# LED typical and forward voltages

So, I was going through the spec sheet of super bright 5mm orange LED, and it was listed that the forward voltage is between 1.8V and 2.5V. Below that there was a property listed as 'typical voltage' and had a value of 3.2V. I'm confused as to what value of resistance I must use.

• Provide said spec sheet please. – Passerby Dec 16 '15 at 16:14
• Here ya go : pcboard.ca/leds-and-led-ribbon/leds/5mm-leds/… – sixter Dec 16 '15 at 16:17
• Yeah. Strange. Perhaps they mean typical circuit voltage, this LED is used in? – Eugene Sh. Dec 16 '15 at 16:22
• Okay. But if that's so what's the relevance of such a value? – sixter Dec 16 '15 at 16:23
• I think it should read 2.3 instead of 3.2. – EM Fields Dec 16 '15 at 17:26

The important thing to realize is that CURRENT is what MATTERS to an LED. The voltage is given as a range because the voltage varies, depending on manufacturing tolerances, temperature, etc.

If you design well for LEDs, you feed them with a current source, and the voltage simply needs to be adequate for the number of LEDs you have in series on each current source. In figuring the voltage compliance your current source should have, look at the highest voltage quoted.

If you use resistors for current-limiting LEDs, use either the LOWEST value of forward voltage provided, or a bit less than that (safety margin) when choosing a resistor to set the current. This is sub-optimal .vs. a current source, as it will generally be providing less than the maximum safe current in order not to provide more than the maximum safe current under conditions (or for parts) where the forward voltage is particularly low.

• Using a current source for LEDs is totally overkill for many applications. Using resistors is just fine for indicators. – whatsisname Dec 16 '15 at 17:53
• ...where sub-optimal is just fine, being merely an indicator. I didn't say it was wrong, I said it was sub-optimal - and it is, if looking for maximum safe output. – Ecnerwal Dec 16 '15 at 18:45

Quick solution. If you want to know the forward voltage hook your LED to a power supply turned down to 0 volts and slowly crank up the voltage.

You LED will turn on once forward voltage is reached.

However, you might blow an LED up doing this so proceed with caution.

To help answer your question. Use forward current to decide on what resistor to use and take the worst case forward voltage.

• It's a pretty bad suggestion... no current limiting? Come on.. – Eugene Sh. Dec 16 '15 at 16:21
• @EugeneSh. Never had a problem with current draw, and very helpful when your LEDs get jumbled up. But like I said, its risky. – Jake Robinson Dec 16 '15 at 16:23
• Always use a resistor in series. – Eugene Sh. Dec 16 '15 at 16:24
• I think it's a bad idea. Once the threshold is reached, even for a slight increase in voltage, the current levels will increase unprecedentedly. – sixter Dec 16 '15 at 16:25
• The power supply proposal is more realistic if you have a lab supply that has variable voltage and variable current limit. Set the voltage to say 4 volts and start with the current set to 0. Connect the LED in forward bias mode and carefully increase the current limit up to allow current to flow to the LED. – Michael Karas Dec 16 '15 at 16:27

If you have doubts on the stated specs, you can make a simple Constant Current regulator to test an LED out with a common adjustable LDO.

R1 can be a fixed resistor, or more conveniently, a variable resistor or trimpot. For 20 mA, that's 62.5 Ω. (R = V/I, in this case VRef / IOut).

Simply measure the voltage across the Load/LED and you know how much it's Forward Voltage Drop at 20mA is.

• Sounds like a good idea. A constant current source will always come in handy. – sixter Dec 17 '15 at 5:41
• A small doubt though.. What's the cap used here for.. – sixter Dec 17 '15 at 5:43
• @user4493465 filtering the input. Use whatever cap is recommended by the regulator you use. It may be 1~100 uF instead. – Passerby Dec 17 '15 at 6:25