Sync mode musings was Re: New Article Posted

Dave,

Syncronous mode is an attempt by Allegro engineers to address the
noise issue inherent with "fixed time" chopping. (You can read posts
by Mariss in the archive to understand why this happens, and how his
drives use a feed-forward component to address it.)

For the rest of us using the typical chipsets, it--noisy drives--
will remain a problem; as fixed on or fixed off time Chopping is
pretty much the standard for National, STmicro, Allegro, Sanken,
etc. driver chips and/or controller chips.

I've written that you can sometimes get this at least a bit under
control by various means: board layout, component selection, motor
matching, and so on. But as long as the chopping time is fixed,
there's GONNA be noise, IMO.

So back to the Sanken chip (note that sync mode of allegro/sanken
crosses bipolar/unipolar product lines.) Also that "sync" when
applied to STmicro--formerly SGS Thomson--chips means something
different.

Anyways, we need to remember that most of these chipsets are
designed to be used in EMBEDDED control situations where the
processor supplying the steps is fully "aware" of the environment
and things like shaft position in relation to needs for quiet, etc.

IOW, these things are added in by the chipset designers for users
who are NOT necessarily doing what we do in CNC. A printer, for
example. It--the embedded controller--knows when the head is at
rest, so we can be sure to stop in an "appropriate" position to
allow the sync mode to work as designed. And still use the benefits
of microstepping DURING the head traversals.

I often try to remind folks that CNC--as we use the term here--is a
relatively small SUBset of the MUCH larger motion control industry.

When we take these "specific use" chipsets and apply them to
our "general" needs/use CNC; we must take care to see that we don't
misuse these added "features". Because for most of us using a PC
controller, there IS no way to KNOW which step a given toolpath
will "stop" upon for "T" amount of time. Even closing the loop with
servos' does not do that for us.

BTW, your last sentence below is true of nearly all microstepping
applications AS WE USE THEM HERE IN CNC. The microstepping does NOT
increase our actual ability to stop at a given position against
load; we'll OFTEN be "up against" the fullstep granularity of the
motor. Many times not even realising it. The reason we use
microstepping in our cnc machines at this level--for the most part--
is to reduce resonance issues, and as you've empirically observed to
increase "performance" (speed) due to the effects of this reduction.

I get a kick every day still about how the actual needs for most of
this stuff is below what many folks think it is...
Which is good because most of the time, folks aren't getting
the "specs" they think they are, and think they need...
Which is okay, because most of the time, it's NOT truly needed'; so
the fact it isn't happening goes unnoticed!<G>

Hope this helps,

Ballendo

>In CCED, Codesuidae <codesuidae@...> wrote:
>I'm curious about the problem with sync mode and lose of steps in
>microstepping mode. Is this as a result of the reduced holding
>torque at non- full or half step positions? I always wondered how
>it was that the computer could run with the controller in microstep
>mode without knowing where the full and half step positions where.
>It always seemed to me that it would be highly likely to stop in
>positions the stepper couldn't hold, possibly resulting in lost
>steps.
>Dave K