Go Back to the Basics with an LED Matrix

Addressable LEDs, LED matrices, and the like have revolutionized what kind of colorful projects we can easily make. In many cases, you can load up a library or two and turn LEDs however you wish, but what’s actually going on here? How is it possible to control, say, 64 LEDs with a MAX7219 driver, when the chip has many fewer pins that the number of LEDs in action?

The answer, as described in some detail by Norbert Heinz (AKA HomoFaciens), is a combination of two concepts. First, while we normally – and correctly – use LEDs for their light emitting properties, we sometimes forget that “D” stands for diode, meaning they only allow current to flow in one direction. Using this property, and a combination of outputs that can be set to low and high on both the row and column connections of a matrix, different LEDs can be lit with fewer outputs than the number of LEDs in action.

The problem, as explained starting at around 8:45 in the video below, is that each LED can’t be switched completely independently. If more than only row and more than one column is switched on at the same time, extra LEDs start lighting up. The solution is to scan through the matrix at a speed higher than around 1/25th of a second, causing human eye to resolve each LED as being on or off at the same time. It’s a concept you may know and/or use, but this clip is a good way to nail down the basics.