Startup of the experimental physics industrial control system at NSTX Page: 3 of 3
This report is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
IV. Development Issues
The EPICS software was successfully used in achieving
the NSTX project's initial milestone, first plasma. A
retrospective of our EPICS development experiences
show that a few precautionary measures during the early
development phase could have increased the control
system's safety margin for operational readiness.
A. Late Start
It would have been beneficial to begin using EPICS
sooner. The NSTX construction schedule had the CI&C
effort as one of the last to begin in earnest (i.e. labor
funding ramp-up). The very first exposure to EPICS was
at a (worthwhile) 4-day EPICS Training course at Los
Alamos. This was less than one year prior to first
plasma. After the course was completed, it was still
several months before an on-site, working version of
EPICS was available for beginning the hands-on learning
B. No Prototype
Not building a small, low-performance prototype system
early on was another problem. A prototype could
probably have been constructed using "spare" parts. To
constrain costs, working on the prototype could have been
approached as a low level of effort task, spanning months.
This should have been done well ahead of the time when
the project's controls wheel needed to be in high gear.
The importance of this cannot be overstressed. The most
beneficial aspect of a prototype would be to provide a
realization of the skills needed to get EPICS built.
C. Get the Right People
It took much longer than anticipated to create the software
environment for building EPICS. Access to an
experienced unix system administrator was essential for
building and installing EPICS.
To increase productivity a graphical configuration tool
was used to produce the record databases that are loaded
into the IOC. The editor of choice was called GDCT,
which was open source and part of the EPICS distribution.
There was an alternative to GDCT, it was based on a
commercial product called Capfast. That was not selected
because of cost. GDCT is no longer being maintained, and
whatever bugs are in it will remain there. However, once
recognized, the bugs can usually be avoided. NSTX did not
have the resources to fix the product for the collaboration.
To improve software maintenance, it would be beneficial
to begin using a better database tool.
It has been over a year since EPICS has been running at
NSTX. The system has been very reliable and new
applications can now be quickly added. One remaining issue
is in the combining of records to implement a new control
algorithm. It is not always intuitive which record fields
should be to linked together to get the desired result.
However, with a little trial and error the algorithm is soon
working as intended.
Control system operations are now routine. Enhancements,
reliability, and maintainability issues have become the focus
of the control group. There are several additional EPICS
tools that can be brought online to enhance the usefulness of
the system. Keeping our system current with new EPICS
releases (about every 12 months) and Solaris releases are
important maintenance tasks.
 EPICS documents at Argonne National Laboratory,
Argonne, IL, available over the internet.
 EPICS documents at Los Alamos National Laboratory, Los
Alamos, NM, available over the internet.
 P. Sichta, et al , "The NSTX Central Instrumentation &
Control System ," 18h Symposium on Fusion Engineering,
Albuquerque, NM (1999).
This report can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Report.
Sichta, P. & Dong, J. Startup of the experimental physics industrial control system at NSTX, report, December 17, 1999; Princeton, New Jersey. (https://digital.library.unt.edu/ark:/67531/metadc702314/m1/3/: accessed March 25, 2019), University of North Texas Libraries, Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.