Campbell Scientific CR23X Micrologger - Manual (Page 98)

SECTION 7. MEASUREMENT PROGRAMMING EXAMPLES
7-10
cell, the excitation voltage actually applied to
the load cell, V
1
would be:
V1 = V
x
Rs/RT = V
x
350/(350+33) = 0.91 V
x
Where V
x
is the excitation voltage. This means
that the voltage output by the load cell would
only be 91% of that expected. If recording of
the lysimeter data was initiated with the load cell
output at 0 volts, and 100mm of
evapotranspiration had occurred, calculation of
the change with Instruction 6 would indicate that
only 91mm of water had been lost. Because the
error is a fixed percentage of the output, the
actual magnitude of the error increases with the
force applied to the load cell. If the resistance of
the wire was constant, one could correct for the
voltage drop with a fixed multiplier. However,
the resistance of copper changes 0.4% per
degree C change in temperature. Assume that
the cable between the load cell and the CR23X
lays on the soil surface and undergoes a 25
o C
diurnal temperature fluctuation. If the resistance
is 33 ohms at the maximum temperature, then at
the minimum temperature, the resistance is:
(1-25x0.004)33 ohms = 29.7 ohms
The actual excitation voltage at the load cell is:
V
1
= 350/(350+29.7) V
x
= .92 V
x
The excitation voltage has increased by 1%,
relative to the voltage applied at the CR23X. In
this case, where we were recording a 91mm
change in water content, there would be a 1mm
diurnal change in the recorded water content that
would actually be due to the change in
temperature. Instruction 9 solves this problem
by actually measuring the voltage drop across
the load cell bridge. The drawbacks to using
Instruction 9 are that it requires an extra
differential channel and the added expense of a
6 wire cable. In this case, the benefits are worth
the expense.
The load cell has a nominal full scale output of 3
millivolts per volt excitation. If the excitation is
3.3 volts, the full scale output is 9.9 millivolts;
thus, the
± 10 millivolt range is selected. The
calibrated output of the load cell is 3.106mV/V1
at a load of 250 pounds. Output is desired in
millimeters of water with respect to a fixed point.
The calibration in mV/V1/mm is:
3.106mV/V
1
/250lb x 2.2lb/kg x
3.1416kg/mm/4 =
0.02147mV/V
1
/mm
The reciprocal of this gives the multiplier to
convert mV/V1 into millimeters. (The result of
Instruction 9 is the ratio of the output voltage to
the actual excitation voltage multiplied by 1000,
which is mV/V1):
1/0.02147mV/V
1
/mm = 46.583mm/mV/V
1
The output from the load cell is connected so
that the voltage increases as the mass of the
lysimeter increases. (If the actual mechanical
linkage was as diagrammed in Figure 7.12-1,
the output voltage would be positive when the
load cell was under tension.)
When the experiment is started, the water
content of the soil in the lysimeter is
approximately 25% on a volume basis. It is
decided to use this as the reference, (i.e., 0.25 x
1500mm = 375 mm). The experiment is started
at the beginning of what is expected to be a
period during which evapotranspiration exceeds
precipitation. Instruction 9 is programmed with
the correct multiplier and no offset. After
hooking everything up, the counterbalance is
adjusted so that the load cell is near the top of
its range; this will allow a longer period before
readjustment is necessary. The result of
Instruction 9 (monitored with the
6 Mode)
is 109. The offset needed to give the desired
initial value of 375mm is 266. However, it is
decided to add this offset in a separate
instruction so that the result of Instruction 9 can
be used as a ready reminder of the strain on the
load cell (range =
± 140mm). When the strain on
the load cell nears its rated limits, the
counterbalance is readjusted and the offset
recalculated to provide a continuous record of
the water budget.
The program table has an execution interval of
10 seconds. The average value in millimeters
is output to Final Storage (not shown in Table)
every hour. The average is used, instead of a
sample, in order to cancel out the effects of
wind loading on the lysimeter.