I am a mechanical engineering student so only have a basic knowledge of electronics. I am using some RGB LEDs and I want to control their colour temperature and brightness. I am controlling the colour temperature with PWM but want to control the brightness with another method so the colour is the same as the brightness varies. I have been looking into variable resistors and constant current reduction as a way to control the brightness but I am struggling to find a way to control the colour and brightness separately.

Is there some way to do this, or any suggestions as to where to find good information on how to do this. Thanks!

 

Welcome to AAC.

I would approach this with using a single resistor for each color. Since each different color will have a different Vf (forward voltage) using the correct resistor for each color should work nicely. I'd opt for having the resistor after the LED's and use PWM voltage input to drive the current into the LED's. Give me a min and I'll bang out a drawing.

 

Pwm will give the Voltage amount of each led, so you could also lower the height of the waveform as well which will alter the led current vua the series resistance., if you can post a circuit diagram of how you're doing it then we can modify the circuit.

 

I am a mechanical engineering student so only have a basic knowledge of electronics. I am using some RGB LEDs and I want to control their colour temperature and brightness. I am controlling the colour temperature with PWM but want to control the brightness with another method so the colour is the same as the brightness varies. I have been looking into variable resistors and constant current reduction as a way to control the brightness but I am struggling to find a way to control the colour and brightness separately.

Is there some way to do this, or any suggestions as to where to find good information on how to do this. Thanks!

The current through a single LED will determine its intensity and the relative intensities of the three LEDs determines the color.

Can you explain what you mean by "controlling"? what is the source of the control signals you are using?

 

Another way is to use a constant-current driver for each color LED.
You can have an individual control to each driver to set the color, and a summed control to each driver for the overall brightness.

How much LED current are you controlling?

 

Finished a schematic. Looking at it I realized I did not give you individual control for each color. How involved do you want this to be?

This is my first thought. I'm certain others have different approaches. Probably better than mine. This circuit allows for individual control. I still don't know exactly what you want. If you want to set each at a specific control and then adjust the brightness of the overall circuit, I'm just not sure of your goal. Each chip is a 555 timer chip. Values of components are to be determined by you, the engineer, who knows what it is you're after.

 

Adding a common PWM control for the three LEDs would require a PWM circuit that is voltage controlled, which cannot be readily done with a 555 circuit.
Linear control, as I proposed, makes separate and common control of the LED currents simpler.
It likely can be done with one or two LM324 quad op-amp IC packages, and three transistors, depending upon how you want to vary those currents.

 

To clarify a few details that seem to be incorrect: PWM control provides an adjustment of the AVERAGE POWER, not the voltage or the current. So the average (apparent) brightness of an LED varies, while the actual brightness is not changed.
So with a white led PWM CAN CHANGE THE EFFECTIVE BRIGHTNESS without changing the color temperature. Color temperature of SOME WHITE LEDs is affected by the current.
The other detail is that colored LEDs usually do not change their color aseither voltage or current are changed.

 

I am a mechanical engineering student so only have a basic knowledge of electronics. I am using some RGB LEDs and I want to control their colour temperature and brightness. I am controlling the colour temperature with PWM but want to control the brightness with another method so the colour is the same as the brightness varies. I have been looking into variable resistors and constant current reduction as a way to control the brightness but I am struggling to find a way to control the colour and brightness separately.

Is there some way to do this, or any suggestions as to where to find good information on how to do this. Thanks!

I did this for a member here that wanted LED colour bank to Fade in fade out each bank every 10sec or so, I did this with simple driver Mosfet and PWM, using a 8 pin Picmicro.

 

PWM control provides an adjustment of the AVERAGE POWER, not the voltage or the current.

When driving an LED with a series resistor, the PWM will indeed control the average current.
Of course, that also controls the average LED power.

 

This would go better if we all knew what you were currently using? Choosing color is simple enough, The RGB LED intensity is commonly referred to as fade in (increase intensity) or fade out (decrease intensity). Both color and brightness or intensity are easily controlled. It's a matter of what you are currently doing and what is available to you to get the effects you want. You could be using a uC or discreet components which some circuit examples have been provided.

Ron

 

There is a possibility that both controls can be PWM.

For example, with a common anode RGB LED you would use PWM on the low side for the color and PWM on the high side for the brightness.

I think the trick would be to use 2 different frequencies such as 6khz to 10khz for the low side and 500hz to 1khz for the high side.

This is something I have thought about many times but never tried, and I believe a micro would be best.

The problem with linear brightness is the fact that the green and blue LEDs will dim before the red so low brightness levels become predominantly red.

Placing a diode in series with the red LED can help to solve this, but that is not a perfect solution because the forward voltage difference between red and blue/green is usually about 1 volt.

 

I would think the easiest and most flexible way to do this would be with an MCU, and in software.

 

The problem with linear brightness is the fact that the green and blue LEDs will dim before the red

Why do you say that?
For all LEDs the brightness should vary approximately the same for a given percentage change in current.
The red LEDs are indeed more sensitive, but I don't think that affects the change in brightness, if all their currents are all set to give the desired color.

 

Why do you say that?
For all LEDs the brightness should vary approximately the same for a given percentage change in current.
The red LEDs are indeed more sensitive, but I don't think that affects the change in brightness, if all their currents are all set to give the desired color.

You are correct in terms of constant current drivers.

I was thinking more in the terms described in post #3 or using linear on the high side in the example I gave. (linear variable voltage)

Your concept should work fine, but I never really thought about the details of such a system.

 

I was thinking more in the terms described in post #3 or using linear on the high side in the example I gave. (linear variable voltage)

But that will still tend to give a linear change in current through the LED resistor unless the voltage is near the LED voltage.

 

Yes, but as I stated in my post it becomes a problem at the lowest brightness levels.

Believe me I have seen this in real life...I have only been building lighted displays for decades.

 

Believe me I have seen this in real life

Is that for a proportional change in all the LED currents?

 

No, it's for simple variable voltage supplied to the LED resistor array.

 

No, it's for simple variable voltage supplied to the LED resistor array.

The that could explain the variation in color you were seeing.