10
\$\begingroup\$

This isn't asking the equation to determine the necessary resistor for LEDs, but more asking the general practice for selecting them.

I've seen multiple circuits that have used much higher resistor values than what I would deem to be necessary. For example, I have seen a design that has used a \$330\Omega\$ resistor for a red LED with a forward voltage of \$2V\$, and forward current of \$20mA\$ on a circuit with \$5V\$ supply. By my calculations that's twice as high as it needs to be (\$150\Omega\$).

I've read else where that this resistor is the 'playing it safe' choice, in that they use that wherever and can be confident they wont blow the LED. But is there any other reason behind it? Other than purposely halving the LED brightness.

Perhaps this prolongs the life of the LED? In my circuit I've selected the theoretical correct resistor value for each LED, but want to know if there's a practical rule I'm missing as the resistor values are quite small at times.

\$\endgroup\$
  • \$\begingroup\$ ROT - Rule of Thumb. Quite often used for non critical design values (such as the resistor for an LED) a designer will use a value they know will work. It generally speaking isn't the optimal value. Hence the variation in circuit values for the same task. Sometimes electronics design is an art rather than a science. \$\endgroup\$ – JIm Dearden Jun 17 '13 at 14:25
  • \$\begingroup\$ The title and introduction of your question are misleading. This has nothing to do with calculating the resistor value to achieve a particular current. Your question is really about how to choose the appropriate LED current. Using a resistor with a fixed supply as you seem to imagine is only one way to set this current. You need to fix this to separate concept from implementation. \$\endgroup\$ – Olin Lathrop Jun 17 '13 at 14:38
7
\$\begingroup\$

While the answers by @Passerby and @MichaelKaras pretty much cover it, there's one more thing to add:

  1. Humans perceive light intensity non-linearly: At very low intensities, we are very sensitive to a even slight variation in brightness. On the other hand, at higher intensity, the unassisted human eye is pretty much incapable of discerning differences in intensity.

    This really interesting graph demonstrates this excellently: Graph (source)

    Essentially the ability to perceive change in intensity is very high when most of the vision is attributable to the rods of the eye (scotopic vision), and drops very low when the cones are doing the sensing (photopic vision), i.e. at slightly higher luminance.

  2. Less critical but good to know: LEDs illuminate somewhat non-linearly versus current, with the graph dropping off the linear as current increases. This is most noticeable with red.

    Graph (source)

So, long story short:

The human eye cannot notice even large intensity changes at the higher levels of light that an LED generates at higher currents. Using half or even less than half of nominal rated current (20 mA typical for indicator LEDs, 50mA or more in high power LEDs) will thus work perfectly fine for most indication purposes.

In my designs, 5 mA is my preferred current for all indicator LEDs: Try it, it works great!

\$\endgroup\$
  • \$\begingroup\$ Another data point: I have a red/green bicolour indicator LED in my design, which are operating at 2.5ma and 5ma respectively. That gives a nice orange when both are on at the same time. The red side is twice as bright as the green one for the same current (and this is described in the datasheet) \$\endgroup\$ – pjc50 Jun 18 '13 at 8:35
  • \$\begingroup\$ @pjc50 Oh, absolutely! I've written about the difference in both emission and perception across the color spectrum in another answer on EE.SE long ago. The perceived intensity versus spectrum graph for the human eye is a roller coaster! \$\endgroup\$ – Anindo Ghosh Jun 18 '13 at 8:37
5
\$\begingroup\$

LED Datasheets, like most datasheets, are based on averages and target goals. Mainly, the listed led life is hours on at x current, y voltage, z temperature. For most "regular" leds, that's 20mA forward current. Drive it harder, it's brighter, but life is shortened. Drive it softer, it's dimmer, and life should be longer.

Aside from allowing the LED to live longer, there are 2 other reasons you might want to drive it with a lower current. First, because you are saving energy. If you are on battery power, every mA counts. Drive the led at a lower current, save energy, you can drive it longer. The second reason is that you don't need them at full brightness. Sometimes the difference between 18mA and 20mA can't even be seen unless you put two side by side to compare. Sometimes, depending on ambient light, distance, led type, and purpose, you could have the led running at 4mA and that's good enough. I can't tell you how many things I have with overly bright leds that I have needed to put tape over, or modded to lower the brightness (the original gameboys for one. Blinding!)

There is one other concern for choosing resistors. And that is standard values. You might not have a resistor with the same exact value that (V - VF) / I would suggest. So you choose the next bigger one.

\$\endgroup\$
1
\$\begingroup\$

Reducing the current in the LED will lower the stress on the diode and will lengthen its life. Running an LED at rated maximum current is OK if you make sure that the applied supply voltage is well regulated. But please consider that many times the rated current through an LED may cause it to emit way more light than would be necessary for some applications.

Designers will also limit the current through LEDs for some other reasons as well including:

1) Lower current operation can extend battery life for battery operated products.

2) Some LEDs will have variation of the emitted wavelength based on amount of current flowing through it. Current limiting may be used in these cases to adjust color purity.

3) When using an array of LEDs on a panel there can be variation of apparent brightness between various LEDs of different sizes, colors and part numbers. Reducing the current in the various LEDs on the panel is a common scheme used to make them all look uniform at similar brightness.

\$\endgroup\$
-4
\$\begingroup\$

Use a potentiometer and when the LED is at max brightness measure the ohms. Use a resistor within 50 or so ohms. For nine to 12 volts, I use a 330 ohm or 470.

\$\endgroup\$
  • 4
    \$\begingroup\$ Problem with your suggestion is: When do you stop turning the pot? When there's a change in emitted color (red to orange-ish to black, I've seen it)? How do you know it's at maximum brightness without taking it past there to a SED (smoke emitting diode) and then a DED? At which point, that particular LED is gone, but the next one even from the same batch need not behave exactly the same at any given current. \$\endgroup\$ – Anindo Ghosh Jun 18 '13 at 6:36

Not the answer you're looking for? Browse other questions tagged or ask your own question.