Results: Testing XM-L, MC-E, SST-50, and XP-G emitters **Updated**

don't worry guys, it is in the works (well, at least the R2)

What about the newest Cree's, any plans, I realise they may not be power houses so maybe not much interested?

Yup. Thanks to dthrckt graciously sending me an xr-e r2, it will soon be tested along with a direct to copper mounted xm-l.

Gentlemen,

The initial star mounted test for the xr-e r2 is complete. The first post has been updated to include the graph. One of these days I'll go through and clean up that first post.

Thanks again to Dthrckt for sending me the emitter for testing....and yes, I'd like to someday get around to testing the new Cree flavors too... I don't need the money, but if y'all could somehow paypal me "15 minutes" that would really help :)

-NeedsavacationMatch

Thanks Match! So it's not worth going over 2.0A. Guess you didn't test it to the point of "pop, burn, fizz, fizz". How was the heat generated at 2A? Manageable for a decent heatsink setup?

-Garry

Thank you Match!

Overdriven XR-E lights seem to be quite popular right now.

Thank you (again ;-) ) Match. Those last results show why xre are still all the rage in flashlights. And a super duper reference as always.

Thanks guys, glad you can use the data.

In a typical star mounted application, regardless of whether or not the led star is mounted on 3lbs of copper or not, the answer to you question is no...not worth going over 2a. If it responds anything like the xm-l did, then the bottleneck is the star. As soon as I get some time, I'll remove the xr-e and directly mount it to copper and retest.

There was no reason to take the xr-e to failure, as the breakover point for output was 2.1a. More importantly, I want it alive for the afformentioned test. As for the heat, think of it in watts for easy comparison. 2a x @4v is about 8watts. Compare this to @ 9.5-10w for an average driven xm-l.

great work, Match! this has got to be THE most useful thread in the 'flashlight web'

As for the XB-D and XT-E, Cutter doesn't have any 4 xb-d PCBs. I suspect the one up versions of each will find some niche w/ beam profile or tint, and maybe eventually price, but I doubt we'll select them for throw or high output. Still, the data will be useful when you do get around to it.

I, for one, will be much more excited about the direct copper mount results :) Still, you shouldn't feel pressured, that would take the fun out of it, and if that happens, what's the point!

I hear that! I spend so much time fixing and maintaining my 'stuff' (particularly IC engines and whatever they're bolted to) that I don't feel like I own it - it owns me!

Thank you for all this information. This is my one of my favorite threads I have ever read and like the updates.

Thank you! :crown: H)

Thanks for all the effort, Match! Much appeciated. :beer:

Hi,

It's already been asked a couple fo times and I can't see a reply, but are you/would you consider testing a SST-90? The interest is huge because so many of the Chinese lights are under driven at 5-6As, where as the quailty units (Olight, VP2000, custom builds) hit out at 9-10 amps plus. It would be interesting though to see just how much difference it makes between the levels regarding lumens output. I shy away from buying the cheaper lights due to the amount of drive but if there is really little between, then maybe there is little reason not to try them.

I think he mentioned his power supply was maxed out w/ the sst-50 test

Thank you very much Match for the XR-E R2 results!

Was it an EZ900 or EZ1000 emitter?

I was inspired by Match's new guess-the-output contest...I plotted his XM-L test results along with Cree's data for T6 and U2 XM-Ls, accounting for thermal resistance. It's not 100% perfect [I made assumptions here and there for lack of data] but I was happy with how close the simulated data mimics the real deal (see chart 3; chart 2 refuses to cooperate). Data in bold font is overdriven territory-looks like Match got a high end T6!

https://docs.google.com/spreadsheet/pub?key=0AsTyaLKLOLAbdDlFUHktTGE1NE5qbkNEQVc5RnllMXc&output=html

Very Nice!

Not a diffrence the human eyethan pick up

agreed-I can *just* detect a difference in output comparing two of my Mags modded with T5 and T3 XM-Ls. Even then, I can't really account for how their tints (3C and 5A-ish) affect that perceived difference. Basically I had to find out if I could find out how well real life stacks up with Cree's data and vice-versa

Some of the images aren't visible, like XM-L and XP-G chart.

Can we have those back please?

Not sure why they disappeared, but I just fixed it.