Extrapolated XM-L Specifications

mvyrmnd

Flashlight Enthusiast
Joined
Sep 4, 2009
Messages
3,391
Location
Australia
I've just bought myself a light that pushes three XM-L's to 4.2A. I wanted to try and work out just how many lumens they might be generating at that sort of current, but the chart in the spec sheet only goes up to 3A.



I've done some black magic in Excel, and extrapolated this chart. The numbers would only be accurate to +/- 10%, as the chart doesn't give exact numbers and it's quite hard to read accurate numbers from the supplied graph.



So without further ado:



GRAPH REMOVED DUE TO THE FACT IT WAS COMPLETE AND UTTER RUBBISH.



Needless to say, my new Triple XM-L would be roughly 4000 emitter lumens on full blast! (Thanks to ma_sha1 for his awesome work!)



I hope this is useful to at least one person here... :ohgeez:
 
Last edited:

tfilandm

Newly Enlightened
Joined
Jul 13, 2009
Messages
15
Is it reasonable to assume that if their package didn't get thermally unviable past the published currents Cree would have given us the data?
 

flashflood

Enlightened
Joined
Mar 9, 2011
Messages
608
If you extrapolate linearly you'll get 30,000 lumens at 100 amps ;) -- but sadly it doesn't work like that. There have been several threads about peak useful drive current for the XM-L. Measurements indicate that somewhere between 3 and 4 amps (depending on the particular sample being tested), output starts to decline and you just get more heat. There does not appear to be much benefit in driving the XM-L above 3.5A, and nothing but downside in driving it above 4A.
 

mvyrmnd

Flashlight Enthusiast
Joined
Sep 4, 2009
Messages
3,391
Location
Australia
My initial graph was very wrong. The establishment of that was lost in The Great Crash. That other linked thread turned into a big argument, and it was determined that there was a clear benefit to overdriving the XM-L, if you had proper Heatsinking, and won the lottery on how well your emitter liked being driven.
 

flashflood

Enlightened
Joined
Mar 9, 2011
Messages
608

Right. However, Wayne Johnson (founder of Elektrolumens) has found benefit in going as high as 4A. In both cases, these results are based on testing a very small number of LEDs (possibly just one). Cree, the manufacturer, stops the chart at 3.5A, and you can see that the graph is concave leading up to that point. Hence my statement that peak output is somewhere between 3 and 4 amps, with diminishing returns beyond 3.5A.
 

tfilandm

Newly Enlightened
Joined
Jul 13, 2009
Messages
15
Well, Yeah. Flashflood's extrapolation provides as much useful information as the OP's. I'm sorry to be a **** about it, and maybe I'm missing something, but I fail to see the point of the spread sheet.
 

tfilandm

Newly Enlightened
Joined
Jul 13, 2009
Messages
15
Aahh..... Okay. Sorry to reflame you after you'd already been beaten up for posting imaginary data.:)
 

mvyrmnd

Flashlight Enthusiast
Joined
Sep 4, 2009
Messages
3,391
Location
Australia
The data wasn't imaginary, just a flawed extrapolation. After that I had worked a graph formula that matched the measured data to 5A and beyond.

Suffice it to say, there's a huge efficiency drop off after 3A, which is why Cree didn't include those values in their chart. For those chasing lumens at whatever the cost, there's a lot to be had at higher currents. Just how far you can push depends entirely on your equipment.
 

ma_sha1

Flashlight Enthusiast
Joined
Jan 10, 2009
Messages
3,042
Location
CT, USA
The Triple XML I sold to you was built on Cutter 20MM star, If I remember correctly.
http://www.candlepowerforums.com/vb...rbo-Triple-XM-L-With-Nuke-Mode-gt-4000-Lumens

for link in post #4, compare Testings are meaning less if they are not done on the same PCBs,
all the copper heatsinking means nothing is teh PCB is the bottle neck.

There is a huge light difference between the nuke mode from the Hi 2.8A mode in my build, that's a fact. also the Nuke is not 4.2A, it's 4.6Amp. :eek: in my testing with fresh batteries.

Would love to see beam shot against the Techjunkie 5xMCE light you just picked up:)

Yeah, a shoot out between the Triple XML Nuke vs. Penta MCE XX ray !!
 
Last edited:

mvyrmnd

Flashlight Enthusiast
Joined
Sep 4, 2009
Messages
3,391
Location
Australia
As soon as the XX-Ray arrives I'll be posting a shoot-out between all my big Mags. Battling it out will be a Elektrolumens MCE-III Mag (Triple MC-E), an Elektrolumens Wall-of-Fire (Quad P7), Your Triple XM-L, The XX-Ray and a Mag623.

It's going to be a bright night!
 
Top