Even though the 60 load cells could be sampled at a frequency of up to 20 Hz, a
sampling frequency of 4 Hz has proven to be sufficient for irregular waves with a
peak period of 2.5 sec. This provides 10 data points per wave cycle at the peak
frequency.
Two smaller data sets also are collected during a typical experiment. The
first is collected prior to starting wave generators and vertical turbine pumps at
the beginning of the test segment. For this data set, the 60 load cells are sampled
at a frequency of 1 Hz for a duration of 100 sec. Another data set is collected in
the same manner immediately after the test segment, when the waves and
currents have been stopped and the water in the basin has become relatively
calm. These two data sets are used to quantify the initial and final submerged
weight of sand in each trap.
There are three issues that complicate the task of collecting high quality LST
data. The first is the accumulation of air pockets beneath the traps and the
neoprene rubber seals. This occurs as the facility is being filled with water, prior
to an experiment. However, a simple method was developed to remove these air
bubbles from beneath the traps and the neoprene rubber seals, as discussed
previously in Chapter 4.
The second issue is related to the observation that some of the sand that is
transported into each flow channel accumulates on top of the neoprene rubber
seals around the perimeter of the traps. In other words, a portion of the
submerged weight of this sand is supported by the walls of the flow channels
rather than by the load cells which support the sediment traps. This problem is
greatest at the upstream end of the traps, because a significant amount of sand
accumulates in that area. To solve this problem, an additional task has been
added to the standard operating procedure used to conduct each experiment. The
results from a previous LST experiment will be used to illustrate this procedure.
In Figure 52, the "PreTest Weight" data are not equal to zero because the traps
had not been dredged prior to this test segment. After the test segment was
completed, the submerged weight of sand in each trap was measured for 100 sec
at a frequency of 1Hz, as indicated by the "PostTest Weight" data set in the
figure. A water hose is used to wash the sand off the rubber seals and into the
traps. This task is carried out in two steps. First, the sand that has accumulated
on the rubber seals along both sides of the traps is washed into the traps. Then
the submerged weight of sand in each trap is measured for the third time, for
100 sec at 1 Hz and is referred to as the "Side Rubber Cleaned" data set. Next,
the sand that has accumulated on the rubber seal at the updrift end of the traps is
washed into the traps. Finally, the submerged weight of sand in each trap is
measured for the fourth time, for 100 sec at 1 Hz and is indicated by the "All
Rubber Cleaned" data set. This data set represents the true submerged weight in
each trap. For this particular test segment, an average of 3.3 and 7.7 percent of
the true submerged weight of sand measured in each trap had initially
accumulated on the side rubber seals and the updrift rubber seal, respectively.
Therefore, a total of 10 percent of the sand had initially accumulated on the
rubber seals, which is an average value for the 20 traps. Although this process
typically takes 2 hr to complete, it is a crucial step in the process to obtain high
quality sediment transport data sets.
95
Chapter 8
Sensor Calibration, Measurement, and Data Analysis
Previous Page
