Newbie Level-0 LED Question

LED are diodes.
Normal LEDs do not conduct significantly below a specific voltage, but once that voltage is reached, trying to increase it will rapidly increase the current until the device destroys itself.

In other words, LEDs should only ever be used with a series resistor to set the current, or some other form on current control/limiting. They are current operated devices, not voltage operated.

The normal working current in the datasheet is what you should consider a maximum value for continuous use.

The LED datasheet forward voltage is useful to allow you calculate the series resistor - you subtract that voltage from the supply voltage to get the voltage that will be across the resistor.

eg. If using a 12V supply and a led with forward voltage of 2V, there will be 10V across the resistor.
With a 3.3V supply, around 1.3V

20mA is 1/50th of an amp so the resistor needs to be a value around 50 times that voltage; for 10mA use 100x the voltage as that is 100th of an amp.

So 500 Ohms for 20mA on the 12V supply (10V dropped on the resistor) or 1000 for 10mA.

For 3.3V directly from a microcontroller pin, 10mA probably better as they have limited drive, so 1.3V * 100 = 130 Ohms.

I normally use ultra-bright LEDs, but run them at a fraction of their full rating, 1mA at most; that puts far less load on the circuit and saves power in battery powered devices.

(You can buy LEDs that have a resistor included to work at a specific voltage, though they are a lot less common and more expensive that normal LEDs; LED replacements for panel lamps or in switches are the same, they have current limiting for a specific voltage & possibly rectifiers so the connection polarity does not matter).

thank you for reply!

Unfortunately Im still wondering why 5mA is all thats being drawn. Am i making a mistake in even caring?

I looked at maybe 8 LED data sheets and every single one of em is max 10 - 20mA. Shouldnt I be able to verify these measurment of current?

I learned some valuable stuff from your reply, for some reason Im finding myself still as confused… am thinking my LED is specd different than average.
I want to find a spec sheet that matches.

qb748 said:

thank you for reply!

Unfortunately Im still wondering why 5mA is all thats being drawn. Am i making a mistake in even caring?

I looked at maybe 8 LED data sheets and every single one of em is max 10 - 20mA. Shouldnt I be able to verify these measurment of current?

I learned some valuable stuff from your reply, for some reason Im finding myself still as confused… am thinking my LED is specd different than average.
I want to find a spec sheet that matches.

Click to expand...

The figure you see quoted in datasheets is the maximum continous current the LED is rated to withstand without failing. The LED is powerless to change the current you put though it. That current is set by the series resistor you put in series with the LED.

Here's an easy way to think about it. It is somewhat simplified but acceptable if you're just starting out. Assume the LED always drops 2V. If your supply voltage (or digital I/O voltage) is 3.3v, that leaves 1.3v for the resistor. If you use a 220 ohm resistor, you will get 1.3/220 = 5.9mA. If you want 20mA (the max current the LED is rated for), you will need a resistor that passes 20mA when the voltage across it is 1.3 volts. R = V/I = 1.3 / 0.02 = 65 ohms.

In practice you want to be on the safe side and not exceed the max current rating of the LED, so round up to the nearest standard value. You can choose 68 ohms, but I would go for 82 ohms to be safe. Also, in my experience standard 5mm LEDs do not get significantly brighter as seen by the human eye above 10mA so I usually go for 5mA - 10mA on my indicator LEDs.

Also, in reality the LED will drop slightly more voltage at higher currents. I've attached the datasheet for an RGB LED as an example. The graph below is for the red colored LED. Although the voltage changes slighlty with forward current, it is always very close to 2v (I usually use 1.8v for my calculations) so you can just assume it will be 2v or whatever. If you open the datasheet you will see that LEDs with different wavelengths (colours) drop different voltages.

qb748 said:

My issue is understanding why I keep getting 4 or 5mA as current draw vs 20mA. I did 3.3V off the esp32, used a 220 Ohm resistor and get 5mA.

Click to expand...

You have DEFINED the current by the choice of resistor and the voltage in use!

Ohms Law; the current through a resistor depends on the voltage and resistance.
Volts / Ohms = Amps.

3.3V out of the MCU:
Less the LED forward voltage (say 2V) leaves 1.3V across the resistor.

