EOS modded with XR-E

I had a few XR-E LEDs that I had purchased and thought I'd try my hand at throwing it in an EOS. The result turned out very nicely.


Here are some beamshots. Nevermind the dates. I don't have a fancy camera, so all I can do is turn off the flash and press the button.

EOS Luxeon (modded with reflector & diffuser, my preference)



XRE (reflector & diffuser, the beam has several artifacts without the diffuser)



Side by side

How did you mod your EOS? Do you have any pics of it? You took out the optic and replaced it with a reflector and diffuser? Does that work better than the optic or is it just to get more of a flood light than a spot?

It would be cool to put a Cree in my Petzl Tikka XP and see how bright boost mode is. I don't know how the cree will work with the luxeon optic though.

First I pulled it apart. The IMS-17 is a drop in replacement for the default optic. The Optic would never work with the XR-E. The reflector can be made to work after a bit of modification. And I use the HDS diffuser from lighthound. The reflector is really bad without the diffuser, even with the diffuser you can see artifacts if you look carefully, they're smoothed out though so you'd never notice if you weren't looking at a white wall.

There are some gotchas in the process. I had issues with the XR-E stars and the reflector is touchy.

My overall impression is the throw is largely the same to perhaps slightly better, but there is a lot more total light. So basically this mod provides a lot more flood. I like the spot+flood of the reflector as opposed to the mostly spot optic. I really like the end result, so I modded my wife's EOS. Apparently mine has better tint though so it looks like she's going to be reaching for mine first.

I'm going to have to get me some more stars...

Sorry, no pictures of the insides.

Hope nobody minds cyclic links, it is quite related though, heh. Here is how I modded the reflector. https://www.candlepowerforums.com/threads/140186

Nice job. If it has a converter in it, it likely pulls less power also, so expect a possible increase in runtime.

Nice job! I am amazed at how much better the XR-E's seem to be compared to the Luxeons.

Could you please post pics of the Cree beam without the diffuser? I'm curious to see the artifacts you are talking about. Any ideas if they're caused by the reflector resizing, poor focus... ?

very nice! I have been thinking about this since the XR-E started its cheer fest here on CPF ...

However, I was waiting for Don's 17mm reflector to become availble... but your method makes sense...

What exactly did you have to do to your reflector? Thanks for sharing!

EDIT:

NM, just found your other thread

https://www.candlepowerforums.com/threads/140186

greenLED said:

Nice job! I am amazed at how much better the XR-E's seem to be compared to the Luxeons.

Could you please post pics of the Cree beam without the diffuser? I'm curious to see the artifacts you are talking about. Any ideas if they're caused by the reflector resizing, poor focus... ?

Click to expand...

Not sure if my camera would pick them up very well, especially on my textured walls. There's another reflector thread floating around, the main problem without the diffuser is the same "dark ring" problem around the spot. Other than that, a few artifacts in the spill due to the dome and a tighter, slighly textured, spot is the result. Fairly typical of non-sputtered reflectors other than the dark ring, still perfectly useable but I'm getting pickier about my beams :naughty:. If you want maximal throw you definately don't want the diffuser. I haven't played with placement extensively but good placement should be able to get you pretty close to stock IMS performance. The dark ring is a mystery.

The main reason I went with the diffuser initially with the Luxeon is the ability to use it around people while shining it at them. A diffuser keeps them from looking straight at the die when they are in the spill, it's much more tollerable. A nice beam is a generous side effect of that.

Some things I've noticed about the LEDs. The domes themselves have imperfections so you get artifacts even with an exposed LED (bumps that cause small holes, etc). The die location varries LED by LED, some aren't exactly centered. They aren't much different but I can see it and it probably has some yet to be explained effects.

The stars I received were flow soldered but the negative contacts were not lined up properly and didn't connect, I had to fix that by soldering a glob on the negative contact to connect it to the pad next to the package. An experience that might be frustrating for someone who is not handy with an iron.

Another note, you might be able to use a slightly filed down IMS-20 in the EOS for more throw. There's a bit more room in there when you plop a reflector right down on the package like that. There's basically no chance of it rattling either which happened occasionally with the luxeons and an IMS-17.

Another side note, I did put some thermal paste on the temp sensor in there so it hopefully still makes contact with the star. I'm really not sure though cause of the extra head space.

