losing steps ( I am losing my mind :)

Hi,

Well a few comments. I am trying to upgrade my netscape, - so I
will be able to fix those line wrappings. Thanks Darrell, but it
don't fix my printouts. My version is a bit old.

Stepper motors:

Well, these are cheap and simple, not only the motor, - but the
drivers also. So these are popular, and do a great job in a floppy
drive. Because they are popular, and many have success in making
them run, - they try to use them everywhere. But in fact, if you
want closed loop control, they get more expensive than servo
drives, and can't match the performance. (IMO)

But you may come very close, - if you have a thick wallet. So on
a scale from 1 to 10, where 10 would be a professional closed
loop drive for step motors, - then I say:

Is it possible to make simple circuits, from - let say - 1 to 7 on
that scale. A #2 design would maybe flash a warning, and do some
corrections, a #3 would just trip an E-stop.

In order to make anything, you have to know what the problem is.
A rapid traverse, could lose steps - due to acceleration and
braking. Okay, - if this is a problem, then the simple circuits
could maybe solve it. Nothing is perfect, - and the more you want
to push it up this "scale", the more difficult it will get. And
it will cost.

Most people on this list is "DIY" types. So if you could improve
your "home made" machine - why not try to gather some info on what
and where is the critical parts of running these machines ?

Well, I lack experience in a machine shop, - I just believe that
here is room for some simple solutions. Most of you use parts you
have hacked from something else, - and I just thought that it
would be nice if more "hacked" parts could be used :)

Earl described something that is just about the same stuff I have
mentioned, - but there is many solutions - on that scale from 1 to
7.

Matt said once that it would be difficult to have a closed loop in
EMC with steppers. What should the PID gain be ? - Well, if you
put some thought to what I have said, - then a DRO board could
read the motion. But if you could have a simple algorithm to
compensate for any lost steps, then this logic could be inserted
into EMC, without much difficulty. EMC will give the steps, and
if this little logic could just make sure that the step motors
would effectuate the steps, then you don't have to include a lot
of this into the "inverse kinematics" motion loop.

So I am back to the question: Where is the critical places -
during the cutting or milling operation ?

If it was possible to rank these "places" on - a scale from 1 to
10, then it would be much easier to see where and what could be
done.

Because there are so many "DIY" people here, then there should be
a lot of ideas for simple circuits also, at least many would find
such things interesting. "An unorthodox stepper motor cookbook ".

//ARNE