A Perfect Dedome?

A cooler MT-G2? Hmmmm……

I’ll check out the Rebel S1 thanks Dale.

As with the 2nd generation of Cree LEDs, not sure about the others, but XML2 and XPG2 seems to be really yellow after dedoming, even if it was CW 1A to begin with. Using the block to elevate the emitter was clever. Unfortunately there is no accurate way of determining candela by eye since your iris contracts and expands to let in less and more light respectively.

Yes, some (not so good) pic here: Accidental MTG2 dedoming..

But the aperture does not :wink:

Click on the pic and open it up in Flickr, then right click on it to see it original size. That barn is 610 yds away! The camera settings were chosen to represent what I was seeing, I was comparing 2 lights. The fence in the foreground is 100 yds from the light. The trees that appear in front of the barn are actually 320 yds from the camera while the barn is 610 yds.

Settings: 0.3 seconds ƒ/2.8 ISO 1600 15.1 mm (35mm equivalent to 28mm) G1X

What light is that? Looks super bright.

The HD2010, running on a Powerizer LiNiMnCo 26650 18A capable cell.

To show that barn?

Yeah cropped photos don’t hurt.
What mods have you done to your 2010? I know with my Shocker, I can barely see trees past 500 meters.

Well, there’s not much I haven’t done! lol Being new this has all been a learning curve. It now has an E1320 driver with 5 modes, memory, no blinkies and is running 3.82A at the tail. You’ve seen the star and emitter. I also bypassed the springs with 20ga wire. Is there something else I can do? Other than the Acrylic UCL AR coated lens that it already has? I’ve hounded MRsDNF to make a copper pill for me but no luck so far. Vesture of blood may do some for these lights, keepin my fingers crossed.

I might could add a couple more 7135s, just got my M3 head up to 4.22A, de-domed, with an E1320 driver that I added 2 chips to.

I have a 1/4” thick piece of copper that I’m thinking of cutting a round for a heat sink, since I can’t do threads I can do a different approach. Why couldn’t I use my dremel tool and cut the shelf of aluminum out of the pill, press fit this thick piece of copper in it’s place? I could reflow the star/emitter onto the copper first so it’d be like one piece, then press it into the sleeve of the pill and thread it back into the light.

Think that’d work? Then I could add like 3 more chips and get it up to 5A or so? This is an XM-L2 we’re talking about, so it should handle it.

Kinda like this? This one has a 1/4” ring of copper for the driver to set in, which I wouldn’t need here as I’ll have the aluminum pill with the threads left.

Replacing part of the aluminum pill with copper is definitely a good idea, the less aluminum it has to travel through to get to the body the better. OldLumens melts copper shot into gaps to make sure there’s maximum contact surface area, you could do that too.

And if you do that you could bring your XML2 to 6A comfortably. I recall someone here has a TN31 modded that gives 6.5A to a dedomed XML2.

Witht the LED inside of the refector now, do you notice any lumens loss compared to before de-doming?

I didn’t and still don’t have any way to measure the lumens, so I don’t know if there was that small loss or not.

I went to check it’s amperage again cause I wasn’t sure I was remembering correctly and it isn’t working properly. Seems there’s a ground short on the front side of the board. It’s happened once before but it’s not easily remedied this time, so yeah, for sure, it’s coming back apart. I’m getting to know this light REAL good! lol

So I will be doing the copper heat-sink, modifying the pill and revising how the board is mounted…which will also allow me to add chips. :slight_smile:

I’ll be back!

Night all! :weary:

Ooops, missed all this last night... Yes, a light meter helps big time.

Dale - that emitter stacked on copper looks sooo cool because that's always an issue of trying to get the emitter in the reflector and battling with the height of the wire connections interfering with the reflector seating. For the HD2010, you'll probably hit a point of diminishing return for the Vf demand of the XM-L2 with a linear driver and a good cell (KK, IMR, etc.). I don't think you could get 5 amps out of that setup, maybe 4.5 amps, maybe, but don't think you can get even that. Every 7135 you add steals more voltage too. The latest HD2010 build I did with a de-domed XM-L2 got 200 kcd with a Nanjg at 4.2A. The more copper you have in the pill the better, but you will probably not see any improvement in light output - the biggest, most important upgrade for light output increase is the copper star itself witha direct thermal transfer.

