Shortest timer pulse width in one-shot mode

damien · ‎2014-02-16

Posted on February 17, 2014 at 06:40

Dear All,

I am looking to generate short pulses (1us) for an application. On first appearance, the datasheet looks to be very imprecise as to the shortest pulse that can be generated, i.e.:

''Pulse lengths and waveform periods can be modulated from a few microseconds to several milliseconds using the timer prescaler and the RCC clock controller prescalers.''

Aside from ''a few microseconds'', what is the shortest pulse width that can be generated using the STM32 timers (or, for that matter, GPIOs)? How is this calculated, and what other factors contribute to this timing (aside from external hardware).

-- Damien

#stm32 #one-pulse

chen · ‎2014-02-17

Posted on February 17, 2014 at 10:07

Hi

''Aside from ''a few microseconds'', what is the shortest pulse width that can be generated using the STM32 timers (or, for that matter, GPIOs)? How is this calculated, and what other factors contribute to this timing (aside from external hardware).''

Depends on the part you are using and the clocking speed.

The Timers can be set to run at the same frequency as the CPU core. In this case, the shortest pulse possible is the period of one clock cycle.

I believe that one an output pin is configured as timer output, this bypasses the IO pin clocking (max IO clocking I have seen is 50Mhz). The pin output is not totally under control of the timer.