Campbell Scientific TX312 High Data Rate GOES Transmitter - Manual (Page 71)

F-1
Appendix F. High Resolution 18-Bit
Binary Format
When using the binary 18 bit signed 2’s complement integer format, all data
values in the datalogger final storage area must be in high resolution format. In
most cases the datalogger program should set the data resolution to high at the
beginning of the program. Use the P78 instruction with parameter 1 set to 1.
Note: P77 Real Time can not write the time or date in high resolution. To send
a time stamp, first write the time back to input locations, then sample the input
locations as high resolution. As an alternative to using P77 for a time stamp,
the GPS time can be retrieved from the transmitter and written to final storage
in high resolution format. See instruction P127 for details.
Because the binary 18 bit integer is an integer, all information to the right of
the decimal point is dropped. This occurs while the datalogger is copying data
to the transmitter. The original data is left intact in final storage of the
datalogger. If transmitted data requires precision to the right of the decimal
place, multiply the number by the required factor of 10 before storing the data
to final storage. After data is received by the ground station, division by the
appropriate factor of 10 will replace the decimal point.
In high resolution format, data stored in final storage has a maximum
magnitude of 99999 and a minimum magnitude of 0.00001.
NESDIS has placed restrictions on the format of data sent over the GOES
satellite network. The first restriction is the use of 7 data bits and one parity bit
per byte. The second restriction is the most significant data bit of each byte, bit
6, is always set. Without data, each byte transmitted over the satellite has the
format of “p1xxxxxx”. The x’s will hold the 6 bits per byte allocated to data
information. The “p” is the parity bit and the “1” is bit 6 which is always set.
Resolution of each data point would be severely limited if the data point
consisted of only 6 bits. We use 3 consecutive bytes to form a data point word.
The first byte sent is byte 3, the most significant byte. A complete word is
created by using 3 consecutive bytes, stripping the 2 most significant bits from
each byte, then combining the 3 bytes into a word. See the examples below.
Each data point is formatted as an 18 bit integer. The format uses the
most significant bit (bit 17) to designate sign. The format of each 3
byte data point is as follows, note the top row shows the bits used and
there significance.
17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
p 1 x x x x x x p 1 x x x x x x p 1 x x x x x x
Where each “p” is the parity bit for that byte.
Where each “1” is bit 6 for that byte and always set to 1
Where the 6 “x”s represent bits 0 through 5, these make up the
information for each byte.
Where the 18 bit data point is made by combining the three bytes
after bit 7 and bit 6 of each byte have been dropped.
Where 0 represents bit 0 - the least significant bit