APS Control System Naming Convention
This document is in the process of change. This line will be removed at the
end of this change process.
Background
EPICS will supply the base elements, as `channels', for the named elements
in the APS.
The restrictions of Channel Access itself will constrain the naming
of the base elements. The names will therefore show up on schematics, wire-lists, and as VxWorks symbols.
The Component names will be used to store the collected data and to
retrieve set points. Thus these names will appear extensively in analysis,
setup, and calibration software used by SDDS. Once a PV name has been developed
and become known it may be changed only with notification of the
PVname mailing list and concurrence of the users of the PVname.
Constraints
Channel Access and EPICS possess a flat namespace and a discovery protocol
that require all integrated components have channel names that:
-
are unique
-
are less than 28 characters in length
-
have no embedded non-alphameric characters other than `:', `_',
`-', `<' and `>'
Desirable Items
-
The name should make explicit the hierarchical, shared view that is most
natural.
-
A balance should be struck between intelligibility and brevity; insofar
as simple string-match tricks will have to be used for name selection,
the convention should help, not hinder this.
-
The mapping from channel names to component name should be obvious.
Requirements
Each name will have the form:
<sys><sub><xxx><xxx><xxx>
Colons may be used to separate fields
where:
-
<sys>
-
is the system, one of {L, P, S, SRF, VM, Mt, ...} see below for current
list.
These are intended to be accelerator subsystems and other independent,
logical, geographic or administrative entities. The acronyms/abbreviations
should be official.
-
These elements are as specified by the ring managers.
-
Revised Naming Conventions for the PAR and LET
-
Nomenclature for the Booster Synchrotron and BTS/BTX Lines
-
Nomenclature and Name Assignment Rules for the APS Storage Ring
Typically a component name will be precisely "<sys>:<sub>:<type>",
including the many devices which comprise it. This portion of the name
should be the most carefully chosen.
Examples:
VM:01:3IP1.VAL Sector 1 Pump Pressure
S:BM:CurrentAO Storage Ring Dipole Current
S:MPS:srBeamCurrentCC Storage Ring beam Current for MPS system
B:QD:Qdrt2MagOTBI Booster Quad D Overtemp trip
Miscellaneous
-
Names should be unique independent of case.
-
Clearly unnecessary name segments should not be used.
-
It is suggested that the PV name end in a "C" when the PV represents
a Control or Command point and thus is eligible to be restored. A PV
name should end in "M" when it is a monitored point, and in "P" when it
is a parameter. When a monitor element refers to the set point and
read-back of the same physical quantity, append a "C" or an "MC" as
appropriate. When a particular PV should never be "looked at" by and
external program an "X" should be appended to the PV name. It is
further suggested that the explicit record type (AI, MBBO, CC ...) not
be included in the PV name as this designation may be changed by the DB
designer but the PV name will not be able to be changed.
-
It is suggested that for sector numbers less than 10 a leading 0 be
used in the schema. (S01, S02, ...)
Standard System names
S Storage Ring
B Booster
SY Booster RF (Not to be used for new PVs)
BRF Booster RF
P Par
L Linac
VM Vacuum
Mt Master timing
It Injection Timing
SRF Storage Ring RF
SRFB Storage Ring Feedback System
BTS Booster to Storage Ring
H2O Water System
IOC IOC Status
ACIS Access Control Interlock
FE Front End (XFD)
ID Insertion Device (XFD)
EPS Equipment protection System (XFD)
PA PSS (XFD)
Standard names
Some Systems use templates to develop databases therefore the PV names for these
systems have a regular form. The information on these PVs will appear in this
document as it become available: