The first thing you do when you learn a new programming language or platform is to write a “hello world” application. This requires something like a “printf” function. That’s not so straightforward when it comes to microcontrollers – where will the output of the “printf” go? That’s where serial communications come in. Arduino users have it easy – they just need to use Serial.print(). But the situation is not so bad if you are using a standalone microcontroller – just choose a chip like ATmega168 which has USART – hardware support for serial communications.

The code needed to transmit serial data is very simple, and the datasheet has most of what you need:

Here is the schematic of a simple setup that will let you send debug data (strings) from an ATmega168 to your computer. If you are completely new to AVR programming, I recommend that you read Hackaday’s tutorial on the subject. Note how the TX/RX lines are flipped when you connect it from the ATmega168 to the FTDI adapter.

A few things to remember in order to for this to work correctly (the Makefile in the GitHub link below takes care of all this):

  • For getting a 9600 baud rate, the chip needs to run at 8 MHz, and for this, you need to unset the CKDIV8 fuse.
  • For full sprintf formatting support, some additional flags are needed in the linker.


Serial Communications with the ATmega168 1


Here is a photo of the setup that I used:


Serial Communications with the ATmega168 2


Here is what the output looks like on CoolTerm, a serial monitor that I use on my Mac:

Serial Communications with the ATmega168 3


Having a “printf” function is very handy for debugging your projects – so choose a chip that will let you support this functionality without too much pain.


Here is all the code used in this project – do pay attention to the flags in the Makefile.