Is there anything wrong with this circuit?

I created a circuit from this example: http://www.instructables.com/id/Power-LED_s---simplest-light-with-constant-current/step2/Specs-Function/

I basically just put 3 of them in parallel because I am using a common anode RGB Led.

Can anyone tell me if there is anything wrong with this circuit? Also, would I be able to switch each individual LED on and off by placing a TLC5941 like in this:

Where the 3 tlc5941 chips are actually one chip only using different channels.

I could be thinking about this completely wrong so any feedback is greatly appreciated.

How I think it works is than when the TLC5941 channel is open (no impedance?), the led will be off because all the current through R1 will just go to ground, and when the channel is closed (Very high impedance?), it will act like the chip isn't there and Q2's gate will be saturated.

Yes that looks like it will work.
However, as the TLC5941 is operating backwards, that is when an LED is fully on your LED will be fully off.
This has the effect, as others have noted, that you can’t get the LEDs to go off completely. This is because fully on for this chip still has a period when the LED is off, but this is so short it is not noticeable as the LED is very bright. However when this is inverted you will still be able to see a faint flicker.

Thanks for the reply. I understand what your saying about the light not fully turning off. Would I be able to put a diode with a very small breakdown voltage in between the tlc5941 and where the collecter of Q1 connects to alleviate that?

Or is there not simple way to get past that? Im not too worried about the flicker affecting the color mixing, I am more concerned about not being able to turn off the individual lamps. Thanks again for your help.

Would I be able to put a diode with a very small breakdown voltage in between the tlc5941 and where the collecter of Q1 connects to alleviate that?

No because this is a full size pulse, the chip produces PWM so anything you do to get rid of it will also get rid of the ability to light up the LED normally.

I think the only way round it is to add another transistor between the TLC chip and the constant current circuit to turn the drive the right way up.

Alright, I guess ill test it out when my parts get in and if I need to put in another transistor I will. If only they made chips like the TLC5940 that could source a bit of current... I haven't found any yet though.

As always, thanks for the timely advice Mike!

If only they made chips like the TLC5940 that could source a bit of current

It is much harder to make a circuit that sources a constant current than one that sinks a constant current. Also you can operate it under a wider variation of loads because the only way to source more current is to up the voltage and there is a limit given by the supply voltage. Where as you can sink as much as you want simply by lowering the impedance to ground.

That's why you see them all having sinks, normally it doesn't matter much.