Class=70 (0x46) Single precision floating point measurement



Floating point single precision measurements. This class mirrors the standard measurement events is CLASS1.MEASUREMENT=10. The measurement unit is always the standard unit.

The value is a "float" - IEEE-754, 32 Bits, single precision.

s eeeeeeeeeee mmmmmmmmmmmmmmmmmmmmmmm
s = sign bit ( 1-bit )
e = exponent ( 8-bits )
m = mantissa ( 23-bits )

That is a total of 32-bits. The most significant byte is stored in byte 0.

Byte Description
0 s e7 e6 e5 e4 e3 e2 e1
1 e0 m22 m21 m20 m19 m18 m17 m16
2 m15 m14 m13 m12 m11 m19 m9 m8
3 m7 m6 m5 m4 m3 m2 m1 m0

Very Simple Control Protocol
This document is licensed under Creative Commons BY 4.0 and can be freely copied, redistributed, remixed, transformed, built upon as long as you give credits to the author.

results matching ""

    No results matching ""