Let's Make Robots!

SED or Smoke Emitting Diode

This is an attempt to record the very edge to which an LED can be driven. A failed attempt.


I tried to post this as a component page, but it would not take embedded video...

The LEDs I tried are rated at 30 mA maximum. The forward voltage is between 1.9 and 2.1 V. I used a power source of 8.1 V  and a variable resistor in series that could range from 5000 to 0 Ohm.

I consider this experiment failed because I could not bring the LED to the brink of extinction (or rather ignition). It proved very difficult to bring the current through the LED to exactly 30 mA using nothing but a simple 5k linear potmeter. Once at 25 mA, it went straight up and off the scale. The difference in the angle of the potmeter was unnoticeable. I suggest someone takes this from my hands (and budget) and continues with a PWM signal generated by a micro controller. That should be much more manageable.

Great fun though! 


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
why not  use a series resistor that would bring it close to the doom level?  20omh or something? or just use a lower ohm pot.
use 'm

Good grief, I am going to have to take a crack at this. Maybe after making my lab coat.

I have a TON of pots (all leftover from school x2 because there was a paperwork issue and I got two sets *g*). For some reason I still hold the notion that every component I have is precious and I should do all that I can to not destroy them for the sake of amusement.

Maybe if I perform a sacrifice of an LED my projects will go more smoothly? 

Spending way too many brain cells on this one. But I now know not to trust these potmeters with any job that requires them to work over the full range.


This is my 10 kΩ / 270° potmeter from the video, turned to an angle of almost 0°. This measures as 230 Ω. When I push it nearer to 0°, it jumps to zero Ω. I noticed the same behaviour in ten-turn pots as well. Maybe this experiment is why logaritmic potmeters were invented. Or fixed resistors in series!


Add something small that by itself would limit current to aomething you believe would be too much for the LED, then have the pot to dial the mA you want. The multiturn pots would probably be better, more precise. And a smaller value pot as well, so more of the "swing" is needed to change the mA. Maybe a 47 ohm resistor in series with a 1k ohm pot. 8.1 volts - 2 volts = 6.1 or so to dropped across the resistor, so with the pot at 0, the 47 ohm would give 127 mA, and 1047 ohm would give 5.8 mA. Now te LED might be lit at both ends of this range, but you'd have a larger amount of the turning that would be more controllable.
All of the wirewound pots I had do the same thing, I can get to around 205 or so, then it jumps to about 18.
I understood 0% of this video. Zero.


Maybe my next video will have a warning for artsies showing the first ten seconds.8-)

Only today I discovered that another one of my videos got censored by YouTube because of my flagrant abuse of copyrighted music (I paraphrase). Instead of re-editing, contesting or finding a new hosting service, I decided to move on and forget about the whole thing.

And be more carefull next time.

The record companies are reallly pushing the limits.

Where would we be without the use of "classic: music?

The golden age cartoons all usesd it. Why do you think they we called : "Looney Tunes", "Merry Melodies",  and "Silly Symphonies"