BTU Shocker Mod - XM-L2 U2's/SinkPAD's

Oh boy, where can I start... This is/was a scary mod - it's difficult taking on a mod job on a light you've never done before, and it's not yours! It's someone else's pride and joy, and their investment, oh - and it's the biggest, most expensive light you've ever worked on, and also, your first multi-emitter upgrade you ever did -- no problem . So what do you do?

  1. the plan - only doing an emitter upgrade, that's it. Well, upgrade the wires to something heavier, better, and XM-L2 U2's on SinkPAD's. Word is the thicker, bigger stars should fit fine. Take pics along the way - the pics, as it turns out, will be your reference (but I never take enough pics! Always forget).
  2. Tear It Apart! Well, the driver assembly unscrews - bad thing - you'll have twisted wires (remember this later). So, it comes apart easy, 1 phillips heard screw in there to hold down that massive reflector, take the screw out, unscrew the bezel (real nice threading by the way on the bezel), pull out the massive, heavy reflector (did I mention the reflector is heavy? Again, remember this for later!!), unscrew the collar that surrounds the reflector to get better access to the LED's
  3. Desolder the emitter wires - funny, why are the main wires from the driver thinner than the short wires between emitters? (think I found out why - twisting heavy wires doesn't work well)
  4. pop off the epoxied 16mm stars - left some scratches, but surface will be sanded/polished anyway later, though should have been more careful
  5. Now sand/polish the pill top to 2500 grit, nice and smooth
  6. re-flow the XM-L2 U2's onto to the SinkPAD - old frying pan method, works great again, test those reflows - looks good!
  7. sand/polish the SinkPAD's to 2500 grit
  8. Plan on mounting the emitters - I believe AS5 is the best way to go, not epoxy. So, after they are wired, you can freely position/align them. And that super heavy weight of the reflector that's screwed down with one center screw will firmly hold them, maximizing contact for thermal xfer. Will there be twisting pressure, putting the wires at risk? HHmm, maybe? So use epoxy on the outer edges of the stars to help locking them in to position.
  9. Position the stars by using the reflector, and mark the positions - perm. marker on the actual pill top. This doesn't have to be precise because these will be on AS5 later, and will be moveable to a certain extent.
  10. All the above was the easy part, now comes the wiring... Plan originally was teflon coated 20 gauge, so soldered to the driver (oh - detailed pics of the driver showed me where to solder!!). I was thinking I could avoid wire twists by using the driver wired, screw the driver in, the after, solder the wires to the stars -- No, stupid, that won't work, because the driver has to be out in order to get to the screw that locks down the reflector! Ok - twist the wire then - No! Teflon coated, silver tinned wire doesn't twist well, pretty much not at all -this won't work. So, delay the job while ordering silicone 20 gauge and 22 gauge wire (use 22 gauge if 20 gauge doesn't work out).
  11. 20 gauge silicone is in - de-solder the teflon wire, solder in the silicone wire - use kapton tape to wrap the red and black wires close to the driver as a strain relief because of the twisting pressure on the actual solder connections.
  12. So all set, leave enough slack so the driver can be pulled out enough to get to the screw, set the stars in the AS5 in the marked positions, cut/solder wires intereconnecting the stars (use your pics as reference again!), then solder the 2 main wires. Ok - this soldering is not so easy - you are working vertically down, into the pill top, and for some reason, it wasn't easy to get good solder connections - maybe it's this new wire, not sure.
  13. Now all soldered, but you need to protect the + and - solder points because the bottom of the reflector is flat and will ground the wire connections. Now I chose to use kapton tape here. Also I re-used the XML alignment rings, but they don't elevate the reflector much off the star, so the kapton tape does it's thing. Unfortunately I don't have any of those XML isolation glue-backed rings that are built for 20mm stars, only for 16 mm stars. Do they even make them?
  14. So now you are set to screw down the reflector, then do several CCW twists of the driver in order to CW screw it in, so maybe twists aren't too bad.
  15. All done? no - assemble the bezel and collar around the reflector -- ooops, something is wrong! I secured the reflector, but it's impossible to get the collar down around it! Crap - the collar has to go on first before the reflector, duh. So, take it all apart, now screw on the collar, and now drop in the heavy reflector -- wait, this thing is damn heavy and I got nothing to hold on to - I gotta let this 5 lb hunk of aluminum drop like a 1/4" onto the emitters? Well, let it drop carefully, and precisely. Did I mention this now $150 plus light is not mine? Well if I could put my greasy fingers into the reflectors, no problem. No - not an option, so, let her drop!! I did, and it seemd to work - positioned it, screwed it down, CCW'ed the driver before screwing it in. Now, easily screwed on the bezel with the glass (AR in this case).
  16. Fired it up - wholla, let there be light, and a whole lotta light!

