Calculate the period of input PWM using the PWM input function of STM32

1. Clock part, the clock frequency of TIMER2 is initialized to 36M.

RCC_PCLK1Config(RCC_HCLK_Div2); //Set low speed AHB clock = system clock/2

RCC_PLLConfig(RCC_PLLSource_HSE_Div1, RCC_PLLMul_9); // Frequency multiplication factor is 9 PLLCLK="8"*9=72

2.TIMER initialization

TIM_DeInit(TIM2);

TIM_TimeBaseStructure.TIM_Period = 65535;

TIM_TimeBaseStructure.TIM_Prescaler = 18;

TIM_TimeBaseStructure.TIM_ClockDivision = 0;

TIM_TimeBaseStructure.TIM_CounterMode = TIM_CounterMode_Up;

TIM_TimeBaseInit(TIM2, &TIM_TimeBaseStructure);

TIM_ICInitStructure.TIM_Channel = TIM_Channel_1;

TIM_ICInitStructure.TIM_ICPolarity = TIM_ICPolarity_Rising;

TIM_ICInitStructure.TIM_ICSelection = TIM_ICSelection_DirectTI;

TIM_ICInitStructure.TIM_ICPrescaler = TIM_ICPSC_DIV4;

TIM_ICInitStructure.TIM_ICFilter = 0x0;

TIM_PWMIConfig(TIM2, &TIM_ICInitStructure);

TIM_SelectInputTrigger(TIM2, TIM_TS_TI1FP1);

TIM_SelectSlaveMode(TIM2, TIM_SlaveMode_Reset); //Reset mode to slave mode

TIM_SelectMasterSlaveMode(TIM2, TIM_MasterSlaveMode_Enable); //Enable master-slave mode

TIM_Cmd(TIM2, ENABLE); //Enable TIM2 counter

TIM_ITConfig(TIM2, TIM_IT_CC2, ENABLE); //Enable CC2 interrupt request

3. TIMER interrupt processing

void TIM2_IRQHandler(void)

{

/* Clear TIM2 Capture compare interrupt pending bit */

TIM_ClearITPendingBit(TIM2, TIM_IT_CC2);

/* Get the Input Capture value */

IC2_Value = TIM_GetCapture2(TIM2);

}

Note:

I don't know why, but after the STM32 TIMER clock is divided, the accuracy error of the clock increases with the increase of the division value. Without division, I detected the 1KHz PWM input signal, and the error was 0.5us. With 18 division, the error is unbearable. When measuring 100Hz, it is calculated based on the captured value to be 105Hz.

It is indeed as my STM32 friend said. I misunderstood the frequency division. The frequency division value = TIM_TimeBaseStructure.TIM_Prescaler + 1. So if you want to divide by 18, just set TIM_Prescaler = 17. After this modification, the measurement error is within 5/10,000. Thank you STM32 friend for your reminder!! (I will not delete the above misunderstanding, but keep it as a warning for the future, haha!)