Calling out to Weezard for LED advice

[mx4intx]

Looks like we lost some posts.

[Weezard]

Aand we're back!

Not the first time Cdot went tits-up ya know.
This one really had Sundance scramblin' though.

I figure he's been havin' a ball
Don't envy da owners here atall.

Looks like fun, but I'll tell ya what.
Re-building a server is a pain in da butt

It's time consuming and often frustrating.
I is happily retired from all dat kine sing.


At least they had a good backup from 2/23.
And I did save what few e-mails I could for interval if ya need 'em.

What's happened since the duck tape test?
Lit up any ladies yet?

Aloha,
Weeze

[mx4intx]

What's happened since the duck tape test?
Lit up any ladies yet?

Aloha,
Weeze

I dont remember where I was here...

I had all the blues mounted and wired, dimming is perfect.

And to make a long story short I totally forgot to wire up the current limiter for the red string. 2A power supply. Lost 2 reds for sure and now the 4 blues outshine the 6 reds left which should be impossible if they are ok.

I ordered this time eight 660's, four 625's, and four 725-745's. If the emerson effect is bunk then at least those last 4 will provide one well known catalyst to chem reactions, heat, but jus' a lil.

These are still the ledengin 5w's and with the four good blues adds up to 100w.

Heres the damage. I also notice that I have ea type of red on the same side. I wanted them alternating I think. My new layout puts all 4 colors onto a quarter of the heatsink with no color adjacent to the same color on a different quarter of the heatsink. Hope that makes sense.

Back to work on this Wed I hope.

[khyberkitsune]

Actually I wouldn't be surprised to see the blues outshine the reds. The red range is much harder to obtain good output efficiencies from, with some of the best Osram and Cree hitting only about 32-35% efficiency per input watt, whereas the best blues are easily hitting 40-45% efficiency per input watt with regards to conversion to light - assuming that the spec sheets per a particular bin are not lying to us.

But the typical pattern seen in spec sheets in almost every bin level is that reds are less efficient as far as radiometric emission is concerned. This is why many companies are using blue diodes as the base of their white LEDs.

[mx4intx]

Actually I wouldn't be surprised to see the blues outshine the reds. The red range is much harder to obtain good output efficiencies from, with some of the best Osram and Cree hitting only about 32-35% efficiency per input watt, whereas the best blues are easily hitting 40-45% efficiency per input watt with regards to conversion to light - assuming that the spec sheets per a particular bin are not lying to us.

But the typical pattern seen in spec sheets in almost every bin level is that reds are less efficient as far as radiometric emission is concerned. This is why many companies are using blue diodes as the base of their white LEDs.

According to the specs blue is 45lm to reds 125lm. At least for the 625nm red. The 660 is just given in mW. I dunno. I planned on saving them anyway. Two for sure are dead theyre both 625nm red. Thx Khyber

[khyberkitsune]

You cannot use luminous flux to measure monochromatic light in red or blue. The lumen is weighted at 550-555nm - green.

The only true measure of diode efficiency is radiometric output.

If red were more efficient (and higher electron-voltage) then it would be used as the base for white diodes. As it is, blue is used because of the ease of downconverting from blue light to red light using phosphors. Upconverting requires special tricks like frequency doubling (making a red diode output a 'green' light) and is horribly inefficient, moreso than a normal HID lamp.

[mx4intx]

You cannot use luminous flux to measure monochromatic light in red or blue. The lumen is weighted at 550-555nm - green.

The only true measure of diode efficiency is radiometric output.

If red were more efficient (and higher electron-voltage) then it would be used as the base for white diodes. As it is, blue is used because of the ease of downconverting from blue light to red light using phosphors. Upconverting requires special tricks like frequency doubling (making a red diode output a 'green' light) and is horribly inefficient, moreso than a normal HID lamp.

I wasnt using it to measure anything. I was wondering why my eye saw the opposite of what the specs stated, a lower brightness.

[khyberkitsune]

Blue light is a higher energy potential, thus a stronger source of light per photon emission. Blue light also has a pretty high optical damage, and the diodes you've picked are classed with laser diodes in Risk Group 2.

Spec sheets are misleading. Why are they trying to measure output in terms of lumens when the lumen is specifically set at 550-555nm? You're not going to measure 630nm or 660nm light as if it were 550-555nm.
Input power/output power. 1 w input, 450mw output. Blue diodes (sometimes) have the proper radiometric output listed, but those are few and far between.

[mx4intx]

Well the new LEDs arrived today. Building resumes tomorrow.

[khyberkitsune]

Can't wait to see how it goes! I just obtatined a rather difficult-to-find copper-cored computer heat sink, now I'm waiting to see if this one particular company can create a super-dense multi-band package in the 100w range to my specifications.