Hi Charles,
 
I am studying the possibilities of having 3 axes with step/dir and encoder 
signals A/B/I to control servos. It's a shame about the eQEP1, I saw the 
pin overlapping too  :(

I have followed your work with PRU's and I'll try by this way. I think it is 
possible 
to use the PRU0 for 9 encoder signals and PRU1 to step/dir, 6 signals, although 
I think  it will not be compatible with Bebopr for example, but make my own 
interface 
board not think it was a problem. 
 
I have to study in depth the programming of PRU for encoders.  I think you 
have something about this (PRU encoder program). Is it working properly? 
maximum 
frequency that you got reading? any problem if I want to read 3 encoders at 
time?

Thanks a lot.
 
 
 

El sábado, 1 de febrero de 2014, 1:02:31 (UTC+1), Charles Steinkuehler 
escribió:

> This is a heads-up to anyone working on BeagleBone based interfaces for 
> LinuxCNC.  I am working on code to support encoders on the BeagleBone, 
> allowing true servo operation (within limits).  Encoders can be 
> supported on the BeagleBone via hardware or software (via the PRU). 
>
> There are three hardware encoders on the BeagleBone: 
>
> * eQEP2, on P8 pins 11, 12, 15, 16 
>
> * eQEP0, on P9 pins 25, 27, 41.1, 42.1 (BeagleBone Black only) 
>
> * eQEP1, on P8 pins 31, 31, 33, 35 (conflicts with HDMI) 
>
> The eQEP2 pins conflict with the typical BeBoPr step/dir pinout (re-used 
> by several CNC capes), but could be used if desired, since step/dir pins 
> from the PRU can be freely assigned. 
>
> The eQEP0 pins do not conflict with the standard BeBoPr pinout, and 
> might make a good primary encoder input for the spindle or other 
> high-speed axis. 
>
> Sadly, eQEP1 cannot be used without disabling HDMI.  :(  I don't 
> generally intend to run without a display, but I'll make sure the HAL 
> BeagleBone encoder logic will support all three hardware encoders if you 
> don't need a display. 
>
> In addition to physical hardware support for encoders, it is possible to 
> support encoders via the PRU.  I expect to be able to fairly easily 
> support up to 100 KHz or so count rates (25 KHz encoder A/B signals), 
> and may be able to go as much as 10x faster, depending on how complex 
> the PRU code becomes and how many channels are enabled.  Regardless, the 
> encoder A/B/Index signals will need to be on direct PRU input pins.  On 
> the BeagleBone Black, PRU0 has 9 direct inputs available on P9 (only 7 
> for the BeagleBone White), meaning it would be possible to support 3 
> encoders with index, or 4 plain encoders on the 'Black. 
>
> Feel free to contact me directly if you are working on a design and are 
> concerned about your pinout. 
>
> -- 
> Charles Steinkuehler 
> cha...@steinkuehler.net <javascript:> 
>
>
>

-- 
For more options, visit http://beagleboard.org/discuss
--- 
You received this message because you are subscribed to the Google Groups 
"BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to beagleboard+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to