what resistor for LED @ 12V ?

[danielzink]

Currently in the middle of a 12V build.

I'm used to using 4.7k resistors on 9V builds - but what value on a 12V build so the LED doesn't fry immediately ?

Thanks, Dan

[R.G.]

(1) Look at the LED datasheet. This is where it tells you what current the LED can withstand. Almost, but not all T 1 3/4 (5mm dia) LEDs are 20ma max average current.
(2) Look at the LED datasheet. This is where it tells you the forward voltage the LED will have when conducting. Each color has a slightly different voltage. REds are *usually* 1.2 or so. The voltage rises through orange, yellow, and green to about 2V. Check the datasheets for blue, violet, pink, pomegranite, puce, lilac, etc. as there are different technologies for these. Some blue LEDs have Vf as much as 4-5V.
(3) Subtract the LED Vf from the power supply voltage. If you have a 1.7V LED, then subtract 1.7 from 12 to get 10.3V. This is the voltage across the resistor.
(4) You want a resistor that has Vsupply-Vf across it and Imax or less through it. So now it's design time. You get to pick the LED current. Light output is approximately proportional to current, and the max current from (1) is as bright as it gets. You now pick a current less than the max average current. This is usually less than 20ma, and sometimes much less, depending on the brightness of the LED per milliampere. As an example, let's pick 10ma.
(5) Now we calculate that resistor. We know the voltage across it and the current through it, so ... yep, Ohm's law time. The resistor is equal to the voltage from (3) divided by the current you picked in (4). In our example that's R = 10.3V/0.01A = 1030 ohms. Since resistors don't come in that fine a gradation, pick the nearest standard value - 1K.
(6) Evaluate how you did. Is that bright enough? Too bright? Change the resistor to make the current and brightness change. We're at half of the max current with 10ma in our made-up example. Need more? You can double it up to 20ma by halving the resistor to 515 ohms, or 510 as a standard value.  Too bright? Cut the current and brightness in about half by doubling the resistor to 2K. Or make it 4.7K, or 8.2K to make it dimmer still.

Then go look up an LED calculator page on the internet and snicker because you don't need that any more as you already understand it. 

[sean k]

Well, if you don't want to do it the hard way  by using ohms law then you've got 30 percent more voltage so 30-40 percent greater resistance will be alright. Kinda like sharpening your axe at midday then cutting the wood at twilight but whatever.

Anyways, take the voltage of the LED, say 1.7V, and take that away from 12, which is 10.3V, now divide that by the current which might be 20mA, and you get about 515 so 510 ohms. 10.3/.02= 515

Hey, am I doing that right? it's been so long since I've even used ohms law. You could put 90VDC on the 4.7k and the LED would not blow( though you might blow the resistor if the wattage rating ain't high enough... yup, you'd need a 2W). The LED mustn't be very bright at 4.7k and 9V?

Thanks for the chance to get Ohms law back in my head  ( I was writing this while RG did his but I'll publish it anyways just to remind myself about ohms law)

[composition4]

I usually just hookup up a 25k/50k whatever linear pot to the power supply positive on an outside pot terminal, then connect the LED to the other side of the power supply and the wiper of the pot.

Start off with full resistance on the pot, slowly bring it back until the LED is bright enough. Once you have it, unhook the LED and measure the resistance on the pot that you've ended up at. There's your resistor value.

To me, that's always been far more valuable than calculating values based on LED Vf - the given specs are for a nominal brightness, and usually I'll want my LED at less than that anyway. Saves batteries and retinas.

Jonathan

[frank_p]

http://www.oksolar.com/led/led_color_chart.htm

[John Lyons]

I usually just hookup up a 25k/50k whatever linear pot to the power supply positive on an outside pot terminal, then connect the LED to the other side of the power supply and the wiper of the pot.

Start off with full resistance on the pot, slowly bring it back until the LED is bright enough. Once you have it, unhook the LED and measure the resistance on the pot that you've ended up at. There's your resistor value.

To me, that's always been far more valuable than calculating values based on LED Vf - the given specs are for a nominal brightness, and usually I'll want my LED at less than that anyway. Saves batteries and retinas.

Jonathan

+1, that's what I do as well. LEDs vary in brightness and I can't keep track of what I have.
One works with 500R another with 4k7.

john

[aron]

From the FAQ:

http://www.muzique.com/schem/led.htm
http://ourworld.compuserve.com/homepages/Bill_Bowden/led.htm

[frank_p]

http://www.oksolar.com/led/led_color_chart.htm

I just posted this link because I have some weird LEDs with no data and with less common colors and was wondering where I can get Vf and Imax for them.  At least this chart give me some guidance to use the calculation approach (described by RG in that same post).  So maybe it will be useful for others.

Well the potentiometer is a god thing too: you can do the claculations the other way after to put some data on your less common LEDs.  (Thanks for the tip)

[R.G.]

I just posted this link because I have some weird LEDs with no data and with less common colors and was wondering where I can get Vf and Imax for them.  At least this chart give me some guidance to use the calculation approach (described by RG in that same post).  So maybe it will be useful for others.

Vf is easy: put a 4.7K in series with the LED and hook a 9V battery across the two. When the LED glows, measure the voltage across it with your meter. That's Vf.

If you have no data, assume that the LED max current is 20ma for the 5mm diameter "T 1 3/4" package. The max current is determined by how much heat can get out of the package, and that's almost constant for a given package.

[frank_p]

I just posted this link because I have some weird LEDs with no data and with less common colors and was wondering where I can get Vf and Imax for them.  At least this chart give me some guidance to use the calculation approach (described by RG in that same post).  So maybe it will be useful for others.

Vf is easy: put a 4.7K in series with the LED and hook a 9V battery across the two. When the LED glows, measure the voltage across it with your meter. That's Vf.

If you have no data, assume that the LED max current is 20ma for the 5mm diameter "T 1 3/4" package. The max current is determined by how much heat can get out of the package, and that's almost constant for a given package.

Thanks RG for the little "push", that is pretty logical and a clear view.  My LEDs are 2.5mm all sort of colors (light green, light yellow, dark orange, etc.)

[R.G.]

If you have 2.5mm LEDs, go look in an electronics catalog (Mouser on line if nothing else) and find out the most common max current for a 2.5mm LED package. The max current is most often fixed by the package.

[frank_p]

Okay, Thanks again...