Re: [CAD_CAM_EDM_DRO] Single-wire step and direction
Posted by
Alan Rothenbush
on 2005-06-02 11:27:50 UTC
Since you did actually solicit opinions, my largely uninformed one follows ...
1. I don't think it would be _more_ reliable, and perhaps even less.
. You are placing enormous faith in that timing cap not drifting in value with
time, and equal faith that some momentary delay in the PC (from some
unexpected interrupt?) never occurs such that a CW pulse would become a CCW
pulse.
. The voltage issue is irrelevant. If you so horribly miswire things that
you'd blow up the pair of optos in your current scheme, you'd still blow up
the single opto in your proposed scheme.
. If crosstalk between signals is solved in the new scheme with twisted pair
wiring, it would be solved in the current scheme with twisted pair wiring.
2. I'm not sure cheaper is worth discussion.
A 3 axis CNC mill project done on a real budget begins with an $800 Taig mill,
adds $350.00 worth of Geckos and another $150.00 to power and house the
controller, to control at least a hundred dollars worth motors. You'd use
EMC (because it's free) with the old PC from the closet (because it's also
free).
Even so, you're still into this for $1400.00 !
Frankly, saving the three optos to reduce the price to $1397.00, while greatly
reducing your choices of controller software just doesn't seem like all that
big a win.
3. Easier to use .. man, if you can't figure out, from all the help that's out
there, and all the lovely drawings done by all the kind souls in this group,
how to connect up pin x of your PC parallel port to Pin x of your Gecko, this
project is NOT for you. Go find some simpler hobby.
4. Freed up interface bits. Now in this I (half) agree with you, as there
never seems to be enough.
However, the writing is clearly on the wall; the parallel port _is_ going to
disappear one day, making even those four extra bits a short term measure at
best.
The future of this business/hobby is in external pulse generators connected
via USB.
So why not use USB for the I/O as well ?
There are any number of cheap microcontrollers on the market now that do USB,
simply and easily. And there are drivers (for Windows, at least) that allow
USB communication to be done as easily as old fashioned serial communication;
indeed, some of the drivers work in exactly the same manner, allowing
old-fashioned serial I/O programs to run unmodified over USB.
First we talk Art and the EMC guys into adding I/O via "serial" as well as the
existing parallel methods, the serial being old-fashioned RS-232 or
new-fangled USB.
Then SOME of the smart electronics types in this group will be sure to jump on
this and bring out products (heck, repackaging an Atmel USB-AVR development
board is all we're talking about), and there's our new, completely backwards
compatible standard.
My two cents.
Alan
--
Alan Rothenbush
Academic Computing Services
Simon Fraser University
Burnaby, B.C., Canada
Before me things create were none, save things
Eternal, and eternal I endure.
All hope abandon ye who enter here.
1. I don't think it would be _more_ reliable, and perhaps even less.
. You are placing enormous faith in that timing cap not drifting in value with
time, and equal faith that some momentary delay in the PC (from some
unexpected interrupt?) never occurs such that a CW pulse would become a CCW
pulse.
. The voltage issue is irrelevant. If you so horribly miswire things that
you'd blow up the pair of optos in your current scheme, you'd still blow up
the single opto in your proposed scheme.
. If crosstalk between signals is solved in the new scheme with twisted pair
wiring, it would be solved in the current scheme with twisted pair wiring.
2. I'm not sure cheaper is worth discussion.
A 3 axis CNC mill project done on a real budget begins with an $800 Taig mill,
adds $350.00 worth of Geckos and another $150.00 to power and house the
controller, to control at least a hundred dollars worth motors. You'd use
EMC (because it's free) with the old PC from the closet (because it's also
free).
Even so, you're still into this for $1400.00 !
Frankly, saving the three optos to reduce the price to $1397.00, while greatly
reducing your choices of controller software just doesn't seem like all that
big a win.
3. Easier to use .. man, if you can't figure out, from all the help that's out
there, and all the lovely drawings done by all the kind souls in this group,
how to connect up pin x of your PC parallel port to Pin x of your Gecko, this
project is NOT for you. Go find some simpler hobby.
4. Freed up interface bits. Now in this I (half) agree with you, as there
never seems to be enough.
However, the writing is clearly on the wall; the parallel port _is_ going to
disappear one day, making even those four extra bits a short term measure at
best.
The future of this business/hobby is in external pulse generators connected
via USB.
So why not use USB for the I/O as well ?
There are any number of cheap microcontrollers on the market now that do USB,
simply and easily. And there are drivers (for Windows, at least) that allow
USB communication to be done as easily as old fashioned serial communication;
indeed, some of the drivers work in exactly the same manner, allowing
old-fashioned serial I/O programs to run unmodified over USB.
First we talk Art and the EMC guys into adding I/O via "serial" as well as the
existing parallel methods, the serial being old-fashioned RS-232 or
new-fangled USB.
Then SOME of the smart electronics types in this group will be sure to jump on
this and bring out products (heck, repackaging an Atmel USB-AVR development
board is all we're talking about), and there's our new, completely backwards
compatible standard.
My two cents.
Alan
--
Alan Rothenbush
Academic Computing Services
Simon Fraser University
Burnaby, B.C., Canada
Before me things create were none, save things
Eternal, and eternal I endure.
All hope abandon ye who enter here.
Discussion Thread
Mariss Freimanis
2005-06-01 22:14:48 UTC
Single-wire step and direction
Alex Holden
2005-06-02 03:09:55 UTC
Re: [CAD_CAM_EDM_DRO] Single-wire step and direction
Alan Rothenbush
2005-06-02 11:27:50 UTC
Re: [CAD_CAM_EDM_DRO] Single-wire step and direction