Did you checked BEFORE installing the board with the new LED if there was a short between + or - and the MCPCB ?
Since itās DTP, it could have shorted the driver because the body/head of Olights is connected to the battery + !
I already fried an S1RII driver because of that, and now it has 4 modes : Ultraultralow, ultralow, verylow and not-bright-Lowā¦ a good childproof Olight 
LOLOL, you made me laugh with those modes Sorry for your trouble too mate :zipper_mouth_face:
Well, I didnāt check it before, no 
I did what I normally do on flashlights, I unsoldered and soldered normally (apart from that eventual longer time over one of the wires on the first unsoldering attemptā¦), and only after noticed the issue.
I never used the battery inside while doing the mod, but still it may have cause the short? :person_facepalming:
Well, the ultraultralow mode I get may be used as presence light, leaving the light ON all night on the bedside, just like a lighted tailswitch :person_facepalming:
EDIT: I should have stayed still and just put the GITD tape :confounded:
Donāt be ashamed, I recently fried my S2RII tooā¦ while trying to put an LH351D in it
Well, I never really liked it anywayā¦ or did I ?
Whaaa? 
Did it happen something similar? And no functioning at all?
Well, Iām just upset, Iād like to keep this one working fine thatās all!
I guess I should report it on āāwhat did you break todayā:https://budgetlightforum.com/t/-/40266ā thread xD
The funny thing is, I wish the moonlight on my S1 were lower.
Realizing an emitter swap is possible here, I'm tempted to try on my S1, but then this is a thread about how it goes wrong, so maybe not...
Well, it was also proven that it works, mine was just a bad example 
I guess that the soldering iron will make the difference. A thin tip and a being hot enough to unsolder without having to touching for a long time may do the difference.
The more I think of it, the more I remember the beginning of the process yesterday, the more I realize that it may have been my mistake. It happened something similar when I tried to disassemble the Sofirn SP10A: the heat in the wire made a component burn (or at least get damaged).
Soā¦get a good soldering iron with nice tip and ā¦maybe it work 
OTHO Iāve swapped every other Olight S series lights and neve encounter any problemā¦ until the S1RII/S2RII
Iāve swapped : S1, S1R, S10RII, S10RIII, S30RII, S30RIII, M2R, H1, H1R, and also made triple in some of them
Did you remove the pill/driver of any of them? If so, can you share how? Like, tiwisting it, or pressing from head or tail?
You surely have a nice Olight collection 
That's really encouraging. I'm much more likely to take a crack (hopefully, not literally!) at my S1 then. It's a useful little dude but the cool white is so angry blue, it's hard to carry at all. I'll admit though, when I do carry it, I'll clip it somewhere I can idly thumb the funny-shaped TIR when I'm standing around. I actually rather like that TIR's beam.
Iām not trying to say you werenāt careful with the mod, but I know it happened at some point because I have several dead Olight drivers with identical symptoms in a box beside me. All it takes is the tiniest slip with tweezers while youāre doing something like positioning the optic. Iāve done emitter swaps on dozens of Zebralights with only a couple duds, but my failure rate with Olights was basically 100% until I realized just how fragile their drivers are. The cell stays out of the light unless there is no possibility I will be touching the MCPCB.
As far as I know, removing the driver is the same as outlined in X3ās S1 Mini teardown. I have disassembled quite a few Olights and it isnāt a very fun job. Unless you are fairly competent at diagnosing electronics issues there wonāt be anything you can do with the driver. From the way the UI behaves and the separate moonlight failure mode, moonlight is distinct from the rest of the modes on the driver, but I have no idea if itās part of the IC failing or some easily replaceable component. Personally I would just put it back together and accept it no longer has a useable moonlight mode. Itās definitely the driver, not your choice of LED or anything like that, and the damage is likely permanent.
I actually posted a cautionary note about this in the same S1 Mini thread:
Hey Bob, I was probably careless with some stuff :zipper_mouth_face: :person_facepalming: I am a bit messy, specially when soldering :zipper_mouth_face: But one thing I know, is that it wasnāt due to the cell, because it was out of the light!
Also, sorry for your dead drivers, not a nice situation, specially being these good and expensive flashlights!
Thanks for pointing me that thread as I hadnāt look to it before! I will do it, but I will probably not venture on opening the light! I will stay as it is, for the good and the bad.
MEANWHILE, I bought a small multimeter (Uni-T UT33D), nothing fancy but it may help solving the case. I just donāt know how as I never used a MMeter before :person_facepalming:
ALSO, when I arrived from work and went to test the light with the MMeter (I donāt know what to search so I started touching thingsā¦. LOL) accidentally one probe touched the POSITIVE wire (solder) and the headās wall (bare aluminium) and it lit up on the true moonlight.
