LED forward voltage vs. color

paulr

Flashaholic
Joined
Mar 29, 2003
Messages
10,832
I know Vf gets higher as wavelength gets shorter. Blue led's have higher Vf than red led's, etc. White led's have high Vf because they're really blue surrounded by a phosphor.

I'm wondering whether there's a continuous Vf vs color curve which I'd expect there'd be. The manufacturer specs don't indicate it. They say 2.4 volts Vf for red, orange, yellow, then 3.6 Vf for green, blue, etc., all at 20 mA If. I'd expect something like 2.4, 2.6, 2.8, and so forth.

What's the real deal? In particular I'm wondering what happens if I try to direct drive a yellow or green 5mm led from a CR123 at 3 volts. Thanks.
 

Doug S

Flashlight Enthusiast
Joined
Jun 20, 2002
Messages
2,712
Location
Chickamauga Georgia
I've have noticed that at least at very low current levels, the greens and cyans will typically have a lower Vf than blues/whites.
 

paulr

Flashaholic
Joined
Mar 29, 2003
Messages
10,832
Doug, can you measure the Vf of a yellow when you get a chance? I'm wondering if 3V direct drive is likely to fry it. Some heat sinking would be provided the way I'd want to use it. But if 3V is too much, I can use a green instead. Application is a led mod for the #222 screwbase bulb in a Tekna Splash-lite. I keep evolving my ideas about this.
 

paulr

Flashaholic
Joined
Mar 29, 2003
Messages
10,832
Well, yes, but they also presumably use different chemistry from each other, or else they'd all be the same color. So do they have differing Vf's?
 

LEDependent

Enlightened
Joined
Jul 30, 2003
Messages
216
Location
Pittsburgh, Pennsylvania
I'm not quite sure what makes the color difference, but taken from the Luxeon datasheet:

10. All red, red-orange and amber
products built with Aluminum Indium
Gallium Phosphide (AlInGaP).
11. All white, green, cyan, blue and royal
blue products built with Indium
Gallium Nitride (InGaN).
 

paulr

Flashaholic
Joined
Mar 29, 2003
Messages
10,832
Interesting, thanks. Doug Owen kindly sent me some LED's in various colors and I'll do some experiments, but unfortunately I can't get to them right now.
 

Doug S

Flashlight Enthusiast
Joined
Jun 20, 2002
Messages
2,712
Location
Chickamauga Georgia
[ QUOTE ]
paulr said:
Doug, can you measure the Vf of a yellow when you get a chance? I'm wondering if 3V direct drive is likely to fry it. Some heat sinking would be provided the way I'd want to use it. But if 3V is too much, I can use a green instead. Application is a led mod for the #222 screwbase bulb in a Tekna Splash-lite. I keep evolving my ideas about this.

[/ QUOTE ]
Sorry, I don't have an amber on hand. In any event, one sample wouldn't mean much anyway since there is considerable variation in Vf bin among samples of a given color. If by 3V you truly mean a stiff 3V source, there might be a problem. If by 3V you mean a CR123 or 2XAA alkaline, I think you would be OK.
 

Doug Owen

Flashlight Enthusiast
Joined
Jan 30, 2003
Messages
1,992
A fairly large part of Vf is indeed from the energy of the photon emitted, as determined by Planck's equation, for practical purposes it's 1240 Ev divided by the wavelength in nanometers. Two to three volts for visible light. See the graph at the end:

The Photoelectric Effect

Doug Owen
 

Doug S

Flashlight Enthusiast
Joined
Jun 20, 2002
Messages
2,712
Location
Chickamauga Georgia
Trivial tidbit of info: Albert Einstein was awarded the Nobel prize for his work on the photoelectric effect, not his general or special theory of relativity.
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
The resistance caused by using semiconductor materials also adds to the Vf of an LED. We see a lot of variation in the Vf (where the Vf of a green may be higher than the Vf of a royal blue). There may be some base Vf that is accounted for by the wavelength of emitted light, but other variations are shadowing the relationship between Vf and wavelength.
 

Doug Owen

Flashlight Enthusiast
Joined
Jan 30, 2003
Messages
1,992
[ QUOTE ]
evan9162 said:
The resistance caused by using semiconductor materials also adds to the Vf of an LED. We see a lot of variation in the Vf (where the Vf of a green may be higher than the Vf of a royal blue). There may be some base Vf that is accounted for by the wavelength of emitted light, but other variations are shadowing the relationship between Vf and wavelength.


[/ QUOTE ]

While I agree that bulk resistance is *one* of the other factors, it and all others are minor against the energy needed by the photon. That is the well depth. This energy is the major part of the threshold energy needed for current at that operating point (whatever it happens to be under those conditions). White (blue) LEDs need a volt more than red because the blue photon is 'hotter' by a volt, not because the semiconductors differ.

Look at the slope of the V/I curve. A ten fold change in current (say from 2.5 to 25 mA for 5 mm parts or 35 to 350 for a LS) gives us only a 10% change in Vf. Ten times more light and a nearly trivial (one percent, relative) change in Vf.

Rules is rules. Can't cheat that quantum stuff.

Doug Owen
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
Right, I was talking more about LEDs of the same pedigree (base materials). Given a single sample of a luxeon (hell, even many samples), you would be able to tell with a pretty high degree of accuracy which materials the LED are constructed of (AlInGaP or InGaN) based on the Vf, but you probably would not be able to conclude which specific wavelength was emitted based on the Vf, due to other variations masking those differences.

In a perfect world, all green LEDs would have a certian Vf, all blues another, etc.
 

Ferrous

Newly Enlightened
Joined
Mar 17, 2003
Messages
62
Location
B\'ham WA
I jumped in late, but I follow and agree with most of what's been said. What I've seen supports the assumption that voltage follows wavelength at an arbitrary curent in a given package with the same recipe.

One exception I stumbled upon was that one manufacturer managed to make LEDs of three different colors (red, yellow, green) come out fairly close at 20mA.

Also, one thing I heard about was a GaN green LED with a Vf in the 2.8-3.2V range but the frequency was similar to the GaAlAs that have a Vf around 1.9V. It was supposedly more efficient, so I'm led to make some wild assumptions.
 

paulr

Flashaholic
Joined
Mar 29, 2003
Messages
10,832
Hmm, I just was at Radio Shack and saw they had some green LED's spec'd at 2.1 volts, maximum rating 2.8 volts. I wonder what's going on with that.
 

BLU3_SHOCK

Newly Enlightened
Joined
Dec 24, 2003
Messages
104
yellow led are almost alway 2.7 volt from 5 mm to ls and yellow has a lot of the same charateristics of green and cyan bout the same with amber but amber is like a cross between red and yellow.
 
Top