If you suspect the 20mm SinkPAD thickness may be an issue, I dont' see it - the reflector may be raised somewhat, but there are so many threads on that bezel, and the way it seats, I don't see anything noticeable. I really like all the threaded connections on the Shocker - seems to be quality and lots of threads.

HHmm, did I mention the reflector is slighty over-weight? I really can't understand why - don't see how it could be used for heat sinking..

PS: I'm sure I left out some details (and mistakes) either forgetting or too embarassed to admit .

^
This is basically what I did too, with one change for Step 16.
16. Reverse polarity the driver on a bench power supply and smoke some unknown components, thereby rendering your stock BTU driver a one-mode, DD, glorified piece of wire. :smiley:

Well I got my BTU in. Awesome job Tom! The light is now even more of a beast. So I topped off my batteries (Keeppower 3400’s) and did a full set of tests. The results -
———————————————————————————————
Stock -
Turbo (start) - 3050
Turbo (30sec) - 2880
High - 2160
Medium - 928
Low - 68
Throw (from 15m) - 124k
————————————————————
Modded (UCL lens, XM-L2’s on copper, etc.) -
Turbo (start) - 3640
Turbo (30sec) - 3670
High - 2520
Medium - 1025
Low - 75
Throw (from 15m) 181k
—————————————————————
So my numbers match almost identical to Toms. And yes the the light actually gains slight power at first before leveling off and doing a slow drop. MUCH nicer to see vs the 170 lumen drop it use to do.
Also recently picked up a Extech EA31 meter. A little higher quality (and priced) meter. Got it mainly because it can read down to .01 lux which will help on long distance throw testing. Anyway, initial testing on the BTU read pretty consistant with my old meter (and cheaper). So basically between two people, two lightboxs, and 3 meters - I’m pretty confident these numbers are pretty spot on and correct.
We’ll be taking the BTU out tomorrow along with some of our other big guns for some long distance beam shots. If all turns out OK, I’ll post them up on a new thread.
And lastly (and again) THANKS TOM FOR THE SWEET MODS!!! I LOVE MY BTU EVEN MORE NOW!

Nice results rdrfronty. When you get a chance, can you take a closeup picture of the reflector surface? :wink:

Sure thing. I’ll try to take one in the morning and post it up for you.

After 30sec in Turbo mode the brightness even increased?!

Yes - I've been seeing this on some lights, some builds. I'm thinking maybe it's the battery carrier springs or the battery carrier assembly because I didn't beef up the springs or do any mods on it. It's like in a warm up phase, then maybe 45 to 60 secs or so into the run, then you see the gradual drop, but those first 45-60 secs, it's rising. Maybe the increasing heat (and/or current) is producing better contact initially?

Quiet - hope this doesn't pi$$ too many owners off, but: Oh btw, Ric says he will be shipping AR lens in the Shocker, due to arrive to him in a few days.... finally!

No one can read this I hope ...

Damn I just ordered one from Ric! I am gonna ask Ric about this!

Tom E, I wish to seek for your advice on how to remove those LED stars on the heat sink. Currently I am going to repair(and mod at the same time) my DRY and those LEDs are being glued on the heat sink and they are tight. Using what method can I remove the stars?

Yes - so were the Shockers, and did several other stubborn ones. Well think about it - damaging those stars? Not a problem - you are tossing them anyway, alum stars are literally pennies. If you have a thin bevel screw driver (good one), that would do it. You gotta take your time though, dont' rush it. Seems like that epoxy will give up after timed pressure. Never tried heat (ex: heat gun), but maybe that would help? I use a tool, cheap one (maybe from DX) that on one end has a flat tip with a pretty narrow bevel edge and I usually use that first to get some lift on a corner (by the cutout for the leads). Again, I don't care that I bend/damage the star, actually i want to get it bent up. Once that is started, then a heavier screw driver pretty much always works. On the Shocker, I did damage the pill top somewhat, but the sanding/polishing flattens it and the any heat xfer loss is probably nothing because it's not near the center of the star anyway.

On conventional lights with stubborn stars, I'll even take a punch and hammer from below through the wire lead holes, totally mangling the star.

I'm sure there's better techniques guys are using, but I kind of brute-force it.

