Campbell Scientific TD Operating System Addendum for CR510, CR10X, and CR23X - Manual (Page 25)

THIS SECTION ENTIRELY REPLACES THE DATALOGGER MANUAL SECTION 2.
AD-2-1
SECTION 2. INTERNAL DATA STORAGE
2.1 FINAL STORAGE AND DATA
TABLES
Final Storage is that portion of memory where
final processed data are stored. It is from Final
Storage that data is transferred to your
computer. With the TD datalogger, Final
Storage is organized into Data Storage Tables.
These data tables should not be confused with
the program tables *1, *2, and *3 that contain
the datalogger program.
Within each data table, data is organized in
records and fields. Each row in a table
represents a record and each column
represents a field. To understand the concept
of tables it may be helpful to consider an
example. A CR10X is to be used to monitor 3
thermocouples (TC). Each hour a temperature
for each of the three TC is to be stored. The
table has 4 fields: "DATE_TIME TEMP1 TEMP2
TEMP3." Each hour a new "record" would be
added. The "hourly" table would then be
organized as follows:
DATE_TIME TEMP1 TEMP2 TEMP3
01/27/91 10:00:00 23.5 24.6 28.2
01/27/91 11:00:00 24.2 22.4 23.4
Only the hourly data is stored in the hourly table,
Each output interval has its own table. Data
tables can also be "event driven" rather than
interval driven, that is a new record is stored
when a specified event occurs rather than
based on time. Each table is completely
independent of any other tables and all records
in a given table have the same number of fields.
Each table is allocated (manually by the user or
automatically by the datalogger) a number of
records. Different tables have different
numbers of records. Each data table is in a ring
memory configuration such that when the
allocated number of records has been stored,
each subsequent new record will overwrite the
oldest stored record. The *9 Mode may be
used to view the size of the Data Storage
Tables. (Section 1.9)
The TD datalogger supports naming of tables and
fields, so any data value can be referenced by the
table and field names. For example, the
temperature data for the first thermocouple is
referenced as "HOURLY.TEMP1." As Data Tables
are allocated in the datalogger program, some
Final Storage Memory is reallocated for the
storage of these labels and other data table
overhead.
NOTE: All Data Storage Tables are
reallocated and erased whenever the
datalogger program is recompiled (*0, *6,
*B), when Input Storage Memory is
reallocated (*A), or when a new datalogger
program is transferred from the computer to
the datalogger. ALWAYS RETRIEVE
UNCOLLECTED DATA BEFORE MAKING
ANY CHANGES.
A time stamp and record number are
automatically included with the each record in
each table. These are used as part of the data
collection protocol.
2.1.1 TIME AND TIMESTAMPS
Each record in a table has a time stamp
associated with it. With Instruction 84 set for
interval output (a interval in seconds is specified
as the second parameter), time is not actually
stored with each record. Using the timestamp
of the last record stored and the table interval,
the datalogger can calculate the timestamp for
any previous record. When retrieved, each
record in the data file will have a timestamp.
This saves 6 bytes per record by not storing
time with each record. A consequence of not
storing time is that if output does not occur at a
scheduled time, the datalogger must keep track
of the discontinuity in the timestamps so it can
correctly calculate timestamps for records older
than the missing record. The datalogger will
keep track of the 10 most recent discontinuities
in each table. If more than ten discontinuities
occur, records with timestamps older than the
oldest discontinuity cannot be reliably
timestamped when collected. For this reason
interval tables should not be used if outputs will
be routinely missed. Outputs can be missed
(discontinuities can occur) when:
• The datalogger clock is changed such that
is passes an output interval.