Controlling Beamline Magnets

[Mathias <steiner@nscl.msu.edu>]

Mesdames, Messieurs:

Asking about Beamline Controls in the BEAMLINE_CONTROLS list is either
natural or redundant, I'm not sure which.  But I would like to solicit
advice...

We have one more month of operations here at the Cyclotron, after which
time we'll shut down for two years to complete a major upgrade of our
accelerators and beamlines.  Among the things to go will be our old
control system, to be replaced with EPICS.

Currently, I'm using VISTA to tap into our Fermilab-style control system
and set beamline magnets.  The chain is: optics calculation for each
magnet -> field for the given rigidity -> magnet current. 

There is no feedback loop, everything gets set and stays set; dipole
magnets are matched to NMR field with a button push.  All of the 
current <---> field calculations are done by the high-level controls
program that performs the actual setting of the magnet.  Very
straightforward, but cumbersome.


The question is, how much of this should we automate?  I'm guessing that
at least the current <-> field stuff should be handled at the VME-crate
level in such a way as to be hidden from the user.  

I should mention that we do radioactive beams, so we change beamline
magnets  *a lot*...

How does everyone here do it?  One could presumably set things up such
that when the BRho for a particular beamline segment is changed, all
magnets adjust automatically.  Anybody tried this?

While I'm asking: any experience with automated beamline tuning?  


I would highly appreciate your feedback and advice.

Thanks,
		Mathias


Mathias Steiner
Radioactive Beam Physicist
Nat'l Superconducting Cyclotron Lab
Michigan State University
517-333-6423
steiner@nscl.msu.edu