Re: Dedicated mb for PID

Jon, what FPGA did you have in mind? I haven't looked at all the
latest and greates, but I don't know of any that have on board DACs
or differential line receivers. Did you intednd to use an FPGA for
bus interface, quadrature decoding, and general purpose I/O (gpio)
with external DACs, opto isolators, and line receivers?

I think this will cost more than $30. If enough people are interested
in designing a low cost servo interface card like this, I'll
volunteer for schematics and FPGA code. Now all we need is PCB
layout/fab, parts, assembly, and software support in EMC.

~petev

with--- In CAD_CAM_EDM_DRO@y..., Jon Elson <elson@p...> wrote:

> Doug Harrison wrote:
>
> > The current methods of converting S/D pulses to servo control use

analog

> > tuning of the PID (or PD) loop. Gecko 320/340 amps do this quite

well.

> > Rutex has a board that will convert S/D to +/-10V for use with

most other

> > servo amps.
> >
> > Suppose a person wanted to do the PID in software. A dedicated PC
> > motherboard could be used to read S/D pulses from the control

computer and

> > compare this to encoder position via a camtronics card. Servo

output would

> > be through an analog output card, also in the dedicated MB next

to the

> > camtronics card. Software for the PID MB would reside on a

Sandisk

> > Flashdrive.
> >
> > Assuming the motherboard was free, this arrangement would cost

about $250

> > which is less than four Rutex boards (one per axis). It would

also allow

> > one to vary the permissable following error.
>
> You could also do this in one FPGA, probably for $30.
>
> Jon