CAD CAM EDM DRO - Yahoo Group Archive

Re: Testing the waters...

Posted by Tom Kulaga
on 2001-05-23 11:39:13 UTC
--- In CAD_CAM_EDM_DRO@e..., "Tim Goldstein" <timg@k...> wrote:
> I had talked with Mariss a few weeks ago and if I remember
correctly a true
> servo amp is the next product he plans to release.

Well that is certainly interesting...

There's been little enthusiasm from the list for adding stepper
control, as well as some threads about others already doing this (are
those things related?), so I don't know if I'm going to go that
route. As I mentioned before, I'm working with Marty on his BP mill,
and he counts as the one and only person leaning toward steppers. Of
course, since he lives only 10 minutes away from me and he's done me
a lot of favors with getting my own shop running, his opinion carries
a lot of weight.

If I were to build only one of the stepper control boards, how hard
would it be to get it working with EMC? I haven't gotten the EMC BDI
CD to successfully install, so I haven't been able to look it over
(something about a corrupt or inaccessible file error right after a
wall 'o text during the initial bootup). If I can pull this off,
I'll probably report it to the group and put all the software on the
net. Marty is on a pretty relaxed schedule to get this done, so I've
got the time to experiment.

For a servo controller, would it be better to build them as complete
4 or 2 axis cards, and users can add cards as needed? I was thinking
about a card with 2 axes of encoder input, 2 DAC channels, and a 24
bit digital I/O interface. Adding the digital I/O would be
simultaneously cheap and very useful, in that you'd get limits and
home position input for all channels, disable inputs, coolant pump
control, etc. If you really really want to run that VFD from the
board, 8 bits of digital I/O could be run into a (cheap) single
channel DAC. Does 255 steps of speed control seem like not enough?
Seeing as I get by with only 8 on my manual mill, I think 255 is
overkill.

Some other thoughts:

Would there be any value to adding either 16 bit bus support (full
length ISA) or interrupt signals, assuming the I/O chip has interrupt
outputs? Should I wire a special E-stop disable input to trigger an
interrupt?

How about differential input and index pulse detection to the encoder
portion of the board? The LS7266 chip already supports index pulses,
and differential input is as simple as adding some differential line
recievers. They cost maybe a couple bucks apiece, and make the
system far more resistant to electrical noise. However, both these
options add more complexity to the board, and until I get better
software to capture schematics and do autorouting, I shy away from
complexity unless the result warrants it.

Does anyone have any links that explain PCI bus design requirements
in a clear and simple way? I'm tired of reading white papers and
technical specifications. I don't have the training necessary to sort
through it all and determine if this is something I can consider. I
only ask because it's obvious that ISA is going the way of the dodo,
and PCI seems to be the bus standard for the foreseeable future. USB
is also on my to-do list, and I've already got my hands on the
Cypress USB microcontroller kit. Lots of possibilites with this
little guy.

Discussion Thread

Tom Kulaga 2001-05-18 18:24:30 UTC Testing the waters... Tim Goldstein 2001-05-18 18:53:50 UTC RE: [CAD_CAM_EDM_DRO] Testing the waters... Jon Anderson 2001-05-18 18:57:35 UTC Re: [CAD_CAM_EDM_DRO] Testing the waters... Jon Anderson 2001-05-18 18:59:39 UTC Re: [CAD_CAM_EDM_DRO] Testing the waters... Tim Goldstein 2001-05-18 19:08:24 UTC RE: [CAD_CAM_EDM_DRO] Testing the waters... mlichty@c... 2001-05-18 19:14:09 UTC Re: Testing the waters... IMService 2001-05-19 04:47:44 UTC Re: Testing the waters... info.host@b... 2001-05-19 07:22:42 UTC Re: [CAD_CAM_EDM_DRO] Testing the waters... ballendo@y... 2001-05-21 20:23:20 UTC Re: Testing the waters... ballendo@y... 2001-05-21 20:26:45 UTC Small tach was Re: Testing the waters... Tom Kulaga 2001-05-23 11:39:13 UTC Re: Testing the waters... Jon Elson 2001-05-23 12:38:21 UTC Re: [CAD_CAM_EDM_DRO] Re: Testing the waters... Eric Keller 2001-05-23 17:02:56 UTC RE: [CAD_CAM_EDM_DRO] Re: Testing the waters... Paul D. Fincato 2001-05-23 18:04:02 UTC RE: [CAD_CAM_EDM_DRO] Re: Testing the waters...