Re: [CAD_CAM_EDM_DRO] Re: USB ports ?

Hi Dave

Not wanting to appear to be rude, but "crap". The discussion on the emc-users
group was all about routing the step and direction signals via USB to a
USB<=>parport converter.

If it was a case of sending vector & velocity data to a smart controller, then
I would concur that it is viable. As you point out, there are a number of
devices now on the market that work. Why, there are even motion sub-systems
to plug in to ethernet and CAN bus. BUT they all use vector/velocity or
coordinate data and spit out the step/dir signals (for stepper motors). In
effect, taking the high speed loop out of the PC and putting it into a smart
controller.

Regards, Paul.

On Thursday 02 December 2004 16:16, turbulatordude wrote:
> > We just went through this discussion on the emc-users group.  The
> > general consensus was that USB parallel ports will not work because
>
> data
>
> > might be queued, thus messing up timing to the servo/stepper motors.
> > Not all of us agreed, however.    
>
> Simply put, they are wrong and you were correct.
>
> The entire project can be sent to a smart controller and just run
> without the PC even being connected.
>
> Yes, e-stop and end switches then need to be in the controller.  
> That's why is would have to be a smart controller.
>
> Also, data transfer can be buffered with 5 minutes of table time
> being sent so the buffer has between 5 and 10 minutes of table
> time.   Even Windoz would have a hard time messing that up.

--
irc channel for EMC users at irc.freenode.org
/join #emc

Pieces of seven! Pieces of seven! (Hrm, parroty error)