testing a red XP-E, and comparing it to a red XR-C, 22/4/'14: XP-E2 results added in post #32

70 posts / 0 new
Last post

Pages

djozz
djozz's picture
Offline
Last seen: 1 hour 57 sec ago
Joined: 09/07/2012 - 17:04
Posts: 16001
Location: Amsterdam
testing a red XP-E, and comparing it to a red XR-C, 22/4/'14: XP-E2 results added in post #32

A bit hidden in this thread here I posted a quick output test of a red XR-E (it appeared to be called XR-C actually although the die size is more like XR-E Undecided). I was told there that the XP-E was much better, but I did not have a red XP-E to test that. In the meantime I got one (from Fasttech), I put it on a 16mm Noctigon and did the test yesterday, the graph below also has the data from the XR-C I tested for the other thread. I do not have a clue what bin was used for both the XR-C and XP-E. The XR-C was a bit more to the orange side than the XP-E, now I found in my emitter-box another red XR-C that did have the exact the same colour as the XP-E (probably 630nm instead of 620nm), in series on the led-tester at 50mA the brightness of both XR-C's looks the same.

As usual, the emitters were tested outside a flashlight, mounted on a big aluminium block, directly connected to a bench power supply, a reflector placed on top facing the ceiling, output was measured with a lux-meter, of light bounced from ceiling.

So this is a test on a few emitters that have been around for quite a while and while the test could be more thorough (like finding out the exact bins), the differences are clear enough, and I believe there's a few people interested anyway.

The die of the XP-E is clearly smaller than the XR-C, if you measure the surface area, it is a bit over 30% less, so at the same output, the surface intensity of the XP-E die should be 50% more. The XR-C was on its aluminium board (copper does not give much gain over aluminium with the XR-C), the XP-E was mounted on a 16mm Noctigon copper board. (the second picture is a stereo pair, for fun).

So indeed this XP-E has all the advantage of the world compared to this XR-C: lower Vf with higher output, higher max. current, smaller die for better throw, and there are dedicated copper boards available (although you can mount the XR-E on a MT-Noctigon).

I hope there's someone who finds this useful. (I have not seen any red XP-E2's in the wild yet, by the way. EDIT: I see now that cutter has the red XP-E2, too expensive for me to order there, but they are around, at least).

Edited by: djozz on 04/21/2014 - 19:14
CRX
CRX's picture
Offline
Last seen: 21 min 20 sec ago
Joined: 04/02/2013 - 15:27
Posts: 3827
Location: Scotland

Great coincidence. I just ordered a couple of red xp-e emitters to try out.

Interesting to see they top out at about 2A, thanks.

RMM
RMM's picture
Offline
Last seen: 1 week 3 days ago
Joined: 07/23/2013 - 13:47
Posts: 4006
Location: USA

Those results match up just about right with what my very primitive light meter and visual measurements told me: 1.9A is just about the sweet spot for the red XP-E mounted on a direct thermal path copper MCPCB.  At 2.2A it didn't get much brighter at all and I didn't try 2.5A.  I've been sticking with around 1.14A on the thin aluminum MCPCB.

Mountain Electronics : batteries, Noctigon, and much more! What's new? 

djozz
djozz's picture
Offline
Last seen: 1 hour 57 sec ago
Joined: 09/07/2012 - 17:04
Posts: 16001
Location: Amsterdam

This red XP-E led got 3.6A for a moment by accident (at the end of the test), and it seemed to have suffered a bit, after that, at 2A the output went down from 237 to 226 lumen, at 1 A it went from 153 to 144 lumen. That is nothing serious, but I have not seen any sensitivity at all to this amount of over-current with white leds.

B25364
B25364's picture
Offline
Last seen: 5 years 8 months ago
Joined: 01/16/2014 - 17:50
Posts: 97

good info
thanks i needed that info
very helpful

Rufusbduck
Rufusbduck's picture
Offline
Last seen: 8 months 1 week ago
Joined: 04/04/2012 - 15:34
Posts: 10389
Location: Golden state

It would be interesting to see how much better the xrc on a noctogon would do given it’s much larger solder point.

Three Tanna leaves to give him life, nine to give him movement. But what if he eats the whole bag?

Scott

UPz
UPz's picture
Offline
Last seen: 2 weeks 5 days ago
Joined: 01/09/2011 - 08:11
Posts: 530
Location: Barcelona

Very interesting, thanks!

djozz
djozz's picture
Offline
Last seen: 1 hour 57 sec ago
Joined: 09/07/2012 - 17:04
Posts: 16001
Location: Amsterdam

Rufusbduck wrote:
It would be interesting to see how much better the xrc on a noctogon would do given it's much larger solder point.

