[time-nuts] PICPET- was Affordable (cheap) COTS (etc)

Azelio Boriani azelio.boriani at screen.it
Thu Jan 23 19:48:50 UTC 2014


As usual the" hardware" peripherals integrated in microprocessors are
not really independent from the core: the capture signal must be
synchronized with the microprocessor's clock to enter the core. This
require usually a clock with double the speed you need for a real
hardware counter implemented, for example, in an FPGA or CPLD (or
double the resolution for the same clock).

On Thu, Jan 23, 2014 at 8:25 PM, Brian Lloyd <brian at lloyd.com> wrote:
> Has anyone else looked at the Parallax Propeller processor for timing
> functions? It has 8 cooperative cores and has a number of intrinsic timing
> functions for measuring intervals accurately, or for generating
> tightly-timed repetitive pulse trains, within the frequency accuracy range
> of the processor's clock. (Up to 80MHz by spec but the chip runs at 100MHz
> just fine.) One can dedicate one or more cores to performing the low-level
> timing functions while using another core to handle housekeeping functions
> asynchronously. The only real limitation I have encountered is no built-in
> A:D or D:A. (Low-pass filtered PWM is how I handle D:A usually.) The
> processor is 32-bit throughout so one can do reasonable timing operations
> using integer arithmetic.
>
> Using the Propeller to discipline an oscillator using 1pps would be
> trivial. Use the oscillator to be disciplined as the processor's clock and
> use the 1pps to gate an accumulator. (Use the internal PLL to multiply the
> reference oscillator up to 40 or 80 MHz to increase resolution in the
> accumulator.) This operation is performed with a single machine
> instruction. You just let one core sit there and process the error
> accumulation from the reference and the 1pps while another core processes
> the error and sets a third core generating the PWM error signal to correct
> the reference. Using multiple cores means that one really doesn't have to
> worry much about interrupt latency (just poll) or race conditions.
>
> --
> Brian Lloyd, WB6RQN/J79BPL
> 706 Flightline Drive
> Spring Branch, TX 78070
> brian at lloyd.com
> +1.916.877.5067
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.



More information about the Time-nuts_lists.febo.com mailing list