This is the light meter I have, and where I bought it: ebay 200-000-LUX-New-Digital-Illuminance-Light-Meter-LX1330B. This is a pretty common model that rdrfronty and manxbuggy1 have been using, though they more recently bought a better, more expensive meter (~ $130) but I believe the readings compare well with this one.

I’ll definately take those tips in mind, have to do some rebuilding on it as there is an issue somewhere so I’ll look into it and might add a chip to get up in the 4A range.

I’ll post when I get into it, and try to remember to take some pics! :slight_smile:

Ohh - ExTech is the better brand of light meters (that rdrfronty and manxbuggy1 bought one). Here's a nice model: http://www.coleparmer.com/Product/Extech_Precision_Light_Meter_LT300/EW-50510-93, and for only $160 more you can get it fully calibrated (NIST traceable). Funny because here at work, we make/sell nuclear medical equipment and we use NIST traceable radiation sources to calibrate and test our equipment...

He’d say “Well, Rosanne Rosana Dana, it’s alaways a sumtin!”

I got the copper in the pill, took some doing but it’s now got a half inch thick copper pill, cut out on the backside for the driver to fit into. Added 2 chips to the 11 that were already there :slight_smile: and got it all put together. My fancy star/emitter shorted out. So I took that apart, made very small pillar’s of copper solid wire to solder in to bridge the gap from emitter to copper star. That worked out beautimously! Tiny stuff that.

Got it all put back together and was only showing 3.42A at the tail, added a spring bypass at the driver and fried the tail switch. My very first toasted tailswitch. :slight_smile: So I’m modding the tailcap with a McClicky that I just happened to have on hand for just such an occasion. Hope to get this all wrapped up before dark-thirty and get some results back from all this work today! lol

Will keep ya posted, possibly have some pics.

Not sure if this has been said yet, but gun cleaner popped the dome off my XML in about 30 minutes. I’m going to give it another hour or so before I test. I’ll post my findings afterwards.

EDIT: Also, by test I meant see if it still works. I have no way to test lumen output. I can say though the tint didn’t seem to change much if at all for me.

I worked my but off on this light, again, and don’t know if I got anywhere or not. Well, I know I solved a glitch and have more reliable performance. But I’m not seeing the amperage the board should be putting out, albeit the light is very bright.

It’s sure bright at 97 yds!

It wasn’t exactly the switch that fried, it was the trace on the board that the switch is mounted on See the copper looking trace on the top hole off center? See the burned line where it fried? Doesn’t smell rosy :slight_smile:

A look into the reflector, showing my little soldered pillars connecting the emitter.

The driver, spring bypass installed, 2 extra chips up top to go with the 3 Erik put on the bottom. Those 3 on the bottom made me work hard, as you’ll see directly…

Sunken McClicky. More work. Although it’s inset considerably further, it works out because my driver is taller from the copper I added to the pill

The copper heat sink, a quarter inch slab of copper for the SinkPAD to sit on, reflowed to the quarter inch slab on the bottom for the driver. I fitted each piece from top/bottom like a wedge, then reflowed them together so they’re not coming out. But I used a copper bearing solder paste that flows at 430º so I got it pretty hot, discolored the copper and made it look ugly.

It was sitting on the burner bottom down, so this area got a lot of heat and looks horrible, I cleaned it up before putting it in the light. The more deeply recessed area is for the stacked chips on the inside of the driver. Also cut clearing for the positive area to ensure it touches nothing.

And I’m tired! This was all done with a dremel tool. Some filework. Cut the switch retaining pill with a cut-off wheel then filed it square, needed very little file work actually. But I had to open it up for the McClicky, work the plastic components for the new switch, one thing after another. LOL

I had the chance to take it out tonight and left it on high for about 5-7 minutes. I forgot it was on high. It was warming up, but not getting hot like it did before. So the copper is doing it’s job. Seems as bright, if not a bit brighter, but I have no way to check it. There’s 4.6A worth of chips on the board, and I’m showing 3.42A at the tail. Might pay to check the cell, I thought it was charged up but might be forgetting how much time I’ve been playing with it. :slight_smile: