It looks like the guy building the project gets the power from the arduino.
If I red correctly the documentation the arduino voltage regulator can supply up to 1A
The project theoretically can turn on all the 125 LED simultaneously and as they connected in parrallel and each led draws about 20mA the total current they would draw is 20mA*125 = 2.5A much more then the arduino voltage regulator can supply.
How does it works for the dude who built the project ? I can only assume he never turned on all the LEDs together
The word Instructables tells you that the dude is an idiot and is not worth looking at. Almost all of them contain mistakes.
The Arduino's regulators will not in practice provide 1A due to thermal considerations, that is it will overheat.
OK, this is a multiplexed display, multiplexing on five lines. Only one of those is active at any one time, so at most 25 LEDs will be lit at any one time, theoretically half an Amp.
The author of that article was powering it through the USB port (or perhaps directly to the 5 V terminal) from a phone power supply capable of at least that much current.
You would certainly not want to power it through the on-board regulator of the Arduino.
I did not understood what you've said, why only one of those 5 can work at any time ?
Because they are multiplexed. That means only one line is switched at any one time. It is how the circuit works. Look at this for a 2-D matrix, the principals are the same. http://www.thebox.myzen.co.uk/Workshop/LED_Matrix.html
What is the correct way to supply power to the LEDs
It depends how much current you are supplying. I would always use a PNP transistor or P-channel FET to switch current over 20mA to LEDs.
Can you recommend a better site the instructables to learn from it
Any site without the word instructables in the url has a good chance but there is a lot of rubbish out there. The AdaFruit web site has some of the best I have seen but they are understandably a bit biased towards their own products.
Grumpy_Mike:
Because they are multiplexed. That means only one line is switched at any one time. It is how the circuit works. Look at this for a 2-D matrix, the principals are the same. http://www.thebox.myzen.co.uk/Workshop/LED_Matrix.html
I did not understand what is the purpose of the current sink, or what it does.
When i connect the current sink, the source of the current is still the arduino pins, so the electricity flows through the arduino, how does he help to the arduino ?
You can see a current source as a relay between 5V and the LED/resistor; similar a current sink is a relay between the LED/resistor and ground. The current flows through the switch that is in the relay. I will call this the 'switch path'.
Now instead of relay you must read transistor or FET. The Arduino controls the on/off state of the transistor / FET and this requires only a small current; the LED current still flows through the 'switch path' of the transistor / FET and not through the Arduino.
I did not understand what is the purpose of the current sink, or what it does.
In order for current to flow through an LED one end of it has to be connected to a current source and the other end to a current sink. This completes a circuit and allows current to flow. If you simply connect an LED ( and resistor ) between an output pin and ground then the output pin is the source and the ground is the sink.
You can connect this same arrangement an other way. Connect it between the 5V line and the output pin. In this case the output pin is sinking current, the LED turns on when the output is set to low.
In both those examples you have control over one of the signals ( source or sink ) but the other line is connected to a permanent source ( 5V ) or sink ( ground ). Now with a matrix you need to be able to control both the source and the sink and depending on if it is the column or row that has the walking one on it the other line will only be sourcing or sinking the current for one LED. The walking one line has to source or sink for 8 LEDs. That is the line that needs help driving the LEDs either by sourcing current, in which case a PNP transistor or p-channel FET can be used or sinking current when an open collector transistor can be used.
The line without the walking one only has to cope with one LEDs worth of current and so can be connected directly to the Arduino output pin if you want without any other form of driver.
This only sound complex because there are two ways you can wire up a matrix and so the explanation has to take these both into account.
For the example, V+ can be anything; 5V from Arduino, IO pin from Arduino, 5V or 12V power supply.
GND is connected to Arduino GND and if you have a different supply to power the LEDs it's also connected to the GND of that power supply.
The wire marked ARDUINO is connected to one of the IO pins of the Arduino.
The current from V+ (whatever it is) flows through the resistor R1, through the LED, through the collector of the transistor to the emitter of the transistor and back through the ground to the power supply.
The control current flows from the Arduino IO pin through the resistor R2 through the base of the transistor through emitter the transistor to GND and back to the Adruino GND.
