Teardown and Mod Thread of Courui XML2 Aka "Big Head 3x18650 Side-Switch Thrower"

If the heat still has to go through the stock threads it really won't matter much what the plate's made of or what's added to it.

Don’t worry. I’ll be posting my rebuilt pill. That little copper nugget is just the tip of the iceberg.

Hmm, I decide to come back on and the first post I read has this ‘Modder’s Dream’ in it… nice. I guess I need to polish off my international credit card :wink:
Anyone measure the throw with a dedomed XML-2 U2 1C in there? That reflector really looks like it should throw, and the beam shots show it as well.

i ordered mine from kaidomain and it’s taking for ever to arrive, its been close to 4 weeks and still waiting

I meanwhile got mine, too, and just did a driver mod like LinusHofmann in #98 (desoldered all components from the board to simply use it as contact board, and get in a NANJG105C 3.0A with custom firmware) and of course used better wires. I'm quite happy with it. My Skilhunt K30GT (firmware modded) blasts it away though :)

Hehe not surprised there! :slight_smile:
Out of interest did you also shorten the centering ring to focus up the emitter better? It’s really far off optimal on these when stock.
I’d be really interested in seeing some beamshot comparison photos between these two lights if you get a chance.

No, I found the beam ok as it was.

Oh that’s interesting maybe the tall centering rings aren’t a problem across the board. If you get the light apart again I’d be interested in a measurement of the emitter centering ring to see if it’s a shorter one. Wouldn’t surprise me to have some variation on these parts.

Was anyone ever able to get a decent increase through resistor modding the stock driver?

I Added an R100 resister and got 4.1A. I could smell some burning at first and saw some smoke curling out the top of the pill. I don’t know if it was just old flux burning off, but I used it for a few days like that with no problems. I intended to use the stock driver as a contact board anyway, so I waited to see what would happen.
Nothing else happened, but the board has since been stripped of components. You could try running it for a while outside the light and see for yourself.

If I bypass all the sense resistors will I get direct drive then?

How do I do a tail-cap reading on this light? I don’t know where to place the leads.

Remove the dinky little wire, place the leads on the solder pads where the wire used to be.

OK, I’m seeing a few places to order this light from, which is the best mod host so far?
Anyone order and receive the DX model yet?
Is ShenZhenWholesale the only seller with an unbranded model?

KD is slow i ordered 2 of them over 4 weeks ago and still nothing

I got my KD COURUI real fast. I ordered another and it was shipped right away. I expect it any day.

Ok guys, I finally have a lux meter so I decided to redo the runtime test for the modded courui (before heatsink improvements) to get an idea of the amount of thermal sag evident in the output.

The meter I used is one of those cheap red HS1010 ones that are good up to 200k

So here’s the graph.

I’m new to taking lux measurements so I just left the figures as I took them off the meter.
Max values after a few seconds read 975 on the 10x setting of the meter so I guess that’s 9750 lux measured over a distance of 3.5m.
Using the convertor at ledrise.com and an angle of 90degrees (read that on this forum somewhere but no idea if this is right for all lights or depends on beam angle?) that comes out at a 119.4kcd or nearly 120k lux at 1m.

Please correct me if I’m wrong on any of this, still a bit confused by the whole thing at the moment.

….

In any case the raw numbers should tell a fair bit of the story by themselves. I’m seeing a fairly steep drop in lux right at the start that levels off after about 2 mins. At the end of that steep 2.5mins decline the output has dropped from 9750 to 9260lux which is about 5%. Not insignificant but I was expecting more tbh.

Considering that the regulation is in full swing at this stage and the current is locked at 5.81A for most of the first 5minutes, this drop in output can only be attributed to thermal sag.

After that the driver drops out of regulation and it’s no longer possible to see what effect the temperature itself has on the output.
I’ll see if I can hook up a power supply to this light and get a fully regulated 10-15minute runtime test done and see how far the output drops over extended periods at full power. Again just to have a good baseline of values for when I come back and test it again with the improved heatsinking.

PS. Looking at this chart compared to the one on my last test, it looks like the light drops out of regulation a fair bit sooner here. Everything was setup the same as last time but I may not have fully solved my recent flickering/bad driver grounding issue and this may be a symptom. In any case it doesn’t effect this test much but I’ll look into it and see if I’m burning off voltage anywhere I shouldn’t be.

Great work!

Can anyone suggest me a reliable source to order this Courui? DX, KD, or Lightmalls?

I’m happy with my KD version - came with a thicker 25mm aluminium LED base plate, & didn’t have the obstructive centering ring.

….still haven’t got around to modding it yet…