Nitto TGF450 Series, Руководство пользователя

Страница: 31 / 47

User’s Manual
Rev1.0
Date:15/07/2015 User’s Manual
31 / 47
Note: Function code 16 is supported to write float date. But for float date, the first register address
and the quantity of the registers must be even, or writing is not allowed.
6.3 Calculation of CRC parity code.
Request
Response
01 : Address N1 CRC=0FFFFH is initial value
10 : Function code N2 XOR operation the CRCL and N1
00 : Register address higher N3 CRC move 1 bit right , if move out is 1 bit
01 : Register address lower N4 CRC=CRC XOR A001H
00 : Register quantity higher N5 if move out is 0 , CRC=CRC
04 : Register quantity lower N6 Move right for 8 times to finish the N1 calculation
04 : Date quantity N7 …
80 : Date 1 N8 XOR operation the CRCL and N11
04 : Date 2 N9 CRC move 1 bit right , if move out is 1 bit
80 : Date 3 N10 CRC=CRC XOR A001H
80 : Date 4 N11 if move out is 0 , CRC=CRC
CRCL : CRC Parity code lower Move right for 8 times to finish the N11 calculation
CRCH
：
CRC Parity code higher Get the CRC calibration value
6.4 The float date format of the instrument
The storage sequence of 4 bits float formate is as below:
Address: 0 1 2 3
Content:
MMMMMMMM MMMMMMMM EMMMMMMM SEEEEEEEE
Use IEEE standard method, do not store 1 on top digit, if top digit is 1, means negative; if top digit is 0,
means positive. So the 23 mantissas and a 1 on top digit, which is concealed, constitute a 24 bits fixed
point true form decimal, which is a decimal have mantissas less than 1 and more than or equal to 0.5.
The lowest 8 bits are exponent-marker using shift code method. The exponent marker equals to the
actual value minus 127. For example: 7=86H-7FH, -10=75H-7FH
e.g.: 100=0x00,0x00,0x42,0xc8
-100=0x00,0x00, 0xc2,0xc8
0=0x00.0x00.0x00.0x00 (exponent-marker is 0, the number is 0)