MSP430FR5994: TimerA1 Frequency using only CCR0

DeMarcus Cousins

Prodigy 30 points

Part Number: MSP430FR5994

Hi, I have the following TimerA1 configuration for a simple square wave generation program:

SMCLK is selected as clock source
Source for SMCLK is DCO with DCOFSEL_6 and DIVS=8, therefore yielding a 1 MHz clock for SMCLK in theory
Up Mode

Whenever there is an interrupt, I toggle a pin.

However, when I set CCR0 to 2500 I get a frequency of 200 Hz on the oscilloscope.

This would mean that:

200 Hz = ( (f_clk)^-1 * 2500)^-1.

f_clk = 500,000Hz

Why is this, shouldn't the frequency read from the oscilloscope be 400Hz in this case, since SMCLK=1MHz? I know I am missing something. Thank you for the explanation in advance.

Edit 1: Also as a note, for TimerA1:

ID = 0
TAIDEX = 0

So there shouldn't be any more division at the TimerA level...

Edit 2: Edited question for clarity

5 months ago

0 Yuhao Zhao 5 months ago

TI__Expert 6110 points

Hi Demarcus,

There are some example code similar to your application.

Could you first refer to these codes:

msp430fr599x_ta0_02.c (ti.com)

Thanks,

Yuhao

0 DeMarcus Cousins 5 months ago in reply to Yuhao Zhao

Prodigy 30 points

Thank you for the reply. In this given example (msp430fr599x_ta0_02.c ), what would be the frequency of the square wave output to P1.0?

+1 Bruce McKenney47378 5 months ago

Guru 89970 points

Toggling the GPIO contributes a half-cycle to your output, so it takes 2x timer cycles to generate a full-cycle in your output.

To get your target (400Hz) run the timer twice as fast (CCR0=1250-1).

0 DeMarcus Cousins 5 months ago in reply to Bruce McKenney47378

Prodigy 30 points

This makes logical sense. For my purposes, it should be seen as a multiplier for CCR0 rather than the frequency of TimerA changing, because the frequency of TimerA is not changed.

**Attention** This is a public forum

MSP low-power microcontrollers

MSP low-power microcontroller forum

MSP430FR5994: TimerA1 Frequency using only CCR0