Howto date LEDS

[RAPo]

I have a set of (red-) LEDs, no ID numbers.
Is there any way to date these leds?

[Gyro]

That's odd (assuming that they're both Red). Early dim Red GaAsP LEDs had a much lower dynamic resistance than modern high brightness types, those started in the '70s though. If comparing the two, I might have put the plots the other way round. The higher voltage one looks very noisy, likely to be defective / reject.

It's hard to tell anything from the plots, you might learn more from looking at the dies.

[SiliconWizard]

Dating LEDs from their I-V characteristics is an ... interesting approach.

[schmitt trigger]

Perhaps the OP would care to elaborate his theory.

[RAPo]

yes indeed, but I double checked the connections/screenshots.

The 19YY LED looks like a desoldered one, maybe that's what we see here.

That's odd (assuming that they're both Red). Early dim Red GaAsP LEDs had a much lower dynamic resistance than modern high brightness types, those started in the '70s though. If comparing the two, I might have put the plots the other way round. The higher voltage one looks very noisy, likely to be defective / reject.

It's hard to tell anything from the plots, you might learn more from looking at the dies.

[RAPo]

It was one of those brainwaves I sometimes have ;-)

Dating LEDs from their I-V characteristics is an ... interesting approach.

[RAPo]

Well, the theory is simple (but maybe practice is not).

An I-V graph of an LED shows where the voltage drop remains constant independent of the current.
For example, a red LED has a different voltage drop than a blue one because they are made of different materials.

In the run of the years, different materials were also used for a red LED.
The earliest LEDS had a semiconducting alloy gallium phosphide arsenide (GaAsP); nowadays, the red LEDs are made of Aluminum indium gallium phosphide (AlInGaP).
These different materials should show up as different I-V diagrams.

If there was a table indicating period->used material->voltage drop, we could measure the voltage drop and then work backwards to get an (approximate) time of manufacturing.

Perhaps the OP would care to elaborate his theory.

[Gyro]

For a long time, you could buy standard LEDs, high brightness LEDs, superbright LEDs, ultrabright LEDs etc. There was a huge overlap.

[BrokenYugo]

IIRC milspec 1970s style LEDs in the hermetic metal can package are still in limited production and available. Huge overlap indeed.

[RAPo]

There goes my theory.

[Gyro]

Old Red GaAsP LEDs are worth keeping around for low voltage bias supply 'Zener' use. In measurements taken by Morgan Jones for his Valve Amplifiers book, he found that these LEDs (forward biassed) have a dynamic resistance of around 4.3R @ 10mA, compared to 6R for a 1N1418 1N4148 - and of course you would need three in series to achieve the same voltage (18R). True low voltage Zeners fare very badly in comparison, LEDs are much quieter too.

[SiliconWizard]

I see the idea, but I'm curious about what you're going to get in practice. Obviously, the first step will be to get ahold of batches of LEDs from various periods and do some stats.

[AVGresponding]

You could try hanging out in the same light-bars, choose one at the same wavelength as yourself, and put your best voltage forward