We’ve also seen the slight rise on a few lights, not many though. Usually its on mildly driven lights though. I’m not in the least worried about it - I know all the components and work is top notch now that Tom’s went through it. And it does drop in power eventually, it just takes more than 30 seconds for the heat to start playing a picture in it. Toms use of 2500 grit sandpaper on the heatsink along with these great copper Sinkpads and quality thermal paste, and the BTU’s mass all make for excellent heat dissipation.
I actually think the amps were slightly building to the LEDs also. I know some report a fluctuation of 3.8 down to 3.6A on BTU’s. I’m thinking, might be wrong, that with the beefier wires Tom put in that mine is possible building UP to the 3.8A and not going down like many stock ones do. Basically the better wires handle the full 11.4A the driver puts out on turbo. Another reason I think this is because when I tested it on high, not Turbo, with its 2520 lumens it did have a slow drop. Perhaps 20 lumens in 30 sec or so. So that sounds like heat does eventually get to it. Though slight. And that likely the driver builds amps very slightly in the first 30sec on turbo. That’s the way I’m thinking. Don’t Or it could be the springs like Tom suggests.
Of course with the poor thermal paste setup stock, it still dropped 170 lumens regardless of if the amps climbed to not. So I’m VERY happy with its setup now.
As a side note - how cool is it to have a light that kicks DOWN to 2520 lumens? Not too shabby. So this BTU can run at over 2500 lumens, never having to touch a button, until the batteries are drained. And I truthfully wouldn’t hesitate to kick it on its 36-3700 lumen turbo for as long as desired. I mean I ran it 21 min.non stop on turbo, uncooled, sitting on a table, when it was stock and had a fairly poor thermal setup. And it did Ok. Now it would do awesome. Would be absolutely no danger to the LED’s.

I can see your point there, but after using brute force to remove the stars there are still some epoxy sticking on the heat sink, how to remove those leftover epoxy or is it not necessary to do so?

And another question, since I will change those LEDs into U3 (XML2 U2 are expensive for now), how can I make sure that three LEDs are sitting in the right place in conjunction with the reflector? Do you do some marking on the heat sink before removing the old LEDs?

- I'll get the bulk off, maybe finger nail, maybe a tool - but try to avoid scratches there, iso. alcohol, then let the sanding begin! Maybe 120/240 to start, slowly going up (400,600,800,1000,1500,2000) to 2500 grit. Epoxy is gone with either beofre or from the first sanding step.

- Oh - cost all depends, but Craig at IS has the best price on XM-L2 U2's for $7.50, oops, you're overseas though. Maybe +$15 for the XM-L2 U2's over the U3's? But U3's on SinkPAD's is not a bad route either, specially if they are 1A tint. But for me, spending all the time and $$$, I'd go for the XM-L2 U2's - you don't want to do this very often...

- Step #9 - marked the positions, doesn't have to be super precise because you'll have some play by using thermal grease, not epoxy, but I did use a little epoxy on the one outer edge of each SinkPAD to lock it in, somewhat.

So using sand paper can remove those leftover epoxy, hmm.

Kaidomain does offer XM-L2 U2 with $9.49 free shipping, but it is on aluminum star, not Sinkpad. Anyway I still don’t know where to get the Sinkpad, probably I will just go with the easy route - U3 1C on aluminium will do, maybe I’m just too lazy to source those better parts :bigsmile:

Easy to remove LED's from alum starts, then reflow onto SinkPAD's - done this many times.

fyi.. straight U3/aluminum isn't gonna be all that much of an improvement. I wouldn't do it for the work and risk factor, but it will be some good experience, and if you have the time of course.

Actually I have to do this because my DRY is broken badly, so instead of calling this as a “mod” it is more likely a repair, lol.

Reflow is something like heating the heat sink and let the thermal paste to melt on it? Anyway I do not own any Sinkpad…

Oh, ok. Reflow is soldering the LED onto the star board, but using solder paste, and applying heat to melt the solder paste, then the emitter gently sits down in the proper position with a bonded (soldered) connection on the pads of the star board.

So a DRY is a different 3 LED light? I know the DRY driver is an option for a Shocker. Ok - think I understand now, yes, for a repair U3's maybe a good cost effective option then.

My method for removing a star on a heat sink.
Use a strong pair of short needle-nose pliers.
Insert the tips into the u-grooves of the star.
Twist while pushing down just enough to hold the tips in the groves. After a few seconds, the star will let go.
Usually the star is undamaged and can be reused.
Sometimes you may get a small scratch on the pill. Since I sand/lap the pill anyway, this isn’t a big deal for me.

Here are a couple shots of my refector. Best I can do with my iphone. There are a few little ripples, especially close to the LEDS, but obviously not enough to hurt my throw much.


.