PDA

View Full Version : Keep it simple and stupid...

Ramboman
04-11-2012, 09:44 AM
I am building a light ball using 60 rgb leds.
Very easy with only one prop-usb and a 10*(6*3) matrix multiplexing, using the 28 available pins.
It works, but 3.3v is very low for the G and B parts of the leds... current is 1.5mA only.
How can I increase the output voltage to 5V or more with a minimum amount of components ?
Thanks

g3cwi
04-11-2012, 10:21 AM
I wonder if it might be possible to float the ground of the prop up by a volt or so to keep it within spec while delivering more current to the LEDs? That would take just two resistors. It's an approach that needs some thought though as there are risks involved.

Cheers

Richard

Mark_T
04-11-2012, 11:37 AM
I wonder if it might be possible to float the ground of the prop up by a volt or so to keep it within spec while delivering more current to the LEDs? That would take just two resistors. It's an approach that needs some thought though as there are risks involved.

Cheers

Richard

OK, developing that idea: Use a 4V supply and use a forward biased diode to raise the Prop's ground about the LED's ground. I'm presuming the multiplexing transistors are NPN to ground driving LED cathodes. To switch off the transistors properly you'll need to tri-state the pins controlling their bases I suspect. Hopefully the diode drop will prevent more than a diode-drop below Prop ground on any prop pin...

Erik Friesen
04-11-2012, 01:18 PM
Remember this. Simplicity and complexity are close relatives. a uln2803, or discrete transistor is cheap.

pik33
04-11-2012, 01:43 PM
Datasheet says Propeller maximum allowed voltage=3.6V. LED is nonlinear device and maybe this 0.3V will be sufficient.

Ramboman
04-11-2012, 05:25 PM
Datasheet says Propeller maximum allowed voltage=3.6V. LED is nonlinear device and maybe this 0.3V will be sufficient.

Difficult to change on a prop-usb...

Ramboman
04-11-2012, 05:27 PM
Remember this. Simplicity and complexity are close relatives. a uln2803, or discrete transistor is cheap.

May I have some drawing ?
Thanks

RDL2004
04-11-2012, 06:12 PM
Check out this Nuts and Volts article (http://www.parallax.com/dl/docs/cols/nv/vol1/col/nv6.pdf) .

Phil Pilgrim (PhiPi)
04-11-2012, 06:18 PM
It seems to me that the question boils down to this: how much voltage/current will the Propeller pin experience with a resistor-coupled LED pulled up to 5V? If the LED's forward voltage is, say 3.2V @ 20 mA, what would its forward voltage be at 500 uA (i.e. the Prop's protection diode limit)? if it's more than 1.1V (i.e. 5V - 3.9V), which it almost certainly is, there should be no harm in driving the LED directly from the Prop pin, through a resistor pulled up to 5V.

-Phil

Ramboman
04-12-2012, 07:06 AM
Check out this Nuts and Volts article (http://www.parallax.com/dl/docs/cols/nv/vol1/col/nv6.pdf) .

To be clear... the rgb leds I use have four wires : a common cathode and three anodes
The sixty leds are multiplexed... ten goups of six (times three colors)... without additional circuits... place in the balloon is limited
Very easy to program in spin using two cogs
So at one time the cathode is tied to the ground (one pin) and the anode to 3.3V (another pin)
My problem is to rise that 3.3V to 5V.
For another project, a 8*8*8 led cube, I used shift registers... I have no place...