I'm using 4 OSRAM LRTBR48G-P9Q7-1+R7S5-26+NP-68-R33-ZB RGB LEDs wired in parallel and I need to use resistors to drop the voltage going in. I'm using a voltage of 2.8v from a voltage reg and the forward voltage for the LEDs are 1.95V, 2.7V and 2.7V respectively. I'm wondering how the number of LEDs I wire in parallel relate to the resistors I should use?
Thanks,
Max

 

I'm wondering how the number of LEDs I wire in parallel relate to the resistors I should use?

Conservative design would use 1 resistor per LED. If the forward voltages are matched, you could get away with fewer, but it depends on how closely they're matched, what current you're operating the LEDs at, etc.

 

I'm using 4 OSRAM LRTBR48G-P9Q7-1+R7S5-26+NP-68-R33-ZB RGB LEDs wired in parallel and I need to use resistors to drop the voltage going in. I'm using a voltage of 2.8v from a voltage reg and the forward voltage for the LEDs are 1.95V, 2.7V and 2.7V respectively. I'm wondering how the number of LEDs I wire in parallel relate to the resistors I should use?
Thanks,
Max

You potentially have bigger problems to deal with.



The forward voltage of your green and blue LEDs could be as high as 3.4 V, which would mean that you would likely get little to no light out of them at 2.8 V if that were the case.

In general, LEDs should not be run in parallel -- each should have it's own current limiting device. As with everything, there are exceptions to this rule, but you should consider carefully whether your case really is an exception and not just a case of wishful thinking.

With just 2.8 V, regulating the current is going to be tricky because you don't have much overhead and your range of Vf is a very large fraction of it.

 

why would you wire them in parallel?

 

why would you wire them in parallel?

Well I have all 4 LEDs and I want them all to output the same colour and brightness at the same time, and it would be a lot easier for me to just output the colours that I need once than 4 times. But, if there's a better way I could do it, please let me know. Currently this is how I have it wired:

 

You potentially have bigger problems to deal with.

View attachment 323545

The forward voltage of your green and blue LEDs could be as high as 3.4 V, which would mean that you would likely get little to no light out of them at 2.8 V if that were the case.

In general, LEDs should not be run in parallel -- each should have it's own current limiting device. As with everything, there are exceptions to this rule, but you should consider carefully whether your case really is an exception and not just a case of wishful thinking.

With just 2.8 V, regulating the current is going to be tricky because you don't have much overhead and your range of Vf is a very large fraction of it.

How much overhead should I look for when I'm selecting LEDs? If I change the LEDs to a lower rated version what kind of forward voltage should I look for?

 

Why are the LEDs driven by pins labeled as NC?

Even if the pins are active GPIO, there is not sufficient current and voltage from a 3.3V MCU device to drive the LEDs.

 

The light output of an LED is approximately proportional its current. But each LED might have a different voltage at desired current. So putting the same voltage on them will not give you the same current and the same light output. This is why you cannot run LEDs in parallel.

To get the same current through each LED, you will need a constant current driver for each one.

But that is not as difficult as it sounds. You can approximate a constant current by using a higher voltage and a resistor. To get reasonably consistent currents, I would use about 12V and a resistor for each individual LED.

You would need a transistor to control each of the 3 colors, driven by the micro. Then you would wire each red LED with a resistor to one of the transistors, each green to another and each blue to the third, all with their own resistors.

 

Each LED only requires about 5-10mA continuous.
Maybe use each of the MCU pins to drive a transistor with software PWM. Then use the transistor to drive the LED(s).

 

As I said in the post above. You also need to drive them form a voltage higher than 3.3V to get any current control.

 

As I said in the post above. You also need to drive them form a voltage higher than 3.3V to get any current control.

How high should the supply voltage be? The higher the better.
What is a reasonable value? At least twice the diode forward voltage. Four times 3V is even better, i.e. 12V.

 

Well I have all 4 LEDs and I want them all to output the same colour and brightness at the same time, and it would be a lot easier for me to just output the colours that I need once than 4 times. But, if there's a better way I could do it, please let me know.

It all comes down to what you NEED.

Saying that you WANT them to output same colour and brightness is nothing but a meaningless vision statement.

What does it mean for them to be the same colour and the same brightness? Specifically, how close to the same colour do they need to be in order to be considered the same colour? The same goes for brightness?

It all comes down to how close is close enough for your application's actual NEEDs.

The closer they have to be, the more involved your circuitry will have to be.

At some point, it's isn't enough to drive two LEDs of the same part number at the same current because there WILL be both a colour and brightness difference between them. The colour and brightness will also change with temperature, both the ambient temperature and, more directly, the junction temperature of the LED. The question is whether these differences are even close to the point of being enough to matter. We have no way to know what your application's needs are, so that is something that you are going to have to determine.

 

the datasheet shows that current per LED should be 10mA (red), 5mA (green) and 5mA (blue).
if you want to drive 4 of them, total current is
red = 40mA
green = 20mA
blue=20mA

the GPIO is rated 3V and 10mA and therefore cannot directly drive the LEDs - you must use some sort of driver.
one variant is to use transistor on each of three channels.
if for example you use 2N2222 then base resistors can be chosen for some 2-5mA. so 270Ohm to 1k should do.

then you need to size resistors to be connected in series with each LED. since diodes are no linear, but you want uniform brightness/color it is a good idea to use higher voltage. at least 5V. the higher the voltage, the more control over individual current you have.

but assuming you chose to limit supply to just 5V so it can be powered by USB charger, you will need a 3 or 3.3V regulator to power the IC, but the LEDs will be powered by this higher voltage (5V).

then you need to choose individual resistor for each R/G/B LED.

then resistor is chosen by
R = (Vsupply - Vf)/If

for red LED, that means
Vsupply=5V
Vf=1.95V
If=10mA

so
Rred = (5V-1.95V)/0.01A = 305 Ohm.

and you would need one for each red element.
similarly for green LED we get
Vsupply=5V
Vf=2.7V
If=5mA
R = (Vsupply - Vf)/If
Rgreen= (5V-2.7V)/0.005A = 460 Ohm

and you would repeat this for the blue LED which would end up with same value of 460 Ohm.

you can try finding those exact values but common practice is to round it to a nearest common value such as 300 or 330 Ohm for red and 450 or 470 Ohm for green and blue.

next you need to confirm LED pinout. according to the datasheet, this LED is CA (common Anode) so your circuit in #5 is wrong.

so... you will probably want something like this:

 

Well I have all 4 LEDs and I want them all to output the same colour and brightness at the same time, and it would be a lot easier for me to just output the colours that I need once than 4 times.

Wanting 4 different RGB LEDs to behave the way you want doesn't happen by accident. The brightness of each LED color is different and there's no guarantee that the 4 LEDs you have will be matched by brightness or forward voltage unless they were binned that way where you obtained them.