I did that test some time ago Smile

Here I compared two 'identical' white XR-E's, one on an aluminium board with dielectric layer, one on a MT-Noctigon. On copper, at 2A, the output is 10% more than on aluminium, the current at maximum output shifts from 2A to 2.4A, on copper at 2.4A you get 20% more output than on aluminium at 2A.

So, even if the red XR-E is mounted on copper, it will be still way outperformed by the red XP-E

NightCrawl
Offline
Last seen: 3 years 7 months ago
Joined: 01/22/2012 - 08:20
Posts: 3071
Location: Karlsruhe, Germany

Rufusbduck wrote:
It would be interesting to see how much better the xrc on a noctogon would do given it's much larger solder point.

I thought I saw something like that done with a white XRE somewhere and there wasnt much to be gained, but I cant find it..

Anyways, I dont think the XRE will do much better on copper. Certainly not better than the XPE.

Btw., digikey, mouser, cutter etc. stock the XPE2 in red but shipping is too expensive. Maybe we can talk RMM into buying a few and then distributing them? Silly

 

/edit: I thought it was djozz who did the XRE on copper test.. and it was him, he just posted the link. Big Smile

Lithium
Lithium's picture
Offline
Last seen: 1 month 3 weeks ago
Joined: 08/18/2012 - 06:01
Posts: 256
Location: Brittany

Are you sure the die of the “XR-C” is bigger than XP-E ? Have you tried dedoming the XR-C to see if it can give anything interesting?

Want DC-fix? PM me!

djozz
djozz's picture
Offline
Last seen: 1 hour 57 sec ago
Joined: 09/07/2012 - 17:04
Posts: 16001
Location: Amsterdam

Lithium wrote:
Are you sure the die of the "XR-C" is bigger than XP-E ? Have you tried dedoming the XR-C to see if it can give anything interesting?

I was not precise enough, the die of the XR-C is bigger, as seen through the dome, which is the most relevant to its use, it is what a reflector or lens will 'see'. Upon dedoming the dies may not differ that much ( I think they will still be different though)

Lithium
Lithium's picture
Offline
Last seen: 1 month 3 weeks ago
Joined: 08/18/2012 - 06:01
Posts: 256
Location: Brittany

Ok Wink

Want DC-fix? PM me!

JaredM
JaredM's picture
Offline
Last seen: 9 hours 36 min ago
Joined: 10/31/2011 - 13:33
Posts: 602
Location: Pittsburgh, Pennsylvania

I second the XP-E2 request. I’d be interested in them if anybody would sell them from say.. Washington. :bigsmile: Especially now that the Gen 1’s are out of stock and I’m looking to make another order..

RMM
RMM's picture
Offline
Last seen: 1 week 3 days ago
Joined: 07/23/2013 - 13:47
Posts: 4006
Location: USA

I have been taking notes. Wink

Mountain Electronics : batteries, Noctigon, and much more! What's new? 

wight
Offline
Last seen: 1 year 3 weeks ago
Joined: 11/27/2013 - 16:40
Posts: 4968
Location: Virginia, USA

Thanks for posting your results djozz. Good work.

Break’s over for now. That was a long one! wight catchup Wink Wink Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

texaspyro
Offline
Last seen: 4 months 2 weeks ago
Joined: 04/29/2011 - 12:43
Posts: 4593

BTW, your lumen numbers are probably bogus. Lumen numbers are done by weighting light output by the human eye response curve. Your lumen conversion constant was probably derived from measurements of white LEDs. Lux meter readings are also weighted by eye response, but the conversion constant will probably be different for colored and white light.

Comparing your red “lumen” numbers between the LEDs should give a reasonable idea of their relative outputs, though.

wight
Offline
Last seen: 1 year 3 weeks ago
Joined: 11/27/2013 - 16:40
Posts: 4968
Location: Virginia, USA

texaspyro wrote:
BTW, your lumen numbers are probably bogus. Lumen numbers are done by weighting light output by the human eye response curve. Your lumen conversion constant was probably derived from measurements of white LEDs. Lux meter readings are also weighted by eye response, but the conversion constant will probably be different for colored and white light.

Comparing your red “lumen” numbers between the LEDs should give a reasonable idea of their relative outputs, though.

Which way would you expect the results to be skewed?

Break’s over for now. That was a long one! wight catchup Wink Wink Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

texaspyro
Offline
Last seen: 4 months 2 weeks ago
Joined: 04/29/2011 - 12:43
Posts: 4593

I suspect that the true lumen numbers will be lower. Or maybe higher.

wight
Offline
Last seen: 1 year 3 weeks ago
Joined: 11/27/2013 - 16:40
Posts: 4968
Location: Virginia, USA

Fair enough.

Break’s over for now. That was a long one! wight catchup Wink Wink Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

RMM
RMM's picture
Offline
Last seen: 1 week 3 days ago
Joined: 07/23/2013 - 13:47
Posts: 4006
Location: USA

Visually the red lights appear brighter to me than their lumen numbers suggest.

