Siemens Argus 7SR21, Technical Manual

Page: 416 / 644

7SR220 Technical Manual
Chapter 4 - Page 36 of 96 © 2017 Siemens Protection Devices Limited
4.2 MODBUS Register Data Types
4.2.1 FLOAT_IEEE_754
The float data type conforms to the IEEE 754 floating point definition. This specifies that 32 bits of data will be
formatted as a sign bit in the most significant bit (MSB) followed by an 8 bit exponent then a 23 bit mantissa, down
to the least significant bit (LSB).
MSB LSB
Sign Exponent Mantissa
FLOAT_IEEE_754 IN DETAIL
The exponent is an 8 bit unsigned integer. To allow for negative exponents, it is offset by 127. Therefore
the actual exponent is e - 127. The following table shows a detailed layout of the exponent.
2
7
2
6
2
5
2
4
2
3
2
2
2
1
2
0
128 64 32 16 8 4 2 1
The mantissa contains the fractional part of a number normalized to the form 1.xyz i.e. in this instance
xyz. The mantissa represents the binary fraction of a number; therefore the MSB represents 2
-1
(or
1/2
1
) and its LSB 2
-23
(or 1/2
23
). The following table shows a detailed layout of the mantissa.
1
2
1
1
2
2
1
2
3
1
2
4
1
2
21
1
2
22
1
2
23
0.5 0.25 0.125 0.0625 4.768e-7 2.384e-7 1.192e-7
As an example 1,000,000 would be represented as follows (hex 49742400).
4 9 7 4 2 4 0 0
0 1 0 0 1 0 0 1 0 1 1 1 0 1 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0
This calculates out as:
Sign = +1
Exponent = 10010010
2
= 128 + 16 + 2 = 146, subtract 127 = 19.
Mantissa = 1 +
1 1
+ +
2
1
2
2
1 1
+
2
3
2
5
1
+ +
2
10
1
2
13
= 1 +
4096 + 2048 + 1024 + 256 + 8 + 1
2
13
= 1 +
7433
2
13
= 1.907348632
Therefore Sign * 2
Exponent
* Mantissa = 1 * 2
19
* 1.907348632 = 1000000
FLOAT_IEEE_754 & MODBUS
In this MODBUS implementation the 32 bit float is stored in 2 16 registers in Big-Endian format. As an
example, if we take the hex representation of 1,000,000 as a float (from above) we have 49742400h.
Assume this is stored in the registers 30001 and 30002, it would look as follows.
Address
30001
Value
4974