If it ain't broke, don't fix it. 
Was that little bit of silicone seriously messing up the beam or something?
Sage advice Gj. I just wanted to see what condition the bond wires were in. Regretting that decision now as I really wanted to see how long the emitter would have lasted at 7amps.
Isn't Wasn't the silicone transparent? I don't see why you needed to remove it to see the wires. Just get a 10-20x loupe.
I talk about that in Post 79 above.
UPDATE:
It looks like removing the silicone was not the issue. The driver I tried out (TR-142B) likes to kill XML's. I ended up killing 3 more emitters with it before I realized the driver is just not xml friendly above 5.5 amps. I spoke to Moderator007 and that is his experience with this driver too.
So, now I need to find a source for that specific 3T6 driver Manafont was selling. It seems to be rock solid on its regulation and it let me run a 7 amps in the wild without blowing my xml2. It may have eventually fried the emitter, but it seems like a good driver regardless. I plan on replacing the xml2 and running to original driver in the OP (with emitter wire mod and potting). We'll see how long it lasts.
So, anyone know were to get this driver other than Manafont?
I ordered the 3T6 driver from wallbuys but it still hasnt shipped yet.
Thanks for the info Flash. That's the same serial xml killer (Moderator007's nickname) that FT sells. The TR-142B. Seems to have a lot of potential for MT-G2's though.
Well darn it. I guess I wont push it pass 5.5A. Thanks for spotting that for me.
Half-assed theory:
It uses a large inductor as part of the regulation circuit. On startup, it takes some time for the magnetic field in the induction coil to reach full strength. Until then, a largely unrestricted current flows to the emitter.
Or not. 
I've forgotten so much of this stuff. 
I have no idea what causes the TR-142b fasttech driver to kill leds when lowering the sense resistor. When it starts delivering around 6 amps to the led poof, that’s all she wrote. No matter if its on copper or aluminum. With the MT-G2 you can crank it up as much as you like.
Looks exactly like the one from fasttech that I’m holding in my hand. I cant tell if all the components are the same spec’s. But I have ever component that is on your pic and its labeled the exact same for each component on the board. The only difference I see is your TR-142b lettering is below the toroid, mine is at the top of the toroid. Oh ya, mine doesn’t have a spring either. :bigsmile:
.
There is more below the TR-142b lettering, probably a date.
Mine says 2011-03-30
Yours says 2011-01-10
What is the date on the vertical board. The one I have and another one I got today for someone else says 2011-03-30, but the label is at the top of the vertical board.
The Manafont 3T6 driver is labeled TR-0159D, 2012-01 and has a couple more components on the horizontal board under the toroidal inductor. Also the existing components have different values. The 2 additional components are SMD and are labled "A07T". They are rectangular and have 3 prongs total along the 2 long sides. Two on one side (far left and far right) and another tab on the other side in the center.
Also, there is another resister bay in the voltage sense resister band under the toroidal inductor.
As far as I can tell, my TR-0142b looks like the exact same board as dthrckt’s in this thread, which is a TR-0159B, 2011-06-18.
Well darn. Its to bad the Chinese keep changing something that works well or cant build a good clone.
Moderator007 wrote:
As far as I can tell, my TR-0142b looks like the exact same board as dthrckt’s in this thread, which is a TR-0159B, 2011-06-18.
https://budgetlightforum.com/t/-/17886
Looks the same as the TR-0142B's I have/had too. The version I have though (TR-0159D, 2012-01) is different. We talk about it a bit towards the beginning of that thread you linked, but I didn't notice the "missing" components in his pictures at the time. I can see it now. The two additional components would be on the horizontal board right under the sense resister back in the 3rd and 4th picture of the OP. His is just blank green there, but our TR-0142B's have a label imprinted there. Additionally, the vertical board is taller on the TR-0142B. I will do my best to take a couple pictures tonight. The version I have is the one I want to get more of.
EDIT: A couple pictures:
TR-0142B. Notice label on horizontal board under the toroidal inductor:
The labels on the bays are Q1 and Q3. Does anyone know what they are?
Added some driver pics to above post.
I decided to dial back to 6.5 amps. Also went with domed XM-L2 T6 3C. It a much tamer light now with only 134kcd, but it is much more useful to me in this configuration. Bigger hot spot and brighter spill. I kind of miss the burning hot hot spot though. Just seemed like. "I'm a pissed off hot spot and I have places to go!" Fun on occasion, but not what I need.
Does anybody have a definitive ID on the little 'A07T' things? I just fried a few, can't find any datasheets that match up.
edit: Well that didn't take long. They're AO7Ts, not A07Ts. http://www.aosmd.com/pdfs/datasheet/ao3400.pdf
Think I can substitute the IRLML2502 N-FETs from the BLF-15DD?
I’m sure it would work but don’t know what sort of gains just that would give ya.
is this the driver?
http://www.fasttech.com/products/0/10005304/1424817-tr-0124-14-66w-5-mode-led-flashlight-driver-for
No, it's missing these 2 components under the inductor. Q1 and Q3 in the pic below. That driver from FT will eat up xml's if you go past around 5 1/2 amps.
EDIT: I'm not sure the driver you linked to (One post up) fries xml's. I was thinking about this driver when I said the above. The driver you linked to above is a 32mm diameter driver. As opposed to the 26mm driver being talked about in this thread.