Mountain Electronics : batteries, Noctigon, and much more! What's new? 

hank
hank's picture
Offline
Last seen: 13 hours 20 min ago
Joined: 09/04/2011 - 21:52
Posts: 8171
Location: Berkeley, California

If anyone has amber to try, I’m always interested in those (for “low/no blue light” use in evening hours)

djozz
djozz's picture
Offline
Last seen: 1 hour 57 sec ago
Joined: 09/07/2012 - 17:04
Posts: 16001
Location: Amsterdam

I give the lux/lumen readings to compare with other leds that are similar in colour, in that relative aspect it is useful , and although it may not be the absolute correct number for all single colour emitters, it does say something of how bright it is perceived (not completely 'bogus' IMO, but a little: yes). I know that it is much to ask of a cheap lux-meter to use the right correction curve so that for all measured wavelengths it is correct for the human perceived brightness (although it would be a simple software correction, I wonder if it is done right, or at all).

wight
Offline
Last seen: 1 year 3 weeks ago
Joined: 11/27/2013 - 16:40
Posts: 4968
Location: Virginia, USA
hank wrote:
If anyone has amber to try, I’m always interested in those (for “low/no blue light” use in evening hours)

Is it not safe to assume that all colors of XP-E are improved over their XR-C counterparts?

Break’s over for now. That was a long one! wight catchup Wink Wink Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

djozz
djozz's picture
Offline
Last seen: 1 hour 57 sec ago
Joined: 09/07/2012 - 17:04
Posts: 16001
Location: Amsterdam

djozz wrote:

I know that it is much to ask of a cheap lux-meter to use the right correction curve so that for all measured wavelengths it is correct for the human perceived brightness (although it would be a simple software correction, I wonder if it is done right, or at all).

I must correct myself: I assume that light detection in a luxmeter is done with a single detection device. If it does not have the correct sensitivity for different wavelengths, according to the definition of the lux (not likely), any sensitivity correction must be done by optical filters before the light reaches the detector, after the detector the luxmeter does of course not know which part of the signal comes from which colour of light. For 24dollars shipped there's no way my luxmeter has such pre-detector correction (if it exists at all).

texaspyro
Offline
Last seen: 4 months 2 weeks ago
Joined: 04/29/2011 - 12:43
Posts: 4593

Pretty much all lux meters have a silicon photodiode sensor with a greenish filter in front of them. The filter is supposed to match the diode response to the human eye response.

The quality of that filter/match varies quite a bit. If you are looking at wide bandwidth white light, differences in wavelength/output tend to average out. Looking at a narrow band light (like a red LED) has greater potential for a mismatch at that wavelength being significant.

Fritz t. Cat
Fritz t. Cat's picture
Offline
Last seen: 11 months 3 weeks ago
Joined: 05/07/2013 - 00:33
Posts: 2535
Location: Si Valley

Nice to see that they survived that much current, as my red XP-E SK58 gets too hot to hold (on a 14500).

Flashlight designers should look at lighthouses and pottery.
这些谁设计的手电筒应该看灯塔,以及在陶器。

nofearek9
nofearek9's picture
Offline
Last seen: 9 hours 29 min ago
Joined: 07/08/2012 - 08:46
Posts: 1348
Location: Cyprus

any news about xp-e2 red?

djozz
djozz's picture
Offline
Last seen: 1 hour 57 sec ago
Joined: 09/07/2012 - 17:04
Posts: 16001
Location: Amsterdam

nofearek9 wrote:
any news about xp-e2 red?

Well, MRsDNF has bought some red xpe2's for me from Cutter Cool, and sent them to me (he even refused payment Surprised), I got them in last week and will do a test one of these days Smile

nofearek9
nofearek9's picture
Offline
Last seen: 9 hours 29 min ago
Joined: 07/08/2012 - 08:46
Posts: 1348
Location: Cyprus

that’s nice,thanks both of you.

RMM
RMM's picture
Offline
Last seen: 1 week 3 days ago
Joined: 07/23/2013 - 13:47
Posts: 4006
Location: USA

I have a few customers who have lights with the XP-E and now some with the highest binned XP-E2 I could find.  They both told me that they feel like the lights are visually more than 10% brighter with the XP-E2 vs. the XP-E.  This is just anecdotal but cool to see that they can tell a difference. 

Mountain Electronics : batteries, Noctigon, and much more! What's new? 

Fritz t. Cat
Fritz t. Cat's picture
Offline
Last seen: 11 months 3 weeks ago
Joined: 05/07/2013 - 00:33
Posts: 2535
Location: Si Valley

I ordered a Philips Rebel ES 660nm Deep Red LED, as discussed in another thread. Probably not as bright but redder.

Flashlight designers should look at lighthouses and pottery.
这些谁设计的手电筒应该看灯塔,以及在陶器。

Pages