Campbell Scientific CR1000 Measurement and Control System - Manual (Page 130)

Section 7. Installation
130
lapse occurs, the SkippedRecords status entry is incremented, and a 16-byte sub-
header with time stamp and record number is inserted into the data frame before
the next record is written. Consequently, programs that lapse frequently waste
significant memory.
If Lapses is set to an argument of 20 , the memory allocated for the data table is
increased by enough memory to accommodate 20 sub-headers (320 bytes). If
more than 20 lapses occur, the actual number of records that are written to the
data table before the oldest is overwritten (ring memory) may be less than what
was specified in the DataTable() or the CF CardOut() instruction or TableFile()
instruction with
Option 64 ..
If a program is planned to experience multiple lapses, and if telecommunications
bandwidth is not a consideration, the
Lapses parameter should be set to 0 to
ensure the CR1000 allocates adequate memory for each data table.
Table 14. DataInterval() Lapse Parameter Options
DataInterval() Lapse
Argument Effect
X > 0
If table record number is fixed, X data frames (1 kB per data
frame) are added to data table if memory is available. If record
number is auto-allocated, no memory is added to table.
X = 0
Time stamp and record number are always stored with each
record.
X < 0
When lapse occurs, no new data frame is created. Record time
stamps calculated at data extraction may be in error.
Scan Time and System Time
In some applications, system time (see System Time
(p. 468)
) is desired, rather than
scan time (see Scan Time
(p. 464)
). To get the system time, the CallTable()
instruction must be run outside the Scan() loop. See section Time Stamps
(p. 273).
OpenInterval() Instruction
By default, the CR1000 uses closed intervals. Data output to a data table based on
DataInterval() includes measurements from only the current interval.
Intermediate memory that contains measurements is cleared the next time the data
table is called regardless of whether a record was written to the data table.
Typically, time series data (averages, totals, maxima, etc.) that are output to a data
table based on an interval only include measurements from the current interval.
After each output interval, the memory that contains the measurements for the
time series data are cleared. If an output interval is missed (because all criteria are
not met for output to occur), the memory is cleared the next time the data table is
called. If the OpenInterval instruction is contained in the DataTable()
declaration, the memory is not cleared. This results in all measurements being
included in the time series data since the last time data were stored (even though
the data may span multiple output intervals).
Note Array-based dataloggers, such as CR10X and CR23X, use open intervals
exclusively.