Using an Unknown LED

Ok... I have another question. :slight_smile:

I purchased a "Sidekick" kit from Amazon with my Arduino. (link: http://www.amazon.com/gp/product/B007B14HM8/ref=ox_sc_act_title_1?ie=UTF8&smid=A1LHQ5G6ONPXVT) I figured that, because I don't have a lot of components (e.g. resistors, etc.), it would be a good start for electrical components as I experiment.

HOWEVER... It doesn't come with documentation. I can figure out the resistors (those are obvious), the mini servo that comes with it has a manufacturer and part # on it, but the LEDs are unmarked. I know they're 2mm (from measuring), but don't know where to find a datasheet on them. There's also an RGB LED that is unmarked.

My question:
What do you do to determine the specs for a component with no documentation? (Or, if there isn't a cut/dry way, what do you do when you're in that situation?)

Thanks!

//Andrew

Compute your resistor for 10mA forward current and go from there... adjust if it's too bright or too dim.

LED information, start here.
http://www.kpsec.freeuk.com/components/led.htm

vasquo:
Compute your resistor for 10mA forward current and go from there... adjust if it's too bright or too dim.

What do I do about forward voltage?

5 - 20ma is a good range for most leds.

Forward voltage: calculate the resistance using 2 volts. Then measure the actual forward voltage with a meter.

anorton:

vasquo:
Compute your resistor for 10mA forward current and go from there... adjust if it's too bright or too dim.

What do I do about forward voltage?

Nothing really, just use a 500 ohm series resistor between the led and wire that network between ground and the 5V pin on your arduino. The led will light up if you have the polarity of the led wired correctly. Once you see it lite up you can measure the voltage drop across the led with your digital multimeter and you will then know what it's forward voltage drop value is. That can then be used to calculate a series resistor value to run the led at whatever current you wish. Most common leds are rated for 20ma maximum continous current, but they are useful indicators lights at any current from 3 ma up to 20ma. There is no reason you have to run an led at a full 20ma, many people spend way to much time worrying and fussing about that, just run them at 10ma and life gets simple. You will find that the color of the led is what has the biggest effect on it's forward voltage drop rating. Red are around 1.5vdc.

Lefty

LarryD:
Forward voltage: calculate the resistance using 2 volts. Then measure the actual forward voltage with a meter.

oh. I should have thought of that... :blush: Thanks!

//Andrew

Thank you very much for all your help (to those who responded)!

I now have my LEDs up and running, and my multimeter correlates that I'm under 10mA (and boy is it bright) with ~2V drop across LED. :slight_smile:

//Andrew

More resistance less light.

anorton:
What do I do about forward voltage?

Measure it with the diode setting on your DMM. If you don't have a DMM with a diode setting, you really should. Or, do as otherwise suggested here--pick a resistor value, wire it up, and then measure it.

danb35:

anorton:
What do I do about forward voltage?

Measure it with the diode setting on your DMM. If you don't have a DMM with a diode setting, you really should. Or, do as otherwise suggested here--pick a resistor value, wire it up, and then measure it.

Now that is a clever idea :wink:

Measure it with the diode setting on your DMM.

Only works if the dmm uses a high voltage battery (9v for example).

danb35:
Measure it with the diode setting on your DMM. If you don't have a DMM with a diode setting, you really should. Or, do as otherwise suggested here--pick a resistor value, wire it up, and then measure it.

I made myself an LED tester. I use it all the time. It's almost indispensable. It's just a current regulator set to 20 milliamps and a 9 volt battery - no power switch... and I can measure the forward voltage of any LED just by sticking my voltmeter across a lit LED.

Krupski:

danb35:
Measure it with the diode setting on your DMM. If you don't have a DMM with a diode setting, you really should. Or, do as otherwise suggested here--pick a resistor value, wire it up, and then measure it.

I made myself an LED tester. I use it all the time. It's almost indispensable. It's just a current regulator set to 20 milliamps and a 9 volt battery - no power switch... and I can measure the forward voltage of any LED just by sticking my voltmeter across a lit LED.

How did you deal with possible reverse polarity connection to the led? (retorical question, you didn't deal with it per your drawing) :wink:

Most standard leds have a maximum reverse voltage rating of only 5vdc. An internal series diode in series with the test leads output would probably be enough.

Lefty

retrolefty:
How did you deal with possible reverse polarity connection to the led? (retorical question, you didn't deal with it per your drawing) :wink:

Most standard leds have a maximum reverse voltage rating of only 5vdc. An internal series diode in series with the test leads output would probably be enough.

Lefty

As you can see from the drawing, I do not deal with reverse polarity. But, since the current is limited to approximately 20 mA, who cares if the LED reverse conducts? Won't hurt it.

since the current is limited to approximately 20 mA, who cares if the LED reverse conducts? Won't hurt it.

You don't think the voltage would play a role here?

I've burned out exactly one LED from reverse voltage. It so rarely happens, it took me a while to figure out why it no longer worked. Oddly enough, it was a PCB-mount LED that was capable of handling over 50mA, and bright enough that you don't want to look at it for long. I would've figured it would be pretty tough to kill. But, of all the 3mm and 5mm LEDs I've ever messed with, I have yet to ruin one with a current limiting resistor in place. Not to say it can't happen, but I don't even try to avoid reversing them if I don't have a way of knowing ahead of time. At 5v, anything from 220R to 10K is likely to suffice, at least for a test.