STM32F4Discovery board is already quite powerful and stacked with many handy features. On a single board you can find 3 axis accelerometer, MEMS microphone, DAC with D class amplifier that can output sound through built in audio jack. Also there are couple LEDs and button for fast access. And of course, the core of this board is STM32F407VGT6 ARM Cortex-M4 microcontroller that packs series of great features including DSP instructions and floating point unit. But eventually with such power you will miss several other handy features that could be used in your project. Think of LCD, Ethernet, RS232 interface and even camera. You could start making your own extension boards that include additional features, but why bother, since there already are several extension boards available for it. Lets go through them and see the features you could get without heating soldering iron.
For some time I’ve been using Codebench ARM GCC tools for developing software for ARM microcontrollers. As IDE I used plain Eclipse which I had to configure by myself. It worked pretty good and there is nothing wrong with this. Anyway sometimes it gets a little annoying to go keep an eye on configurations and manual settings. So I decided to give a try CooCox IDE which claims to be free and open. It seems that already supports all micros I like to use. Along to this change, I am also moving to different GCC tool collection. Codebench free tools are great, but in other hand, there are some sort limitations. One of them is release times. They are releasing their free tools twice a year, so updates and other improvements cannot reach as fast as you’d expect. Another thing I am concerned – disabled hard float functionality. If you would like to ta take advantage of floating point unit in Cortex-M4, then you get stuck. If you are not using hardware floating point module, then this tool works fine and you can stick with it. Anyway I don’t like limitations, especially with free tools, so I am switching to …
You’ve probably noticed that day by day more complex hardware gets closer to user. Of course I am talking about microcontrollers and processors. Several years ago it was a challenge to have running Linux on board. Now almost everyone have Raspberry Pi and doing crazy things. As I have noticed not many hobbyists are building their own boards to get things done. There is a wide variety of ready made developing boards depending on size, speed, price that it looks not worth spending time on whats already done. Everyone seems to be grabbing Arduino and building amazing projects. In couple years enthusiasts used to crunch ARM projects like cookies. This is thanks to choice of cheap development boards and improving free software tools. But mostly important role in this is the manufacturers. They finally noticed that hobby market can be powerful driving mechanism. Ignoring it means loosing the game. Now almost every hardware company is trying to get their development board closer to maker and hacker community. Because they like to build stuff and spread the word about it.
I got few questions from our readers about bit-band feature in ARM Cortex microcontrollers. This seems to be so obvious but also may lead to come confusion while using bit-banding. So let’s look at this feature a little bit closer. Why use bit band Simply speaking Bit banding method allows to perform atomic bitwise operations to memory areas. Why use bit banding? The simplest answer is because ARM Cortex doesn’t have something like BIT CSET or BIT CLEAR commands like most of 8-microcontrollers do. So this is somewhat a work around solution. Another question may rise – Why not using read-modify-write method? Again this method is not reliable in some cases. For instance f there is an interrupt during this operation it can cause data corruption. Other situation may occur in embedded OS when different tasks may modify same memory location. So we want a method that allows to set or clear individual bits with single instruction. This is where bit band method helps.
STM32103ZET6 prototyping board comes with LCD having touch screen capability. It is a great way to interact with device. Practically speaking Touch screen is a resistive film that can be read as regular potentiometer which value depends on touch point. Depending on voltage drop it is possible to calculate the coordinates. To make things easy normally there are touch screen controllers used that takes most of hard work – they have internal ADC that measure the voltage and sends value to microcontroller using a selected interface (I2C or SPI). In our board there is a popular ADS7843 controller used, which talks to microcontroller using SPI. After playing around I’ve put a messy code that reads touch screen coordinates. It is a glued code from various sources, so it is only to fix some results. Currently code reads a bunch of values, then averages to get rid of most garbage and then calculates screen matching coordinates. This is a trickiest part to do. You can do this empirically by getting min and max ADC values for each axis and then calculate coordinates using formula: