then I will get some shots with those. I think the output on the 14500 before the swap was about where the output is now with an alkaline. I plan on using the NIMH for a daily carry since the Efest is unprotected. I have been twisting it off but it’s only a matter of time till I forget and the parasitic off current kills the 14500. Another reason to try making my own driver!

You could try asking Quadrupel if he still has some drivers on hand for the D10. He offers them with a Zebralight-like UI or Anduril.

Anduril would be awesome on a piston light. Thanks I will have to message him.

Just in case anyone is wondering, with the Efest at 3.8V it is drawing 1.2-1.3A. with an alkaline at 1.4V it is drawing 1.5A after this mod. On lowest output both were drawing maybe 30mA. I had to leave the DMM in the 20A range so the 30mA measurement won’t be too accurate.

thanks for all that info
Do you have a way to confirm if output is regulated on LiIon?

great info, did not know
here is his metion of Anduril and Zebra drivers

I have nitecore D10 Q5.
I want emitter swap with stock driver.

What’s the best between 219c 4000k or sst20 4000k.

congratulations on planning to get a High CRI LED

between those two, the sst20 produces more Red spectrum, so that is the one I would pick. Red Color Rendering is a priority for me.

the one I like best, is 219b 4500k, because the tint is pink, instead of green for the others you mention. NonGreen Tint is a priority for me.

If Lumen Quantity is your priority, and you dont mind green tint, consider the LH351d 5000k, it is about 50% brighter than the others, it also has about 50% less Red output (similar to 219c that is also green tinted but lower output).

the following charts are all from data posted by maukka… you can find it by google using the LED name and his handle, like this… “219c maukka” etc

focus on the red bar (CRI R9), and the duv (tint):
sst:

219c:

219b:

LH351d:

for reference, duv of 0 is the tint of incandescent, above 0 is green tint, below 0 is pink tint,
duv of 0.0032 is the tint of sunlight

looking at the duv of the charts, the LH351d has the greenest tint, while the sw45k is the complete opposite

the sst and 219c both have the same amount of green tint as sunlight

so for a light used during the day the green tint wont be obvious, because the brain adjusts itself to ambient light, and when we turn on a flashlight, the brain is still using the environmental reference… hope that makes sense

since sunlight is CCT 5600k, using a 4000k LED during the day is not ideal, as the lower CCT will look relatively warmer (“yellower”). This is why I prefer the 4500k and 5000k options for daytime EDC.

Thank you very much.