74HC595 driving 12V RGB ledstrips

Hi!

I would like to drive 5x 12V RGB ledstrips (only 5x3 leds) on 2x 74HC595 shiftregisters (using
the ShiftPWM library). The ledstrips have an common anode. They should stay individual
addressable.

Do I have to use transitors for each R,G,B line of all the 5 ledstrips?
That would be 15 transistors on the pcb!? :frowning: Is there a better solution?

Regards and thanks in advance!

Do I have to use transitors for each R,G,B line of all the 5 ledstrips?

If using a 74HC595 then yes.

That would be 15 transistors on the pcb!

Yes.

s there a better solution?

Depends on how you define better, lower cost or lower component count?
What current does each of these strips take? That will determine your other choices.

Thanks for your reply!

I think "better" would be less component count. I dont have much space
since Im bulidng a little mood lamp.

I saw the ULN2803 but that chip would assume that the strips have a
common kathode, right?

Each strip takes approximately max. 30-40mA while fading through all
possible colors.

I saw the ULN2803 but that chip would assume that the strips have a
common kathode, right?

No that is a current sink so it assumes you have a common anode strip.

Each strip takes approximately max. 30-40mA while fading through all
possible colors.

Are you sure? It doesn't sound much, one LED normally takes 20mA, you need to know the maximum current and you need to know it precisely.

No that is a current sink so it assumes you have a common anode strip.

So that means that I can connect the outputs of the chip directly to the kathode of each R,G,B lines?
Sorry... to be honest in the ULN2803 schematics I believed that the chip has common GND and
all the outputs are positive. :blush: Do you think I can use that chip?

Are you sure?

Hmm... I thought fading through all colors would give an good average of the current. Its actually jumping between values and
I saw 30-40mA max.
At what color would be the max. current? All PWM ports on 255?

Sorry for my noobienes :~

I measured with all RGB ports set to 255 without fading and
I get 47.5mA for one RGB strip.

The strip is rated at 1.2A when 1 meter (100cm) is lighted but im using just 5cm. So
converting the current would assume:

1200mA/100cm = 12mA -> 1cm = 12mA -> 5cm = 60mA

Sorry... to be honest in the ULN2803 schematics I believed that the chip has common GND and
all the outputs are positive.

No all the outputs are not positave, the outputs pull down to ground. That is the output switches between being connected to ground and being connected to nothing. There is no voltage output from this chip.

That current looks better.
There are lots of chips you can use at that sort of current. Iin addition to the one above you can also use a TLC5940.

Thanks Mike for your help! :slight_smile:

Should I put limiting resistors before the inputs of the ULN2803 or
could I directly route the traces from the 74HC595 to these inputs?

Im just curious:
What purpose has pin 10 (COM) of the ULN2803 if not VCC?

Use TPIC6B595 shift registers instead of 74HC595. They can drive your common-anode LEDs directly, all you need is the series resistors.

What purpose has pin 10 (COM) of the ULN2803 if not VCC?

It is the common connection for the reverse bias diodes used when switching inductive loads. Normally you have no need to connect pin 10 unless you are driving things like a motor.

Should I put limiting resistors before the inputs of the ULN2803 or
could I directly route the traces from the 74HC595 to these inputs?

It depends on how you are wiring things and how you are driving the multiplexer. Generally the resistor goes in the line that will see only one LED on at any one time.