Experimental Physics and
Industrial Control System

"J. Lewis Muir" <[email protected]> · Thu, 12 Aug 2010 09:59:28 -0500

On 8/12/10 8:48 AM, Dirk Zimoch wrote:
> Hi all,
> 
> I am looking for an alternative to our current MAXv motion controller
> solution.
> 
> Can anyone tell me about experiences with
> 
> * DeltaTau pmac
> * Newport XPS
> * Galil
> * other systems?

Hi, Dirk.

At 17-ID at the APS we use two Delta Tau Turbo PMAC2 VME Ultralite
boards and two Delta Tau UMAC MACRO stations.  (We don't use Delta Tau
controllers exclusively, but we do use these.)  Each Ultralite board is
connected via two fiber optic cables to one UMAC MACRO station.  The
UMAC MACRO stations may be deployed at a physical location fairly far
away from the VME boards if desired.  The only cabling between them is
the two fiber optic cables.

The Delta Tau Turbo PMAC2 VME Ultralite supports multi-axis coordinated
motion, motion programs, and PLC programs.

The Delta Tau software development suite is called PMAC Executive Pro2
Suite.  In my opinion, it is poorly written and buggy.  However, you can
do a lot with it and we use it.

We use tpmac <http://www.gmca.anl.gov/TPMAC2/> for EPICS control.  The
tpmac web site has more information about the hardware and setup.

Sergey Stepanov and Oleg Makarov at GM/CA-CAT at the APS have a lot of
expertise with this.  (There are a number of people at the DLS who have
a lot of expertise with Delta Tau PMAC2 controllers too.)

I would say the Ultralite board is *very* complex.  It takes a *lot* of
work just to configure the Ultralite and the UMAC MACRO station.  That
can be a negative if you just want to plug something in and go like you
can w/ the MAXv, or it can be a positive if it gives you the power to do
something you couldn't do with some other controller.

One nice thing about the Delta Tau Turbo PMAC2 VME Ultralite is that
there are various products from Delta Tau in the PMAC2 motion controller
family.  For example, there's a Turbo PMAC2 PCI Ultralite that is a
PCI-bus card.  There's also a small-form-factor board called the Turbo
PMAC Clipper.  We use the Clipper for our goniometer control and
exposure shutter synchronization with our crystal rotation axis at
17-ID.  So, if you do spend the time to learn all about configuring and
programming the PMAC2, you at least have the possibility of applying
that knowledge to other applications.

One annoyance is that there's no officially supported way to explicitly
set a motor's position.  While I tend toward the belief that one should
always home a motor to determine its true position, not everyone
believes that, and there are some cases where one just wants to set the
motor's position without homing it.  There are some workarounds to
setting a motor's position, and we use them sometimes at 17-ID, but they
are all a bit of a hack.

Lewis

-- 
J. Lewis Muir
Software Engineer
IMCA-CAT

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System