Accuracy of the Floating-point Numbers
The 32-bit floating-point numbers have a 23-bit mantissa. Integer numbers can be represented in the following
ranges without loss of accuracy:
•
Binary:
•
Hexadecimal: ±FF FF FF
•
Decimal:
32-bit floating-point numbers are accurate to about 7 decimal digits. An accuracy of 4 decimal digits (0.2 mea-
suring error) is required for measuring alternating current quantities.
2.2.6.2 Data Type - Date/Time (Read and Write)
The Date/Time data type is used to transmit the local time. The following format is used:
Date/Time Status
10H set:
Daylight saving time active
20H set:
Date/time error: equivalent to FAIL bit, see Device Manual SICAM Q100, Time Synchronization.
NOTE
For the time synchronization via Ethernet, Siemens recommends the use of NTP.
SICAM Q100, 7KG95xx, System Manual
E50417-C1040-C522-A8, Edition 09.2020
±(1)111 1111 1111 1111 1111 1111
±16777216
Milliseconds (0 to 59 999)
Byte 1 (MSB)
...
Hours (0 to 23)
Byte 3
Month (1 = Jan. to 12 = Dec.)
Byte 5
Date/time status
Byte 7
Byte 0 (LSB)
Holding register 0065
Minutes (0 to 59)
Holding register 0066
Day (1 to 31)
Holding register 0067
Year (0 = 1900)
Holding register 0068
Byte 2
Byte 4
Byte 6
...
Communication
2.2 Modbus
25