Campbell CSAT3, Instruction Manual

Page: 69 / 72

Appendix D. SDM Communications and Long Signal Cables
D-3
Take the following case, a 575 foot SDM signal and power cable is used to
connect a CR23X datalogger and a single CSAT3 anemometer.
Communication between the CR23X and CSAT3 fails at the default SDM
clock rate. Thus, a slower SDM clock rate must be used. The example CR3X
SDM clock rate search program generated the data in TABLE D-2.
TABLE D-2. Data from Example CR23X
SDM Clock Rate Search Program
150,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16, ... ,253,254,255
151,100,99999,99999,100,99999,99999,100,99999,100,96,0,0,0,99999,99999,
99999, ... ,99999,99999,99999
152,1,0,0,1,0,0,1,0,17,450,449,449,432,0,0,0, ... ,0,0,0
Run the SPLIT parameter file SDM_TRAN.PAR on the above data set to
transpose the data into columns.
Now run the SPLIT parameter file SDM.PAR, to filter out the SDM clock
parameters that were not tested, 99999 in array 151. This parameter file also
creates a report of the SDM clock parameters that were tested and their failure
rate as a percent of attempts.
The following report was created from the above SPLIT parameter file,
SDM.PAR.
TABLE D-3. SDM Clock Rate Report
SDM
Param
Failures %
Attmpt Attempts
1 100 1
4 100 1
7 100 1
9 100 17
10 96 450
11 0 449
12 0 449
13 0 432
From the above results, the fastest rate that the SDM bus could run, without
encountering communication errors, would be using a value of 11 for the SDM
clock parameter in Instruction Set SDM Clock (P115). However, to have a
safety margin, a value of 12 is used.
The SDM parameter was set to 12 and tested the eddy covariance program.
The test showed that using SDM parameter of 12, in the Set SDM Clock
(P115), did not cause the CR23X to overrun the Tables (
∗
B mode), thus the
SDM Clock parameter is left at 12.