1.3 divided by 220 (ohms) = 0.0059 so around six mA. The MCU is likely giving fractionally less than 3.3V

To neared get 10mA, use a 100 Ohm resistor on the 3.3V output.

The current is totally dependant on the external circuit that feeds the LED, which are the parts you select.

qb748 said:

I figured if I set my little Alientek powersupply to 2V and see what current it draws, that Id also get the fwd current that way… sure enough it’s like 3, or 4’ish mA there.

Click to expand...

If you are testing the LED without using a series resistor then the LED should have burned up.
Perhaps your power supply is current limiting.

eTech said:

If you are testing the LED without using a series resistor then the LED should have burned up.
Perhaps your power supply is current limiting.

Click to expand...

It is not very difficult to set the voltage to the exact forward voltage on a modern bench supply without the LED getting fried. It is risky but possible.

Galgso said:

It is not very difficult to set the voltage to the exact forward voltage on a modern bench supply without the LED getting fried. It is risky but possible.

Click to expand...

But it's a REALLY silly thing to do - you need to stick a resistor in to limit the current - the accuracy required, which will change as the LED heats up, putting the LED at considerable risk for no reason.

Galgso said:

It is not very difficult to set the voltage to the exact forward voltage on a modern bench supply without the LED getting fried. It is risky but possible.

Click to expand...

If your supply has current limit set then you would be OK to do that.

But PN junctions, as they heat up, their Vforward drops by 1.8 mV / degree C, so
they can have thermal runaway if there is not a fdbk mechanism or resistor. Hence
driving with a V source asking for flame......

Regards, Dana.

Wow thank you everybody!

Yes I have my alientek voltage supply and i set a current limit, and then dialed up the voltage from 0 to 2V in mV increments… and was hoping to see the 20mA, which I did not. This was all to remove effect of resistor on circuit.

I understand that the current limiting resistor basically defines the current… and yes, the 220 (the lowest resistor I have, barring potentiometer) yields the 5-6mA after crunchng through V=IR (verified with real circuit).

I suppose Im going to reach out to starter kit mfg. and get a spec sheet on my LED.

Galgso said:

If you want 20mA (the max current the LED is rated for), you will need a resistor that passes 20mA when the voltage across it is 1.3 volts. R = V/I = 1.3 / 0.02 = 65 ohms.

Click to expand...

Yeah I think im going to set my 10k pot (with a DMM) to 65 ohm, or something lower than 220 and hopefully I will see this 20mA. Tbh, I would think the “LeD w only power supply” wouldve shown, but meh.

Also I should test out other LEDs, maybe this one is weird.

Another thanks to you all for bearing with my neurosis. I hope you all have a good, valuable day!

qb748 said:

Yeah I think im going to set my 10k pot (with a DMM) to 65 ohm,

Click to expand...

You could damage the pot doing that!

That is 1/154 of it's full value and would be dissipating 26mW; proportionally that current is equivalent to around 4W for the whole track!

It is very easy to damage small potentiometers by trying to use them to directly control current.

qb748 said:

Yeah I think im going to set my 10k pot (with a DMM) to 65 ohm

Click to expand...

I agree with rjenkinsgb, a pot should be used as a voltage divider, not as a rheostat (current limiter) unless you're working very low current at all dial settings.

Be aware, LEDs come in a very wide range, even for parts that you might think of as being equivalent.

A quick search on DigiKey for Red, T-1 3/4 LEDs gave over 600 items.
- The forward voltage ranged from 1.6V to 14V, allthough anything above ~2.5V probably has an integrated resistor.
- The mcd (light output) from 0.01 to 65000

Also, the forward voltage rang differs for colors other than red. Green, yellow, blue and white will be different.

I assume that you're just exploring to expand your understanding. (Which I thing is a wonderful idea)

Try this experiment.
Using your 220 ohm resistor, adjust the output of your power supply up until you see 20mA:
Measure the voltage across the LED and observe the brightness.
Lower the voltage to get 10mA. Repeat the measurements.
Lower the voltage again to get 5mA. Repeat the measurements.

Did the kit include other colors of LEDs? Repeat the above test with them.