I have a LED matrix I need to light up. The problem is that if all the LEDs are lit at once, the ampere would be too high for the circuit to hold.
There are layers, and columns. Since there are fewer layers than columns, I plan to use something in the direction of a swift LED chaser. Is this a solid solution?
There are a few common ways to do it, all based on the idea of turning the 'active' LEDs on and off at a high enough frequency that the human eye perceives it as being steadily on. The easiest technique is multiplexing. With this approach you organise your LEDs into a matrix (set of rows and columns) with each LED connected at the junction of one row and one column. In this way you can turn the individual LED on by driving the column high and the row low (or the other way round, depending which way you wired the LED). This approach gives you the option of either lighting each LED on its own (you would walk through each combination of row/column turning that LED on when required) or to light a whole row (or a whole column) at once.
Obviously it is up to you to work out the power requirements and decide whether you wanted to power the LEDs individually or in rows (or columns).
Note that using the multiplexing approach has a limitation that LEDs which are notionally 'on' are actually only powered a small percentage of the time which limits the brightness that can be achieved compared to power an LED permanently on.
PeterH:
Note that using the multiplexing approach...
Thank you, Peter.
Can this be done with a shift register? When the shift register shifts out, do all the outputs give of 5V at the same time, or one at a time (switching the current swiftly between them)?