initial evaluation of Seoul Semiconductor Co. (SSC) LED ... with 24h burn in data

Candle Power Flashlight Forum

Help Support CPF:

wquiles

Flashaholic
CPF Supporter
Joined
Jan 10, 2005
Messages
8,516
Location
Texas, USA, Earth
PhotonFanatic (Fred) recently sent me one of the new LED's that he is selling here on the forums, and he asked me to do a formal test on my LED Lightbox

Here are some close-ups (SSC on the left) against a "normal" U-bin Luxeon 3W LED:
IMG_6093.JPG


IMG_6096.JPG



Note in this shot, how thin the star base is compared to the Luxeon LED:
IMG_6095.JPG



I compared the new LED to an S-bin 3W and to an U-bin LED, both white, star LED's from Luxeon. Here is the data that I collected so far, at 500mA, 750mA, and 1000mA, using constant current converters from the Shoppe (downboys):
SSC_vs_Luxeon_v1.JPG



Even though this is just a sample of one LED, some things to point out:
- this SSC LED is very bright, and specially at 500mA and 700mA, this particular sample was brigther than my U-bin 3W.
- this SSC LED was particularly very efficient at low currents, specially 500mA. Who knows how much better it might be at 250-350mA :naughty:
- as expected the vf was higher than even the "L" bin I tested against, specially at the higher drive levels, but at 500mA was not bad at all, being the most efficient LED I tested.
- at 1000mA the intensity peaks quicky and drops 3-5%. I am doing some burn-in tests to get additional information, but it might be that these 2.5W LED's do perform better at the lower current settings 😉

My next test (now that I have a fan-cooled copper HS) was going to do a 12 hour (and then continue for a full 24-hour) burn-in at 1000mA, to see if the vf changes, but unfortunately my downboy 1000mA almost thermally shuts down (LED dims a LOT!) after running for 5 minutes or so. I am now as we speak trying it again with a dedicated fan running on the downboy to see if things work better this way. I will report later what happens 😉

EDIT: The initial drop in output at 1000mA was due to the downboy heating up. Now that I have had the DB with a fan right on top of it, the output remains fairly constant, as I would have expected given the cooled HS I am using.

EDIT #2: At 12 hours, the intensity has gone from 450 to 460, and the vf has gone down from 4.62V to 4.39 volts. I will report tomorrow again at the 24 hour mark 😉

EDIT #3: Now at 22 hours, the vf is down to 4.38Volts, but what is most interesting is that the intensity is up a little to 465 (it started at about 450 or so) !!!

EDIT #4: After 24 hours at 1000mA, I let everything cool and repeated the tests on the SSC LED. The vf at 1000, 750, and 500mA went down, but the intensity went up as well !!!

vf before (1000mA) = 4.65v
vf after (1000mA) = 4.38v
intensity before (1000mA) = 456
intensity after (1000mA) = 473

vf before (750mA) = 4.20v
vf after (750mA) = 3.99v
intensity before (750mA) = 409
intensity after (750mA) = 430

vf before (500mA) = 3.75v
vf after (500mA) = 3.64v
intensity before (500mA) = 305
intensity after (500mA) = 318

So basically, this LED did benefit from the 24-hour burn in period - it got brighter "and" more efficient 😉 . Although things got better at all 3 current drive levels tested, the best efficiency gain came at the 750mA drive leve - 10.7% improvement :naughty:

Thanks Fred for providing the star for testing :rock:

Will
 
Last edited:
Re: initial evaluation of Seoul Semiconductor Co. (SSC) White LED ...

How do optics, reflectors and beam patterns compare? Thanks!
 
Re: initial evaluation of Seoul Semiconductor Co. (SSC) White LED ...

goldserve said:
How do optics, reflectors and beam patterns compare? Thanks!
I honestly have no idea - so far I am only testing the individual LED by itself, on a very large fan-cooled HS - far from what the LED will see in an actual light, but a fair platform to compare one LED against another.

To compare and talk about optics, reflectors, and beam patterns you would have to actually build a light with this LED to be able to answer those questions. Plus to really compare them in a fair maner, you would need to use a Luxeon of the same intensity and build a second, indentical light, so that you can then compare how well optics and reflectors do with the LED's output. Not "difficult", but not easy either as there are just too many variables :candle:

Will
 
Re: initial evaluation of Seoul Semiconductor Co. (SSC) White LED ...

Edited initial post. Seems that the initial intensity drop at 1000mA was due to the DB heating up 😉

The LED has been running for 9 hours at 1000mA just fine :naughty:

Will
 
Re: initial evaluation of Seoul Semiconductor Co. (SSC) White LED ...

that's a heckuva Vf drop for 12 hours. However, given the high Vf, it is quite welcome.

The die appears larger than a luxeon - is that just because of extra phosphor?
 
Re: initial evaluation of Seoul Semiconductor Co. (SSC) White LED ...

First post updated with results from 24-hour burn in at 1000mA. Performance is up, most specially at the 750mA drive level 😉

Will
 
Re: initial evaluation of Seoul Semiconductor Co. (SSC) White LED ...

I'm a bit confused...
What means LBU ?
How could we see if the SSC is efficient against Luxeon? Lumen/W ?
Thanks.
 
Re: initial evaluation of Seoul Semiconductor Co. (SSC) White LED ...

LBU = Light Box Units
There's a link in the first post to his light box thread.
 
Will, what's the tint like with the SSC LED? Also, in the picture the die on the SSC looks a little bigger than the Lux die. Is that the case?
 
MillerMods said:
Will, what's the tint like with the SSC LED? Also, in the picture the die on the SSC looks a little bigger than the Lux die. Is that the case?
Sorry about the very late reply. Tint, I don't have one built into a light to do a side-by-side comparison, so I don't know. From looking at it, it looks fairly nice and white.

About the die size, yes, it is phisically larger than the die in a 3W Luxeon LED, but noticeably smaller than the die on the 5W Luxeon LED.

Will
 
Back
Top