Electronics > Projects, Designs, and Technical Stuff
Deep Red vs High Efficiency Red in this datasheet?
(1/3) > >>
alank2:
http://www.avagotech.com/docs/AV02-0699EN

Deep Red says on page 6 that its Luminous intensity is typically 770 ucd, but High Efficiency Red says 270 ucd.

Does this make sense?  Shouldn't the "high efficiency" be brighter for a similar current?

In the peak pixel current on 5 it separates the deep red from the others, but it is 18.7mA vs 17.1mA, so current is very similar.

My question is, which one would use less current for a similar level of brightness?

Why two reds?  Is one a different LED type that is more efficient?
T3sl4co1l:
Haha, their "high efficiency red" isn't.

Yeah, I don't see any reason to doubt that their currents are similar, or that their emissions are different from what they say.  That means the "deep red" is REALLY bright, because it's on the far side of the luminosity spectrum.  Go figure!

Incidentally, that also means their green is shit, in radiative terms -- it's merely comparable by luminosity, but it's at (nearly) the peak.  Old fashioned GaP green I imagine -- not very good efficiency in general (and part of the reason why green laser pointers aren't just a laser diode and that's it, they're an IR-pumped doubler instead!).  There are phosphor+GaInN greens today (you can tell because they drop 3V just like blue LEDs, and have similar quantum efficiency but correspondingly worse power efficiency which is actually worse than white LED phosphors, which is why they aren't favored for illumination purposes), but they probably never updated the product line to add them (or can't because of the voltage drop).

Tim
alank2:
It is odd.  I do have all but the "deep red".

I set them to a low current setting, around 10% of maximum:

green "2963" is very nice
her "2962" is close
orange "2964" is decent
yellow "2961" is much dimmer than the rest

I wonder how the deep red would compare, but i don't have one.

The goal is bright display with least current use...
james_s:

--- Quote from: T3sl4co1l on December 19, 2018, 07:21:47 pm ---Haha, their "high efficiency red" isn't.

Yeah, I don't see any reason to doubt that their currents are similar, or that their emissions are different from what they say.  That means the "deep red" is REALLY bright, because it's on the far side of the luminosity spectrum.  Go figure!

Incidentally, that also means their green is shit, in radiative terms -- it's merely comparable by luminosity, but it's at (nearly) the peak.  Old fashioned GaP green I imagine -- not very good efficiency in general (and part of the reason why green laser pointers aren't just a laser diode and that's it, they're an IR-pumped doubler instead!).  There are phosphor+GaInN greens today (you can tell because they drop 3V just like blue LEDs, and have similar quantum efficiency but correspondingly worse power efficiency which is actually worse than white LED phosphors, which is why they aren't favored for illumination purposes), but they probably never updated the product line to add them (or can't because of the voltage drop).

Tim

--- End quote ---

Off topic here but there are direct injection green laser diodes now. They cost more than the DPSS green lasers (which became insanely cheap) but I'm sure somebody must sell pointers using them by now. I bought one of the diodes and built it into a module, the green color is a tad more saturated looking and slightly more pleasing to me than the yellowy DPSS green.
ajb:

--- Quote from: alank2 on December 19, 2018, 06:33:11 pm ---Does this make sense?  Shouldn't the "high efficiency" be brighter for a similar current?
--- End quote ---

Quantifying the subjective experience of "brightness" is complicated.  The Candela is a unit of luminous intensity and is equal to lumens per steradian--in other words, it's the amount of radiant energy (watts) corrected for human spectral sensitivity (watts->lumens) emitted in a given solid angle (cone).  This is typically measured over a fairly small angle near the viewing axis of the emitter.

This reveals two big confounding factors when comparing the brightness of LEDs via the the cd figure:

- The two LEDs may have very different distributions of their radiated power in space.  A wide angle LED will have a much lower cd rating than a narrow angle LED even if the two emit exactly the same number of photons.  The narrow angle LED will be much more intense when viewed straight-on, but the wide angle LED will be more intense at wider viewing angles.  It's up to you which is more important to you.
- The perceived brightness of the LED depends on how sensitive the human eye is to its spectral content.  This has a related confounding factor: Human vision is complicated.  More on that later.  The big thing here is to understand that the human eye has a spectral response that peaks somewhere around 500-550nm and tails off at wavelengths above and below that.  When you get towards the ends of the spectrum, the relative change in brightness over a small change in wavelength can be huge.  The spectral response is generally defined for two different vision conditions: the photopic regime, where the eye is adapted to bright conditions (better color perception and detailed vision in the center of the FOV, because the color-sensing rods, which are concentrated in the fovea at the center of the retina are dominant), and the scotopic regime, where the eye is adapted to dark conditions (much higher sensitivity at the expense of poor color perception and peak acuity, because the color-insensitive rods, which are more spread out over the retina and less concentrated in the fovea, are dominant).  Luminosity functions have been defined for both of these conditions (actually, several curves have been defined empirically), and by multiplying the power spectrum of a source by the correct luminosity function you get a luminosity spectrum.  If you integrate the power spectrum you get watts, and if you integrate the luminosity spectrum you get lumens.  Wikipedia has a nice graph showing a couple of the standard curves here: https://en.wikipedia.org/wiki/Candela#/media/File:Luminosity.png

That all said, assuming that the two displays have similar angular emission distribution, and assuming that they have been quantified under a suitable luminosity function, the one that is subjectively brighter SHOULD have a higher candela rating.  If the radiated power of the two devices is the same, then the source at 635nm should be about 1.57x times brighter than the source at 645nm (using the CIE 1978 photopic curve, they have weights of 0.21702 and 0.13812, respectively).

However, eye sensitivity at the very edges of the visual spectrum is poorly quantified and highly sensitive to viewing conditions--far more sensitive than can be understood via a fixed luminosity curve.  If you look at some other LED datasheets, like the Luxeon Rebel Color line, you'll find that they bin most of the colors by luminous flux (lumens), except for the deep red and blue wavelengths, which are binned by power.  So I suspect that in this case the candela rating of your displays is simply not a good reflection of the subjective brightness of the display due to the area of the spectrum involved.

As a general rule, I would never solely rely on datasheet figures for LED brightness.  Datasheets are fine for a basic idea, but the exact measurement methods and conditions are pretty much never specified, and the appearance of an indicator is highly subjective and context-dependent anyway.  It's always going to be best to get some samples in and look at them subjectively in different conditions and with different drive currents and see what works best.  This is especially important if you have lightpipes or lenses involved!
Navigation
Message Index
Next page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod