I’m getting my nerve up to do a mod of a cheapie Harbor Freight “25 Million Candlepower” spotlight for my father-in-law. It has a dual-filament halogen bulb that really isn’t very bright. I was thinking of using an MT-G2. I see that James3 has some MT-G2 on Noctigon which he has de-domed, and he just lowered the price on them. But, he’s in the UK and even the cheapest shipping method makes his product higher priced by a few dollars than a not de-domed MT-G2 from RMM. I’m prepared to pay the price difference if de-doming is going to help me get the most from the emitter in this spotlight. I’m certainly not gonna try to de-dome it myself. The light has a reflector that is about 8” wide but not very deep (I didn’t measure, but I think it is around three inches). I know the reason we de-dome emitters is to get more throw, and that is exactly what I’m going for. But, I have no experience with MT-G2’s at all and really I don’t understand how (or if) the de-doming changes the emission angle of the emitter. The spotlight is about the same size as the one in grantman321’s thread here, but of course Harbor Freight quality level. So help me understand. Will a de-domed MT-G2 give me more throw from this spotlight than an MT-G2 with the dome on?
i personally have never seen a successful dedome of an mt-g2.
Dedoming an MT-G2 works best when you slice the dome off with a scapel, don’t try to soak it in gas or anything.
Most peaple use a spacer ring (washer) to get a propper hight.
Dedoming makes a huge difference in throw, tint, and CRI. Check this out for some measurements by BLF members. So far I have ended up dedoming every MT-G2 I have put in a light. The tint and CRI just seem blah with that giant bubble of silicone there.
I just pick off the dome with a pick while it is very hot. I heat the dome with a hot air gun almost to the point the solder reflows. I try to keep the emitter itself cool on a heat sink because I think it helps the phosphor stay bonded to the die better. 100% success rate, but none of my dedomes look pretty as some silicon always remains over the crystals. That area is 3 dimensional and holds on to the silicone.
!
! 
wow impressive guys, these are the first dedomed mt-g2 i have seen.
I’ve done a few now successfully, problem is the price, if you mess one up its a lot! thats why my normal price is higher than XML XPG or XPL, as David says I’ve put the price lower (£12.00) but they won’t stay that low for long as theres too much risk in damaging one and being well out of pocket
This is about as close i can zoom in with my camera, it is quite difficult to take off all the silicone without all of the phosphor, but this is the best i can do.
!
!
^ That's a pretty good job. It may not look as pretty and clean as a xml2 gasoline dedome, but you forget all of that once you turn on your torch.
Has the remaining silicone ever had any bad effects on the beam?
I never use gas, even for xml2, there might be tiny bits of silicone left, but it’s only cosmetic!
None whatsoever, the beam is clean, it’s only big chunks of silicone that would effect it
Wow, great info there! Thanks!
Dome or not, that host reflector will be a disappointment for throw. Most of the light from the LED will come straight out the front without ever hitting the reflector; A big mule light with a small but dim spot in the middle.
These reflectors are made for filament bulbs that emit light in all directions equally. Test it with a mock-up first to see what it will do.
Agreed. I recently had this explained to me too. (Thanks Boaz)
Looks like we need to learn about HID for these lights.
This is what I was afraid of. Actually, this is what I need to know about de-domed versus dome-on. Which has a wider angle? The wider angle will work better in this spotlight, obviously. I do understand that the reflector was designed for incandescent bulbs and as such is not optimal for LED emitters. However, this being such a cheapie spotlight, nothing about it is optimal, and in changing to LED, I’m just hoping for a noticeable improvement from the original. Meanwhile, I may also check into Fresnel lenses. 
As for a mock-up, I already did something like that, with an old 100mA “5mm” LED I had soldered wires onto. I stuck a battery onto it, and held it up inside the reflector. At just about 3/4” into the reflector, it does converge upon a wall about 10 feet away with a hotspot about 18” wide. I was impressed that I could see the light from that LED that well at that distance with my overhead lights on. So, I don’t think the reflector situation is completely hopeless.
I'm guessing keeping the dome would give you a wider angle. I would also think the big apparent light source that the dome creates will help make up for the reflector shape better than dedome.
If you ever dedome, you may want to touch base with tjeret. He used heat and a blade to get a fantastic result on his second dedome attempt here.