Thanks for the additonal info, BackBlast.

Any veredict yet as to what causes the darker area around the hotspot? Is it reflector curvature, improper focus (because they're designed for Lux), user limitations, die properties (a la LuxV)?

greenLED said:

Thanks for the additonal info, BackBlast.

Any veredict yet as to what causes the darker area around the hotspot? Is it reflector curvature, improper focus (because they're designed for Lux), user limitations, die properties (a la LuxV)?

Click to expand...

Greenie, this is hardly the last word! But my guess is the dark area has something to do with the way light gets deflected by the edge of the metal ring around the emitter dome. Similar things happen with deep bezels on lights.

Then again, just a guess! :thinking:

greenLED said:

Thanks for the additonal info, BackBlast.

Any veredict yet as to what causes the darker area around the hotspot? Is it reflector curvature, improper focus (because they're designed for Lux), user limitations, die properties (a la LuxV)?

Click to expand...

If I were to make a wild guess, I'd say that there is actually multiple "die locations" due to the optical oddities with the package. So, it's a side effect of their integrated pseudo optic so to speak. Don mentioned "ghosting" with some of his experiments. So what we might be seeing is a "dark ring" between one image and the next, they are first of all distinct locations. They're far enough yet close enough together that the parabola focuses the other (much weaker) die in a distinct ring (donut hole!) outside of the spot which makes the space between look like a dark ring.

If this is the case, the only way I think you can be rid of it is to tune the parabola so it's not focused on just the main die location, shift it out of focus a bit to blur the image locations so the 2nd die location isn't so distinct. Basically you're spreading the spot out to cover the donut hole, but as you do so the ring is also shrinking so you don't have to go as far as it might initially appear. Your actual focus point is going to be somewhere between the two image locations, closer to the main one. This could probably be accomplished in an existing reflector by sanding off more of the rear of a reflector that's exibiting "dark ring" symptoms. It looks to me that Don has already done this with his reflectors in the A19, at least what I see from the beams shots. :goodjob: I'll try and get around to messing with one of my reflectors to verify my theory or through some other testing of some kind.

So, if I'm correct. The best throw is going to get the "dark ring", the best beam isn't going to have the best throw. At least, with a reflector. Pick your poison.

Anyway, I hope that was at least entertaining to read :candle:

one big negitive with this mod... as it is brighter... it'll also make the low setting EVEN brightER!

hope PT is watching this LED

BackBlast said:

So, if I'm correct. The best throw is going to get the "dark ring", the best beam isn't going to have the best throw. At least, with a reflector. Pick your poison.

Click to expand...

Deja vú... SF U2 donut hole? :green:

greenLED said:

Deja vú... SF U2 donut hole? :green:

Click to expand...

Actually, I think this is a slightly different beast. With the Lux5 there are 4 dies all on the same Z coordinate, only moving around the X, Y realm. If you're out of focus you're out of focus, and you're in donut hole land. In this case we have the real die, and (at least) one "virtual die" that is the same X, Y but different Z. Because the virtual die is out of focus you get a "donut hole" and a real spot, combine the two and we get the "dark ring".

jar3ds said:

one big negitive with this mod... as it is brighter... it'll also make the low setting EVEN brightER!

hope PT is watching this LED

Click to expand...

Actually, it's not as bad as you think. The XR-E gets ~80% more brightness of a Luxeon at high drive levels. At low levels it's more like a T bin Lux I (or V Lux III), just slightly brighter. If you look at the chart in this thread https://www.candlepowerforums.com/threads/138503 provided by user hakstooy you'll see that low level should be pretty close to a luxeon low.

In reality, the major improvement to the Crees is thermal regulation more so than die efficiency. So... It's efficiency drop off happens at higher current than a Luxeon.

If I look at his numbers, this is what it looks like compared to the U bin.

So, the Cree scales better. On the lower end there's not much difference. My low is brighter, but I can barely tell the difference.

BackBlast, what are the units under the Ubin and P3 columns? lumens? What's the current drive for those?

They come from an uncalibrated sphere, they're meaningless other than for comparison purposes. I suggest you goto the thread and look at the actual chart. I just pulled them from it.

thanks bb, for that post.... those percentages are linear though not logrithmic.... and so the percentages are going to have a bigger differnet to ours eyes