Really Basic LED Question

[OilsFan]

What are some ways to control LED brightens when you switch them in and out of a circuit? I recently noticed that when I put larger than normal current limiting resistors on to Super Bright 5mm LED's, to dim them way down, that when I then switched one or two out of circuit the third one got brighter. I'm assuming this effect is always there but much less noticeable when you are running say 15mA through them rather than 0.5mA.

[singapol]

Revisit Ohm's Law for series and parallel circuits you should find the answer.

[Zeyneb]

The current through the LED controls the brightness. If you can implement an voltage to current circuit to control the LED you have more control over the brightness. I think the difference you noticed between the LEDs is due to the variation of the forward voltage. The voltage to current circuit eliminates this issue.

So I'd suggest you to make an opamp voltage to current circuit to control the LEDs. Beware of the current driving capabilities of the opamp.

Good luck!

[katzohki]

One common way of controlling the brightness of LEDs is to use a pulse width scheme.
Rather than just "on" or "off" the LEDs see a square wave. Depending on the ratio of on/off time this controls the brightness that we visually see. If the square wave is 50% duty cycle (on half the time, off half the time) then the brightness would be about 50% of max. 25% duty cycle would be 25% brightness and so on.

[Zeyneb]

1% duty cycle is still significantly brighter than analog current control.

[KL27x]

1% duty cycle is still significantly brighter than analog current control.

Despite what I see in cheap electronics, there should be a current limiting (e.g. a resistor) even when you are using PWM. "Eliminating" a resistor via PWM does not increase efficiency, despite what might appear to be the opposite on first glance.

And even if you omit the resistor, this statement is not true. It depends if the input current is near the FVD of the LED vs > ~0.5V+FVD.

[OilsFan]

I had a play around with an LM317 in both CV and CC configurations and that seems to be a much more stable way to drive the LED's at less intensity. Only problem I have is getting the green one to match the others. It's still much brighter. I think I will just replace it with one that is better matched to the others in the first place.

[katzohki]

I had a play around with an LM317 in both CV and CC configurations and that seems to be a much more stable way to drive the LED's at less intensity. Only problem I have is getting the green one to match the others. It's still much brighter. I think I will just replace it with one that is better matched to the others in the first place.

Hope you don't mind this rather late reply, but I thought I'd mention this. The human eye is most sensitive to green colors under normal light conditions. This means that, measuring by straight light intensity, a green light will appear more bright compared to a red one even though they may emit the same intensity.

https://www.nde-ed.org/EducationResources/CommunityCollege/PenetrantTest/Introduction/lightresponse.htm

[alsetalokin4017]

AP2502 linear current sink (instead of resistor) + PWM to the Enable pin of the AP2502. Try it, you'll like it.

[OilsFan]

AP2502 linear current sink (instead of resistor) + PWM to the Enable pin of the AP2502. Try it, you'll like it.

This is really interesting. I've avoided PWM because I'm working on audio circuits and I'm not sure how much effect it is going to have. It's be nice to PWM out of the range of human hearing but I'm not sure how to implement that.

[tooki]

Well what exactly are you working on? There are many ways to create PWM. Nothing about it inherently dictates what frequencies it uses.

[bcbeck96]

Most PWM will be in the audible range, but at low currents and decent shielding or isolation on your audio circuits, it shouldn't matter. There is also always i2c pots and other things like them