I think I screwed it up... Or did I?

So, I've been collecting the stuff I needed to make a 3D direct drive XPG R4 Mag for about 2 months now, and I'm on the verge on frying my LED.

I followed the steps from the P7 Mag mod tutorial (the one with all the pictures), just replacing the P7 with an XPG. To mount the LED to the heatink, I used a 70% Arctic Alumina 30% JB Weld mix, and it has held well (very well). I also used solid core wire to attach the LED to the switch (again, direct driving). The instant I power up the light, I am getting the angry blue tint shift (within about 2 seconds). The contact between the LED and the heatsink is perfect, and the batteries (alkaline) are only at about 75% charge each. What's going on here? I thought the XPG could handle DD from 3xalkaline cells no problem?

Oh well let's hear what the pros have to say...

Thanks in advance guys, any help would be much appreciated.

Re: I think I screwed it up...

What resistor are you using?

Re: I think I screwed it up...

On handling DD of 3 Alk. D, that depends, some XPG has super low Vf.

For direct drive, you always want measure the Amp to make sure,
don't rely on other people's findings.

Measure the tail cap Amp & see if it's over 2 Amp.

Also, don't mix with JB weld, that might be another problem area.
The Arctic Alumina or silver is plenty strong.

Re: I think I screwed it up...

ma_sha1 said:

On handling DD of 3 Alk. D, that depends, some XPG has super low Vf.

For direct drive, you always want measure the Amp to make sure,
don't rely on other people's findings.

Measure the tail cap Amp & see if it's over 2 Amp.

Also, don't mix with JB weld, that might be another problem area.
The Arctic Alumina or silver is plenty strong.

Click to expand...

+1 on that, IVe seen AA rip the soldermask off a circuit board. Soldermask is some of the toughest stuff IVe seen too, a pain to remove, heat resistant, solvent resistant etc...

Re: I think I screwed it up...

an xpg usually can't handle 3s AA's direct drive. So 3s D cells is probably way out.
an xre or an xpe could do 3s, but the new xpg's are not a direct drive option. (your emiter vf could be as low as 3volts, so a combination of almost dead batteries could be in spec (eg 1.0v closed circuit / with voltage sag))

Also that JBweld under your emitter, ugh, I'm not in favour but definately let us know in 6mo if the emitter's held up.
Usually the absolute thinnest amount of AA is best - any thermal paste is a worse conductor then actual metal (the mcpc, or heatsink), the thermal paste just makes up for deficiencies in surface smoothness of the two mating materials (heatsink, emitter). Adding a thermal insulator (the JB weld) to the thermal paste could be a real hinderance.

Re: I think I screwed it up...

Once again, thanks for the replys guys!

Long story short I felt like the light was a fail, so I put it on a shelf and forgot about it. Tonight I needed to get my mind off some stuff so I picked it up again and started messing with it. I checked all my solder joints, made sure I didn't overdo it on the heatsink paste (I don't think I did, but I very well may have) and cleaned the lens and reflector (alcohol).

Along the way I made myself a dummy D cell and popped it in in place of one of the cells. The light works perfectly, without the slightest hint of overheating or blue shift. I need to do a long term runtime/heat management test, which I'll probably do at work tomorrow night, but I'm very impressed with the light thus far. The throw is great, spill is bright and usable, and it should run a long time on 2xD cells.

Once again thanks for the posts and encouragement, I may have to try a P7 now.