Potentiometer problem..?

[PspKicks316]

Maybe this is normal. I'm not sure. Hopefully someone has an explanation for this.

Okay. I've got an RGB led with a common anode setup. 3 pots, one for each color.
I'm using PWM to drive the led cathodes.

PWM --> pot --> led --> +

I have 3 pots setup so that I can create any color I want.
They each work fine at their least resistance (led fully on).
When blue is fully on, and green is fully on, you can see that both lamps are lit.
HOWEVER, when I turn red to being fully on, both green and blue shut off and only red is on.
All 3 pots are wired in the same direction, same input source, and each outputs to a cathode.

What the hell is going on???

[jrfhoutx]

TBH I have no clue, my only guess would be that it could be overloading the circuit. if they use a common anode, and there isn't enough voltage from your source to power all 3 at the same time then it's possible that the legs requiring higher operating voltages would not operate. (at least as far as my limited understanding of electrical theory is concerned. I know I've seen it happen in cars anyway...)

Now, I know you're into PICs and you said you were using PWM, so I assume there is a PIC involved in the circuit somewhere. So in my quick, unsuccessful, search to refresh my memory on RGB LED operation I found THIS, which you may be interested in, and may help you...

[spurgurgle]

i dont really know but i might be beacause of the red in the led requirering less forward voltage........

try getting a red and blue and a green led wire them blue and green up to a power supply.... then add the red on... the blue and green will get dim... unless u resistor the red down by about 100ohm's

[GhoSt]

spurgurgle why would you add a resistor when a potentiometer is a resistor (a variable one).

[spurgurgle]

spurgurgle why would you add a resistor when a potentiometer is a resistor (a variable one).

i know what i pot is ghost.

i dont really know but i might be beacause of the red in the led requirering less forward voltage........

try getting a red and blue and a green led wire them blue and green up to a power supply.... then add the red on... the blue and green will get dim... unless u resistor the red down by about 100ohm's

in that quote im not refering to kicks's circuit wit pots in . im just saying test it out with 3 leds  red blue and green on say a 3v supply ......the green and blue will be fine until u ad in the red.

so

i guess u could say to get it to work properly u'll need a pot with a higher overal resistance.... or add in extra resistance with say a normal reistor to lower the power to the red led.

[PspKicks316]

The pots I'm currently using aren't powerful enough to completely bring down the voltage to shut off any one led color...unless red is on (I plan on getting 200ohm and 500ohm and seeing which works better).
Maybe that could affect it?

I'll add in another resistor going to the red on later on today and see if that helps at all.
The only other problem with my project is that between the 22 leds, maybe 3-4 don't change color as well. It's a premade pcb I'm hijacking, but they're wired in parallel.
I guess the only thing for that would be to rewire the circuit and wire each led separately and jump the power to the leds instead of it going through the circuit (to be safe with testing).

Oh well lol

And thanks for the link Jr. I'm not using a PIC (using an Atmel AVR) but I may be able to figure out some things from that source code.
This would be alot easier (and a lot less intrusive) if I could just get my 3 pwm timers to run at the same friggin time...
Post Merge: December 22, 2009, 09:22:11 AMReplaced pots with switches. Same thing. Tried resistors, and all it does it lower the max brightness.
WTF.
Post Merge: December 22, 2009, 04:28:00 PMGot PWM running on all 3 pins at the same time, but lo and behold I'm still having the lighting problem. It's not because I'm splitting the signal then.

God, this is weird.