RE: Steppers or servos

[Tim Goldstein <tgoldstein-at-mlh-services.com>]

I will put a little different spin on this. I was running steppers with EMC
and I have now converted to servos that are driven with step and direction
signals just as if they were servos. What I have found is that EMC can put
out as high a pulse rate as I need. The limit I am running into for maximum
rapid moves is the maximum RPM the motors are capable of. Currently I am
having EMC output 30,000 steps a second per axis and it is having no problem
at all with this rate (gives me 180 ipm). With stepper motors I was not at
all able to achieve these feed rates. I think the reason is the step and
direction servos continue to function just fine as the pulse rates jump from
mode to mode where as the stepper will stall out.

If you want to take a look at the step and direction servos go to
www.geckodrive.com and look at the G320. I agree with Jon that a traditional
servo better capability at extremely low move rates, but I have found the
step and direction servos to be a vast improvement over steppers as a cost
well below a traditional servo setup. Heck, you can buy 8 of the drivers for
just the price of the Servo to Go card!

Tim
[Denver, CO]

> 
>  Jon, what is this fixed upper step rate, that limits EMC's
> use. You did say that did you not.
> 
> Steve Carlisle
> 
> Steve Carlisle
> 
> 
> ----------
> > From: Jon Elson <jmelson-at-artsci.wustl.edu>
> > To: Multiple recipients of list <emc-at-nist.gov>
> > Subject: Re: Steppers or servos
> > Date: Wednesday, October 18, 2000 1:02 PM
> > 
> > 
> > 
> > 
> > David Preston wrote:
> > 
> > > I am curious to know whether steppers or servo motors are better
> > > suited to EMC. Perhaps someone knowledgable in this area 
> could explain
> > > the pros and cons of each, I also have ready access to numerous
> > > smaller servo motors and was wondering if it's possible, 
> and if so,
> > > how difficult it would be to wire several smaller motors 
> in parallel
> > > driving a common leadscrew. Last but not least, can someone please
> > > tell me if EMC can easily run a fourth and fifth axis 
> (rotary i.e. an
> > > indexing head) as well as xyz?
> > 
> > Well, it's not EMC really.  EMC is one of the few programs 
> that can run
> > EITHER
> > servos or steppers.  There is a limit on step rate with EMC, now.
> > Servos don't
> > suffer this limitation.  If you want very smooth, accurate parts, I
> > recommend
> > servos.  A good servo amp will keep the motion smooth down 
> to extremely
> > low velocities (like .01"/min) without any stair-steps.
> > 
> > Wiring motors in parallel is no big deal, but COUPLING them 
> mechanically
> > 
> > in parallel is more difficult.  I really don't think it is 
> practical.
> > Surplus servo
> > motors in larger sizes are available at low cost, if you look
> > carefully.  You
> > might try eBay, where a lot of servo and motion control 
> stuff is sold.
> > 
> > EMC will soon have capability of running up to 6 axes of 
> mixed linear/
> > rotary type, as well as Stewart platforms in 6 axes, I'm told.
> > Some of the software components that prevented this have now
> > been changed to allow the extra axes.  There is one component
> > left, the kinematics routines, and work is underway now to add the
> > extra axes to that.  There will probably be several versions made,
> > and either scripts or configuration files that will allow 
> you to set up
> > the axes and offsets for the machine.  This should allow the complex
> > calculations involving the distance between the centers of rotation
> > of the rotary axes to be performed, so the XYZ-AB(C) coordinates
> > can be entered in workpiece coordinates, without regard for the
> > physical configuration of the stacked axes on the machine.  The
> > kinematics routines will compute the necessary movements to
> > put the tool where the programmer wants it on the workpiece.
> > 
> > When this is all working, it will move EMC a lightyear ahead
> > of the other low-cost CNC's, which don't do this for you.
> > 
> > Jon
> > 
>