Configuration timer interrupt 1ms

[mahmoud_dhaiwi]

Posted on June 03, 2015 at 11:24

hello,

I use the stm32f103VD,Clock 48MHz from the 8MHz HSI.Can you help me and tell me if I configured the timer 3 correctly for counter each 1ms ?I am not sure it's correct because the iterruption comes each 0.000999 ms ! I didn't understand why. any help plz ?the code:    //enable the timer clock    RCC_APB1PeriphClockCmd(RCC_APB1Periph_TIM3, ENABLE);    //compute the prescaler value    PrescalerValue = (uint16_t) (SystemCoreClock / 1000000) - 1;    //time base configuration    TIM_TimeBaseStructure.TIM_Period = 1 * 1000;    TIM_TimeBaseStructure.TIM_Prescaler = PrescalerValue;    TIM_TimeBaseStructure.TIM_ClockDivision = 0;    TIM_TimeBaseStructure.TIM_CounterMode = TIM_CounterMode_Up;    TIM_TimeBaseInit(TIM3, &TIM_TimeBaseStructure);    //enable timer interrupt    TIM_ITConfig(TIM3, TIM_IT_Update, ENABLE);    //enable timer counter    TIM_Cmd(TIM3, ENABLE);thanks #stm32-timer-inerrupt

[Tesla DeLorean]

Posted on June 03, 2015 at 14:49

iterruption comes each 0.000999 ms !

I'm just going to assume you mistyped that, as interrupting faster than a micro-second seems rather implausible. A micro-second short of a milli-second? Well the HSI isn't a high accuracy clock, have you looked at this in percent-error term? Consider using a more accuracy clock source. Also the Period should be 1000-1, but that will make it a tad faster still.

[mahmoud_dhaiwi]

Posted on June 03, 2015 at 15:37

Sorry ! there is an error...

'' the iterruption comes each

0.000999 s

 so each

0.999 ms

! ''

If I understand, so the period must be 1000 - 1 to have an interrupt at 1ms ??

[Tesla DeLorean]

Posted on June 03, 2015 at 16:46

Well making the number smaller will shrink the time further

999us instead of 1000us represents an error of 0.1%. HSI is specified to have error in the +/-2.5% or so range, over temperature. You might want to measure the HSI or PLL frequencies via the MCO (PA8) pin to understand what they look like.