Campbell CR23X, Руководство оператора

Страница: 209 / 374

SECTION 13. CR23X MEASUREMENTS
13-9
NOTE: Excitation transients are eliminated
if excitation leads are contained in a shield
independent from the signal leads.
The size of the peak transient is linearly related
to the excitation voltage and increases as the
bridge resistor, R
f
, increases. Table 13.3-4
shows measured levels of V
eo
for 1000 foot
lengths of three Belden wires used in Campbell
Scientific sensors. Values are given for R
f
equal to 1 kohm and 10 kohm. Table 13.3-4 is
meant only to provide estimates of the size of
excitation transients encountered; the exact
level will depend upon the specific sensor
configuration.
Equation 13.3-7 can be solved for the maximum
lead length, L, permitted to maintain a specified
error limit. Combining Equations 13.3-7 and
13.3-4 and solving for L gives:
L = -(R
o
C
f
+ (t/ln(V
e
/V
eo
)))/R
o
C
w
[13.3-15]
where V
e
is the measurement error limit.
EXAMPLE LEAD LENGTH CALCULATION
FOR 107 TEMPERATURE SENSOR
Assume a limit of 0.05
°
C over a 0
°
C to +40
°
C
range is established for the transient settling
error. This limit is a reasonable choice since it
approximates the linearization error over that
range. The output signal from the thermistor
bridge varies nonlinearly with temperature
ranging from about 100 µV/
°
C at 0
°
C to 50
µV/
°
C at 40
°
C. Taking the most conservative
figure yields an error limit of V
e
= 2.5 µV. The
other values needed to calculate the maximum
lead length are summarized in Table 13.3-5 and
listed below:
1) V
eo
~
50 mV, peak transient at 2 V excitation
2) V
e
~ 2.5 µV, allowable measurement error
3) t = 450 µs, CR23X input settling time
4) R
o
= 1 kohm, 107 probe source resistance
5) C
f
= 3.3 nfd, CR23X input capacitance
6) C
w
~
42 pfd/ft., lead wire capacitance
Solving Equation 13.3-15 gives a maximum
lead length of:
L
~
1003 ft., error
~
0.05
°
C
Setting the allowable error at 0.1
°
C or
approximately 5 µV, the maximum lead length
increases to:
L
~
1085 ft., error
~
0.1
°
C
13.3.4 SUMMARY OF SETTLING ERRORS FOR
CAMPBELL SCIENTIFIC RESISTIVE
SENSORS
Table 13.3-5 summarizes the data required to
estimate the effect of lead length on settling
errors for Campbell Scientific's resistive
sensors. Comparing the transient level, V
eo
, to
the input range, one suspects that transient
errors are the most likely limitation for the 107
sensor. The sensors in the WVU-7 are the
same as in the Model 107 (the lead wire is
different), but the signal leads for the WVU-7
wet- and dry-bulbs are not subject to excitation
transients because they are shielded
independently from the excitation.
TABLE 13.3-5. Summary of Input Settling Data For Campbell Scientific Resistive Sensors
Sensor Belden Ro Cw
ττττ
* Input
Model # Wire # (kohms) (pfd/ft.) (us) Range(mV) V
x
(mV) V
eo
(mV)**
107 9661 1 42 45 10 2500 50
227 9661 0.1-1 42 5-45 1000 500 0
237 9661 1 42 45 50 5000 65
034A 9721 1-6 41 1-222 5000 5000 0-90
* Estimated time constants are for 1000 foot lead lengths and include 3.3nfd CR23X input
capacitance.
** Measured peak transients for 1000 foot lead lengths at corresponding excitation, V
x
.