PWM Period (timer ticks)

Bus Frequency

Timer Prescaler

PWM Frequency

×

---------------------------------------------------------------------------------------------

=

PWM Mark Time (timer ticks)

PWM Period

Duty Cycle (as an integer percentage)

×

100

---------------------------------------------------------------------------------------------------------------------------------------------

=

PWM Space Time (timer ticks)

PWM Period

PWM Mark Time

–

=

PWM Generation During Break or Software Interrupt

PWM Generation with the HC08 Timer, Rev. 0

Freescale Semiconductor

9

Eq. 1

Eq. 2

Eq. 3

Once the timer channel is configured, the PWM signal can be generated using the timer channel interrupt.

This should be configured to call the channel interrupt service routine (ISR), loading the timer compare

register with its current value added to the length desired (mark or space). This is achieved by identifying

whether the last action was a negative or positive edge (the port pin toggles on output compare), and

loading the compare register with the next appropriate value.

If the port pin is at logic one (or zero), a mark (or space) is to be added.

NOTE

In this case, the overflow flag and interrupt are not relevant as the counter

is used as free-running with roll-over.

The sample code in the appendix at the end of this document gives an example of how to apply this

principle.

Limitations

The MCU must be able to service the timer PWM interrupts in time for the next edge to be configured.

This sets an upper limit on the frequency/resolution of PWM signal that can be reproduced, and will vary

from system to system depending on the MCU application. It is the responsibility of the system designer

to ensure that the core can update the compare registers in sufficient time using the ISR under maximum

core loading conditions.

A secondary limitation is that using the timer module incurs a greater degree of core overhead, as the

timer module has to be serviced at every edge transition (interrupt). This does not happen with the

unbuffered PWM via modulo counter, as the toggling mechanism is independent of the core and code

execution. The described method of PWM generation is likely to be more suited to slower rate PWM

requirements, due to the overhead generated by having to service an interrupt for each edge of each

PWM signal.

PWM Generation During Break or Software Interrupt

The HC08 data sheet states that the timer is stopped during a break. It is important to distinguish a Break

from a software interrupt (SWI) instruction.

The Break Module (BRK) is primarily used in a debugging/development context. For example,

breakpoints in monitor mode are generated using this module. This explains why the debugger stops

program execution when an address match or BRKA bit-mask is performed.