Deadband used with Data Logging to the SCADA Node Hard Drive. These appear in the Data Log Trend displays. A Data Log Deadband is applied before writing to file (for one-second data. If the change in the Tag's value from last record exceeds this deadband, then a new record is written (record-by-exception).
It is highly recommended to use a Data Log deadband to reduce Disk Access. This will reduce the size of the Data Log Trend files and time used to write to the Hard Drive. The Hard Drive is usually the bottle neck in Data Logging. An IDE Hard Drive at 7200 rpm can handle about 800 Tags per second of Data Logging.
Log Data must be YES for logging to occur. This deadband applies to one-second data. Minute, Hourly and Daily records are not affected by this deadband.
DATA LOG Deadband is in percent of SPAN (% of SPANHI - SPANLO).
SIGNIFICANT CHANGE: Beginning with Version 6.0 candidate, the DataLog Deadband is now applied to one-minute trend data. Previously, the dead band was ignored and a value recorded at the end of each minute. Now, the deadband (multiplied by the difference in high and low span) is used to determine if the new value should be recorded. The result is that many trends that oscillated within the deadband will now appear flat. However, less data will be recorded, and the SCADA node will start faster and show faster response for users with large amounts of trend data.
See also:
4.2.15 Log Data
3.2.23.1 Disable Data Log Seconds recording
6.2.3 Add Data Log Trend Group