Karl Schmidt wrote:
> There seems to be a limit as to how fast one can drive a printer port for 
> stepper control.   Has 
> anyone thought of making a PCI-E board for very high-speed I/O? Just where is 
> the bottle neck?
>   
The port, itself, is not the bottleneck.  I use the printer port for EPP 
transfers to a controller board.  The typical motherboard PP can 
transfer a byte every 800 ns with handshaking, a regular PCI port can 
get down to 600 ns.  The hangup is that software that would send fast 
step pulses would completely take over the CPU.  For faster step pulse 
rates, say much over 50 KHz, an external step generator is what you 
want.  There are several of these, some work over the parallel port, or 
PCI slot.  PCI-E is not really much faster that PCI unless you go to 
multi-lane versions.

So, my Universal Stepper Controller is conservatively designed to go up 
to 300,000 steps/second on up to 4 axes simultaneously.  (Two boards can 
be chained for up to 8 axes.)  EMC sends a velocity command to the 
board, it puts out step pulses and also counts the pulses issued (or 
optionally counts position from an encoder).  EMC reads the position 
1000 times a second and updates the commanded velocity.

Jon

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to