CRAP there goes $30.
Yeah, it’s a beaut!
I like how the reflector screws in the pill, making the whole thing a heatsink.
I have a 2 mode V3 XM-L U3 I'm extremely happy with. As much as I'm tempted to get a V5 XM-L2 U2, I think I'll hold off for a while until someone reviews it.
Yeah :)
Link wasn't good before. https://budgetlightforum.com/t/-/19982
I had to reflow 3 LEDs in my Shadow SL3. I ordered the warm emitter version, but Fasttech sent me the cool version. I had to remove the electronics and the rubber switch cover then heat the entire head using the pre-heater in this hot air station. Worked great.
At $170 this might be slightly out of budget, though. :bigsmile:
But it will certainly get the job done.
I’ll borrow some kitchen gas burner, read somewhere they get up to 800°C, but can be adjusted. XP-G2 flambé…
I used my toaster oven, set to 360 deg F to remove the old LEDs. Worked fine. I could have used it to reflow the new emitters, but I wanted to be able to see them as the solder melted.
I do a lot of electronic repair, so I decided to go ahead and upgrade my soldering station to the hot air rework unit I show above.
oh poo the xpg2 is sold out
i see it says 8x 380ma 7135 chips - does that mean they are pushing 3.04a to that xpg2 ? 
Fast!
Wrote some emails with Hank, he said that it’s driven at 3A.
I thought that was a little… weird, then I looked at Match’s test results Emitter Test Results pt4: XML2, XP-E2, XP-G2, and Nichia219! (along with sinkpad tests) and the Vf of an xpg2 at 3A is surprisingly low (at least to me). It’ll have better regulated runtime than an xml.
As far as I remember Match himself didn’t trust his Vf measurements (whereas his lumen recording seemed pretty accurate and I regularly rely to his charts).
When djozz did his “crash testing” he stated Vf as well and had 3.55V at 3A for an XP-G2. That sounded very reasonable - and is higher than XM-L2 with 3.4V at 3A and much higher than an XM-L, which seems to have a Vf of 3.15-3.20 at 3A. Here is the post of djozz with the graphs for XM-L2 and XP-G2.
Whoah! XP-G2 pumping 3A…that thing has GOT to be making some heat! Nice!
It looks like the best just got better.
My XinTD C8 V4 (XM-L2 T6 3C) gets about 33 kcd of lux (~360m); I wouldn’t be surprised if this XP-G2 version can get 60 kcd (~490m). It’s too bad they don’t sell a de-domed version; then we could probably have a stock C8 with 90+ kcd (~600m) of throw.
Does a XP-G (G2) really have that much more throw over a XM-L (L2)? Or since you are pushing the XP-G at 3A which is 2x times the 1500mA max current the reason? I know my XP-G at 1400mA gets CRAZY hot…at 3A…boy howdy!
The XP-G2 is roughly half the size of the XM-L2, or one quarter the surface area. However, it’s not one quarter as bright; at 1.5A it’s roughly half as bright as the XM-L2 would be at 3A. So, this means its surface brightness is twice as high as XM-L2. Run it at higher amps and it’s more like three quarters as bright as XM-L2, in one quarter the surface area. And if you de-dome it, the apparent size gets even smaller and more light goes sideways into the reflector instead of forward into the spill area.
It’s not because the XP-G2 is over-driven. It’s less efficient than XM-L2, putting out fewer lumens per watt. Instead, it’s because the XP-G2 has a higher surface brightness — more lumens per square millimeter.
So, long story short… XP-G2 will get you a smaller, more intense hot spot than XM-L2.
My numbers earlier were based on actual lights people have made with similar parts. My XinTD C8 V4 gets about 33 kcd lux with a XM-L2 at 3.04A. At the same power level, a cheaper C8 with de-domed XP-G2 gets 90 kcd lux. And some people have gotten 125 kcd.
This works for both reflector-based lights and aspherics. In my Ultrafire 838 zoomie, I just tried a XM-L at ~2A versus de-domed XP-G2 at ~2A, The XM-L got 14 kcd lux, and the de-domed XP-G2 got 60 kcd. The XP-G2 really does have that much more throw.
Whoah…nice!
Thanks!
I have just ordered one with an XM-L2 T6 4D. I will be adding another 4 x 380mA 7135s to it and will upgrade the the driver wires and add some braid to the springs.
I hope you have a powerful soldering iron. The heat should transfer really fast to the pill, since the mcpcb is soldered, not attatched.
+1. This one is definitely going to be harder to modify exactly because of the soldered MCPCB. You've got to be careful how much heat you put into the pill; if you get it hot enough to break the MCPCB/LED loose it may be difficult to try and recenter it. It's pretty hard to get the solder blobs off of these, I always add a lot of fresh solder and flux to them before trying to remove it. It may seem counterintuitive but it is really is the easiest way to get them off.