Free to a good home

J

Jaguarjoe

Senior Member
I have 20 new and unused Nichia high brightness LED's that were engineering samples for an LED street lamp project.

15 are #NS6W183JE
5 are #NS9L153MJH3E
I have no data for Vf or If or anything else other than each unit contains 8 LED's. All are mounted on aluminum substrates.

Free to a good home, just pay shipping anywhere in the world.
 
J

JoeFromOzarks

Senior Member
I could play with such a toy! :) PM sent.


Jaguarjoe said:
I have 20 new and unused Nichia high brightness LED's that were engineering samples for an LED street lamp project.

15 are #NS6W183JE
5 are #NS9L153MJH3E
I have no data for Vf or If or anything else other than each unit contains 8 LED's. All are mounted on aluminum substrates.

Free to a good home, just pay shipping anywhere in the world.
Click to expand...
 
J

John West

Senior Member
I went to the website of the company that produced the LED's. While they listed no data on the specific part numbers provided, I'm inferring that there's a fair chance that each of the 8 LED's is a 1W "warm white" light with a 120 degree dispersion angle.

For the benefit of those who are getting them, I'll post the results here of my testing once mine arrive and I can get a module on the bench.

Thank you, Joe.
 
J

Jaguarjoe

Senior Member
They gotta be bright 'cause 10 of them along with a huge heat sink made one street light.

This guy probably had 200 or 300 free samples from Nichia. I couldn't believe they were still sending him more.
 
J

John West

Senior Member
I'm hoping they work out as lighting in my motorhome. I live in it full time, and durable, efficient LED lighting that I can set up to run on 12V would be ideal. :)

I've found this info so far. It should be a good starting point:
http://www.national.com/analog/partner/nichia

It looks like the NS6W183JE units are pure white * 3.5V drop * 700mA * 3.2W per LED. I note that the listed I*E doesn't add up to 3.2W though. Purdy darn nice in any event.

I'm still digging around for more info on the other number, but it looks like those modules will require experimentation to figure out. I've found nothing on them on the web.
 
Last edited:
A

Andrew Cowan

Senior Member
I received my LEDs today - many thanks Joe!

I'm surprised by the forward voltage - with 250mA going through the LED, there is a 9.4v drop across the LED.

10v drop at 400mA.

I'd like to know how high I can go with the current (obviously with a good heatsink) - any one else done any testing yet?

Andrew
 
J

John West

Senior Member
I wouldn't advise going over the levels you are already up to, Andrew. That's already 4W, and if the LED's are the ones I dug up the specs on (link up above,) then you're already pushing them too hard. Unfortunately, there's only one way to know for sure how hard is "too hard."

I don't know which part number my LED's are, but I'll run tests on them as soon as I can dig out my thermal compound, and post the results.

I'll run one up to the 3.2W power spec listed in the doc linked above, (post #11,) and if it's nice and bright I'll call it "good enough." I'm thermal gooing my test LED to a large fan-cooled processor heat-sink, just in case.

I don't have enough of these LED's lying about to want to learn the hard way at just what point driving them "too hard" is located. I reserve destructive testing for when I have a bin overflowing with the parts.
 
Last edited:
A

Andrew Cowan

Senior Member
Although in post 11, you mention a 3.5v drop (which does nothing for these LEDs).

I think I'll stick with the 10v @ 400mA figure - that stops them getting hot when coupled to a CPU heatsink.

Andrew
 
D

Dippy

Moderator
I'm glad people have mentioned healthy heatsinking.

So many people think that these new high-power LEDs are so efficient that they don't get hot.
They do, very, and even a partial overheat will reduce light levels and lifetime.
They definitely need good heatsinking.

I'd be interested to see your opinions on LED streetlighting.
A local petrol filling station use 12 overhead panels of LEDs for lighting, each panel maybe more than 200 (two hundred) or LEDs. I couldn't tell you what type, but they are the cool-white colour probably from Ebay.

My local council are replacing all minor road SON lighting with Philips CFLs.
They are far dimmer. In fact, according to my £150 calibrated Lux meter, just under half the brightness of the SONs (those peachy coloured lights).
Apparently it is a energy-saving project.
In my little cul-de-sac they removed 6 x 250W SONs with 12 x 150W CFLs and we have less illumination. And the streetlamps are priced at £1200 each plus installation. Energy saving ? My Arm!
 
J

John West

Senior Member
Andrew Cowan said:
Although in post 11, you mention a 3.5v drop (which does nothing for these LEDs).

I think I'll stick with the 10v @ 400mA figure - that stops them getting hot when coupled to a CPU heatsink.

Andrew
Click to expand...
There were two different part numbered device types made available. Apparently you got the type not listed. Multiple Watt LED's typically consist of multiple devices on the die. At a guess I'd say the dies in your units are wired in series, while the ones in the spec sheet are in parallel.

After a close look at the die on the ones I received it appears the die consists of six LED's, possibly wired as two parallel strings of three diodes that are series wired. They would then require about 10V to properly drive them, (with the understanding that LED's should be controlled by a current source, not a voltage source.)

I expect you and I have the same devices and will get the same results. But I wouldn't advise whoever got the other part numbered devices to immediately hook them up to 10V. If they have only a single junction drop (3+Volt) then not even an infinite heatsink will save them.
 
Last edited:
D

Dippy

Moderator
For small items and to save worry, just include a Zero Invoice and say "Free Samples".
It works.
 
J

John West

Senior Member
I just tested my LED's.

3.3V gives 950mA = 3.13W. Very good stuff. Appears to be pure white light to my eye. No yellow, blue or green tint. 120 degree dispersion.

Thanks, Joe.

Apparently, based on the specs I found, I have the NS6W183 type LED's. At least, my measured spec is close to theirs.

I see the company also makes some 10V LED's as well. That's likely the type Andrew got.

I also learned that these "point-source" high-power LED's are absolutely brutal on a migraine headache. They're probably pretty rough on anyone, even without a headache. A good diffuser is a requirement for these little beasts.

Unless, of course, you want to do a bit of hand shadow pantomime. Without a diffuser, they will work splendidly for that. But mine will most definitely have diffusers.
 
Last edited:
J

John West

Senior Member
I'd call it cool. There is no hint of yellow in it at all. However, there is no hint of cool blue, either. A very white white.

But also very intense as a point source of light. Very intense. Wow!
 
Last edited:
D

Dippy

Moderator
Good find.
Check the spectrum graph. Pretty cool.

Did you see QI about slang?
"Cool" "Dude" etc. are ancient slang words.
Come on "cats" (same era) get some new ones :)
 
J

John West

Senior Member
I caught the slang reference, but didn't want to appear dated - even if I am. By the way, "cool" has had something of a resurgence in the USA. It's hard to kill a good colloquialism.

I've yet to find a data sheet on exactly the part number chips I seem to have. Not that it matters. They look very good when driven to 3.2W. I'm happy with that.
 
Last edited:
J

Jaguarjoe

Senior Member
If Nichia was willing to send hundreds of engineering samples to a couple of guys trying to make LED street lights, I'd bet they would send out samples to you guys.
Tell them you're making an LED Shake 'n Bake oven or something.
 
D

Dippy

Moderator
Na, just tell them you use a Forum where loads of people love scrounging freebies. You'll get extra ones for honesty ;)
 
You must log in or register to reply here.
Top