|
1. Modelling Facilities
2. Laboratory Data Collection
3. Test Programme
A large flume at HR Wallingford (UK) was modified
to undertake the first of three series of laboratory test.
The flume is approximately 50 * 10 m in dimension and is equipped
with a number of pumps that allow over 1 m3/s to be pumped
through the facility. A large volume of storage upstream of
the test section allows for continued flow as breach formation
through an embankment occurs. An automated control system
was developed to control operation of the pumps so as to maintain
a steady upstream water level for as long period as possible.
A large pit at the lower end of the facility acts as a sediment
trap before water is released into a re-circulating sump system.
Figure 1: Flood channel facility
Back to WP2
2. Laboratory Data Collection Laboratory Data Collection
For Lab Tests 1-17, the following data has been collected:
1. Inflow into the flume (P6).
2. Water levels upstream and downstream of the built embankment
(L1, L2, and L3).
3. Approach velocity (Upstream of the built embankment).
4. Pore water pressure in the built embankment (P1-P4).
5. Photos and videos for monitoring breach development.
Limited data was collected for tests 18 and 19, as these tests were primarily undertaken to provide our partners in Norway with information about piping behaviour in moraine
embankments. Data from tests 20 to 22 is currently being analysed by UK University and will not be included in
the numerical modelling programme. However, data from these tests will be available for further analysis in
the future by interested modellers.
Back to WP2
3. Test programme
Nine laboratory tests out of the project programme
total of 22 tests were undertaken in the U.K. at HR Wallingford
during August-September 2002. This series of tests was based
around Field Test #2 – the minimum cohesive, homogeneous
embankment. Each embankment was built in the modelling flume
(See Figure) from non-cohesive material, however, more than
one grading of sediment was used along with different embankment
geometry, breach location and time before failure (seepage
effect). . This approach allows investigation of the extent
to which varying these parameters affect the breach growth.
The different sediment gradings used were:
1. Uniform grading with D50 = 0.70-0.90 mm
2. Uniform grading with D50 = 0.25 mm
3. Wide grading (4 types of sand were used) with D50 = 0.25
mm
[Note: Uniform grading = as steep a grading curve as possible,
dependent upon sediment suppliers. Wide grading = combination
of different materials to match the grading distribution observed
in the field. All of these gradings were based upon a 1:10
direct scaling of field test sediments).
Table 1 and Figure 2 below provide details of each test:
Table 1: Details of Laboratory Tests 1-9
|
Lab. Test
Description
|
Lab. Test
Objective
|
Grading
|
| Lab
Test #1 |
Trial
/ test experiment |
Facility
set-up / trial |
1
|
| Lab
Test #2 |
Uniform
grading with the same D50
as Field Test #2 |
Effect
of sediment uniformity |
2
|
| Lab
Test #3 |
Same
as Lab Test #2 |
Assess
repeatability of test |
2
|
| Lab
Test #4 |
Same
as Lab Test #2, breach initiation notch against abutment |
Assess
effect of breach location |
2
|
| Lab
Test #5 |
Replicate
Field Test #2 |
Direct
replication to field event |
3
|
| Lab
Test #6 |
Vary
geometry, different face slope (1:2 instead of 1:1.7) |
Assess
effect of face slope |
3
|
| Lab
Test #7 |
Vary
geometry, different crest width (0.30 m instead of 0.2
m) |
Assess
effect of crest width |
3
|
| Lab
Test #8 |
Same
geometry as Lab Test #2, Uniform grading with higher
D50 |
Assess
effect of sediment size |
1
|
| Lab
Test #9 |
Same
geometry as Lab Test #2, allowing seepage before failure |
Assess
effet of seepage |
3
|
 
Figure
2: Grading curves for laboratory tests 1-9
A further 13 tests were undertaken in 2003:
Tests 10 – 17 were based around Field Test #1 at a scale
of 1:10 to the field test. Test 10-16 were built from clay;
Test 17 was built from moraine (see graph above for grading
curves). Table 2 and Figure 3 provide details of each test:
Table 2: Details of Laboratory Tests 10-17
| Lab.
Test Description |
Lab.
Test Objective |
| Lab
Test #10 |
Replicate
Field Test #1 |
Direct
replication to field event |
| Lab
Test #11 |
Same
as Lab Test #10 |
Assess
repeatability of test |
| Lab
Test #12 |
Same
as Lab Test #10 but compacted with half of
the compaction effort used for that test |
Assess
effect of compaction |
| Lab
Test #13 |
Same
as Lab Test #10 but at optimum moisture content (Partially
failed) |
Assess
effect of moisture content |
| Lab
Test #14 |
Remains
of Lab Test #13 but left overnight |
Assess
effect of seepage |
| Lab
Test #15 |
Vary
geometry, different downstream face slope (1:1 instead
of 1:2) |
Assess
effect of downstream face slope |
| Lab
Test #16 |
Vary
geometry, different downstream face slope (1:3 instead
of 1:2) |
Assess
effect of downstream face slope |
| Lab
Test #17 |
Same
as Lab Test #10 but built from moraine
instead of clay.
|
Assess
effect of construction material |
ame geometry as # 10 with different material
Figure
3: Grading curves for laboratory tests 10-17
Tests #18 and 19 were undertaken in support
of Field test # 5, whilst Tests 20-22 monitored the initiation
and development of piping failures in samples of real embankment
material. Table 3 provides details of each test: Figure 4
shows the grading curve used for tests 18-22.
Table 3: Details of Laboratory Tests 18-22
| Lab.
Test Description |
Lab.
Test Objective |
| Lab
Test #18 |
Replicate
initiation of piping for Field Test #5 |
Provide
information about
the pipe formation to assist in
development of the field test
failure mechanism |
| Lab
Test #19 |
Same
as Lab Test #18 |
Assess
repeatability of the
test |
| Lab
Test #20 |
Material brought from an UK flood embankment. Samples
were 1m (W) x 1m (L) x 0.8m (D)
|
Monitor
piping initiation and
development |
| Lab
Test #21 |
| Lab
Test #22 |
Figure 4: Grading curves for laboratory tests 18-22
|
back to WP2