SO, I guess the driver is not damaged as the ML mode works, but something is not making the right connection. When I touch both the wall and the + wire, the ML mode activates. I tried to put solder from the wire to the wall but that gives a continuous ON even the light isā¦OFF.
But now I know the driver may still be OK, just any connection is not being done correctly.
What do you think it may be?
And will the MMeter be useful to detect it? If so, how?
I am sorry for so many questions And I thank for your information and help in advance! :+1:
So, when I pick the multimeter on the ādiode testā (?) and put the negative probe in the negative wire / positive probe in the positive wire, the led lights up (as it should).
When I put the positive probe in the positive wire and the negative probe on the wall of the pill/head, he multimeter gives a sound and some values appear in the screen!
Does it mean I have a short somewhere?
I really donāt know what to do diagnose thisā¦ 
EDIT: I tried the same test in other lights and they donāt act like this, namely on the second part of the test.
On those Olights, the head is battery+ which is also what the led+ gets. Then led- goes to the driver, and batt- comes out. So led+ should have continuity to the flashlight body on that particular light.
Most lights, batt+ basically bypasses the driver and goes straight to led+, then led- to the driver, and batt- is the body of the light.
This is because of the type of power regulators we use in drivers. There's also some details glossed over here, like the driver leeching some power off of batt+ that goes through it to batt- and never to the led.
If you can still work it with the bezel/reflector/etc. off, try pushing down on the LEDās dome lightly with a fingertip or eraser-end of a pencil, see if it goes to normal brightness.
I had an issue with a TK05 where it āworkedā (all modes, etc.), just at about 1/10th the brightness. āLooseā(??) LED? Either way, reflowing/resoldering fixed it.
Thanks for the clarification Scallywag! I tried some stuff but itt didnāt help after all! Still, now I know how it works
Thanks for the tip Lightbringer.
After testing and trying some stuff, last night I decided to call it a end! Pressing the led or the board didnāt help, the only thing that made contact were the tweezers but on the wall.
So, I decided not to invest more time trying to fix it, I replaced the XML2 in the DTP copper board for an XP-G2 (S4-3D, apparently) and put everything back on place and closed the light.
So that you know, the ML mode is lower than the Reylights have (I have an Amutorch SS light with that driver and compared it).
Despite I donāt have the 0.5lm moonlight, I can live with the current one. And the tint of the XP-G2 will suit me better
Thanks for the help, the information, the time and the tips! All are much appreciated
After emitter swap, the moon mode of my S1R is slightly higher now. Hard to tell. It is higher than moon on my S1RII I think. I guess itās just forward voltage. Hopefully. Still ok to preserve night vision, and much nicer due to the warmer tint.
However: I noticed something similar in my RRT01 2019. When on lowest position, there never was any light before emitter swap. Now there is, even if only visible in total darkness with eyes adapted to it.
I saw your mod and was going to comment and to congratulate for doing it without the flashlight ātrippingā. Butā¦it seems youāre facing some issues too! :zipper_mouth_face:
Unlike mine, your moonlight went higher, but Iām glad it still works
And giving that it is warmer tint, it will look less brighter than if it was a cool tint.
There may be something happening on these drivers when we mod the lights.
BTW, on mine I am having issues with Turbo and Turbo S, it doesnāt hang on there as it used to :zipper_mouth_face:
Even with the battery fully charged it decreases fast. I am not sure if it is something related to the driver or the LED, but it only gets there for few seconds and then decreasesā¦.
Hope youāre more lucky than me
I came back to your thread because you linked it somewhere.
The S1R driver is not good. If really everything is ok, the battery new and the light warm, Turbo S sustains 1ā30 or more (my timer is off, giving it about 1ā40 or so). The slightest amout of dirt, cold battery or full moon, and the light behaves like you describe. A Keeppower 700mAh protected cell gives better results but is quite long. At some point, I removed the collar of a 10C cell for the S1RII and use it in the S1R since. Now the turbo modes are almost always stable.
I guess the trouble in mine was that I am now using a non-DTP aluminium board, 3535, instead of the original copper one. I am not sure if it was DTP, but I am guessing that it is making the difference in the capacity to hold a longer time on Turboā¦
It was a tricky mod from the beginning so I guess I downgraded the light except in terms of tint :zipper_mouth_face: In any case, I think Iām not going back to the original one. I would like to find a 12 to 14mm diameter board, 1mm thick to use a 3535 footprint led, as I donāt want to shave/sand/fill a larger/thicker one.
Thanks for your report Unheard
From my understanding - though the testing happened before my time and Iād be glad to be corrected if I have the wrong impression - MCPCBs are the primary place in a light where copper vs aluminum makes a measurable difference, and thatās even if theyāre both otherwise DTP.