On Tuesday 25 August 2015 09:19:15 Marius Liebenberg wrote: > Thanks all for the help. I used a Teensy 3.0 bord from the drawer and > it works great. I am reading a 2khz signal from the parport and > converting it to a 40khz signal. Duty cycle is from 0% to 99%.
And this is what I love about this list. I have long since given up trying to count how many ways a cat can be skinned. So the education continues apace. I would not have thought of that because I didn't have a suitable bit of hardware laying around. I have to give credit where inguinuity (my speal cheeker fails this one) solves the problem in an unusual way that makes perfect sense. Deserves a tip of the hat. Thanks Marius. > >I have not checked the AVR but most micro controllers have timers > > with hardware support or duty cycle measurement. XMC4xxx or STM334 > > devopment boards have high resolution timers equal to a frequency > > about a few gigahertz. > > > > > >On Tue, 25 Aug 2015 21:09:08 +1000 > > > >Erik Christiansen <[email protected]> wrote: > >> On 25.08.15 10:30, Marius Liebenberg wrote: > >> > I have a couple of Teensy 3.0 boards in the drawer here and they > >> > >>are > >> > >> > pretty fast enough I think. > >> > I am a bit allergic to the USB interface when one uses high > >> > current switching devices on the machine. The USB interface is > >> > not that reliable. > >> > >> It would not be hard to use one Arduino timer in capture mode to > >>measure > >> M/S ratio of a low frequency PWM input (from LinuxCNC), then > >>replicate > >> the ratio on a higher frequency output on another timer. If the > >> ratio of > >> input and output frequencies were a binary multiple, then dividing > >>the > >> input mark to fit the higher frequency output would only be a > >> right shift of the requisite number of bits. (The AVR does not have > >> hardware > >> multiply, but software multiply would be OK, if preferred, if the > >> PWM ratio doesn't change at a crazy high rate.) > >> > >> An interrupt on "capture complete" would be infinitely more > >> efficient than any polling hack, for input PWM measurement. The > >> output timer would > >> just run with auto-reload, responding to a new M/S ratio in the > >> next cycle > >> after it is written to the hardware register. > >> > >> A 20 MHz part could run 10 bits of count at 20 kHz, giving 0.1% > >> resolution. Go for something like 9 bits, 40 kHz, and 0.2%, maybe? > >> > >> But mebbe you have all that figured, already. > >> > >> Erik > >> > >> -- > >> Since late last year, Saudi Arabia has been flooding the world > >> with cheap oil in > >> an attempt to squeeze the life out of America's nascent shale oil > >>industry, > >> sending crude to six-year lows around $US40 a barrel. > >> - > >>http://www.abc.net.au/news/2015-08-20/qantas-profit-propelled-by-low > >>er-fuel-prices/6711174 > >> > >> > >>-------------------------------------------------------------------- > >>---------- _______________________________________________ > >> Emc-users mailing list > >> [email protected] > >> https://lists.sourceforge.net/lists/listinfo/emc-users > > > >--------------------------------------------------------------------- > >--------- _______________________________________________ > >Emc-users mailing list > >[email protected] > >https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
