Nuts & Volts - February 2006

(MC68HC908MR16CFU) and set the parameters for the PLL. I then performed a compile and link (called a make by the IDE) operation, which generated a Cpu.c startup source file and a skeleton nutsvolts.c application source file. Various other .c and .h support files were also generated.

At this point, I decided that everything looked okay. So, I once again used the Bean Selector and

accessed the Port I/O area of the CPU Internal Peripherals folder. I double-clicked on the BitIO Bean to add it to my project. Then, using the Bean Inspector, I named the newly included BitIO Bean LED_BIT. Since I only want to make sure I can compile, link, load, and debug the project code, I went with the LED_BIT Bean’s default I/O pin setting of PB2, which happens to be pin 1 of the MC68HC908MR16.

In the Methods area of the LED_BIT Bean, I told the Bean to only generate code to set and clear the

Port B I/O pin. I then recompiled the project, which brought in the new LED_BIT Bean and its associated code. As you can see in Listing 1 (Listing 1 is available at the Nuts Volts website; www.nutsvolts.com), I added my LED blinker code to the nutsvolts.c application source file, which uses the LED_BIT LED BIT_ClrVal and LED_BIT_SetVa macros and a homebrew for-loop delay routine. Everything compiled and linked just fine. I cheerfully clicked on the CodeWarrior IDE’s Debug icon.

UH OH

The Cyclone Pro drivers loaded and the Cyclone Pro contacted and identified the MC68HC908MR16 MCU that was planted on the solderless breadboard. According to the various popup messages, the MC68HC908MR16 erased and programmed successfully. However, the program didn’t seem to be running and the LED was not blinking. I rechecked my MCU module wiring and traced out all of the solderless breadboard connections. I even swapped in a new 4.9152 MHz crystal.

The stupid thing still would not run. I was at the point of building up another MC68HC908MR16 MCU module. However, I decided to debug the code to see if everything was working as designed. From experience, I know that even the simplest of LED blinker code can harbor a nasty eat-you-alive bug.

After single stepping through the Cpu.c code you see in Listing 2 (Listing 2 is also available at the Nuts & Volts website.), I noticed that the program was hanging while waiting for the PLL to lock. So, I commented out the whil (!PBWC_LOCK) statement and its associated braces and recompiled the project. The LED blinked as it should have after reloading the MC68HC908MR16 with the new code sans the PBWC_LOCK statement.