Proposal for the Maryland State Highway Administration (MD SHA) Phase II Bottomless Culvert Study

Problem Statement and Study Objective

The objective of this study is to extend previous research done at the Federal Highway Administration's (FHWA) Turner-Fairbank Highway Research Center (TFHRC) J. Sterling Jones Hydraulics Research Laboratory on the hydraulic performance of bottomless culverts. This study will focus on three hydraulic attributes for bottomless culverts:

• Fixed bed tests will be conducted to measure pre-scour conditions, which are the best for the methodology proposed in the earlier study to predict scour.
• Submerged Entrance conditions will be tested for both fixed bed and moveable bed conditions.
• Entrance loss coefficients will be determined for prescour and scoured conditions.
• Design coefficients for riprap stability at the culvert entrance will be evaluated and revised as necessary.
• Cross vanes near the culvert entrance, in the approach flow will be tested as a countermeasure for culvert scour/ channel instability. Optimize cross vane geometry and location.
• Detailed velocity distributions will be measured at the culvert entrance using advanced techniques to test assumptions made in GKY&A's and Maryland Department of Transportation (MD DOT) methodologies.

Laboratory Facilities

The FHWA J. Sterling Jones Hydraulics Research Laboratory is located in McLean, VA. The laboratory is equipped with the following:

• A tilting flume which is 70 feet long, 6 feet wide, and 1.5 feet deep with a sediment recess for scour experiments
• Two pumps: one 5 cfs; one 10 cfs (the two pumps can be combined to produce 15 cfs maximum flowrate in the flume)
• A culvert test facility that consists of an 8-foot by 8-foot by 4-foot deep headbox, and a 3-foot wide by 8-foot long, by 2-foot deep tailbox with incremental tailwater control.
• A miniature flume which is 6-feet long, 1-foot wide, and 1-foot deep
• Silicon class pressure sensors
• Velocity probes (one acoustic Doppler velocimetry probe and three electromagnetic velocity probes)
• Special digital camera for particle image velocimetry (PIV)
• Flow visualization with hydrogen bubble technique
• Laser to measure scour and flow depth
• Automation to control pumps and probes is done with a PXI/Compact PCI modular instrumentation
• Data-Image acquisition and control is done with LabVIEW software

These facilities have been applied to study culvert hydraulics for Iowa, South Dakota, West Virginia, and MD DOT as well as the Federal Emergency Management Agency.

General Approach

The experiments will be conducted in the 6-foot wide tilting flume. A rectangular culvert shape will be tested for several flows at various velocities. Experiments will include fixed bed as well as moveable bed tests under submerged as well as unsubmerged conditions.  Entrance loss coefficients will be calculated for pre-scour (fixed bed) and scoured (moveable bed) conditions based on pressure tranducer measurements.  Additionally, various riprap sizes will be retested to improve the regression results from the previous study.  We propose to use PIV to verify/modify the velocity field assumptions at the culvert entrance that were presented in the previous study.

Tasks

The following tasks will be completed under this study:

TASK 1. Experimental setup

The culvert will be installed in the test section of the 70-foot long flume, which located 39 feet from the upstream end of he flume. Silicon class pressure sensors will be mounted in the centerline on the bottom of the experimental setup (Figure 1) to measure instantaneous hydraulic grade lines. Taking time averages will lead to more precise loss coefficient computation. The discharge will be provided by 10cfs computer controlled pump. Flow depths and mean velocities can be computed from pressure sensor measurements in the culvert barrel flow is parallel to the invert. PIV and/or velocity probes will be used to augment these measurements in the highly turbulent region generated in the flow separation zone as the blocked flow mixes with the main channel flow at the upstream end of the culvert.

Figure 1 - Arrangement of the silicon class pressure sensors

TASK 2. Conduct Experiments in the Test Facility

Test a culvert shape (rectangular) for fixed and moveable bed, to determine entrance loss coefficients, velocity distributions inside the culvert and maximum scour depths and to determine sizes of rock riprap and extent of coverage that might be required to reduce scour in the most critical zones. Conduct experiments with combinations described in the test matrix below.

Test Matrix

To measure instantaneous velocity flow fields the particle image velocimetry technique (PIV) is used. PIV utilizes a focused light source, a high-resolution digital camera, and sophisticated computer logic to trace particle movements. This technology makes it possible to accurately measure velocity in complex situations such as flow into culverts.

The experimental setup of a PIV system typically consists of several subsystems (Figure 2). In most applications tracer particles have to be added to the flow. These particles have to be illuminated in a plane of the flow at least twice within a short time interval. The light scattered by the particles has to be recorded either on a single frame or on a sequence of frames. The displacement of the particle images between the light pulses has to be determined through evaluation of the PIV recordings. The local displacement vector for the images of the tracer particles of the first and second frame is determined by statistical methods (Figure 3).

Figure 2 - Experimental set up for Particle Image Velocimetry

Figure 3 - Instantaneous velocity flow field

Riprap experiments will be conducted for 3 different uniform particle sizes. The velocity will be increased incrementally until a discernible area of particles will be dislodged, which will be considered to define the failure condition for that particle size.

Conduct experiments with Rosgin type cross vanes placed in front of the culvert entrance as a countermeasure for scour and in the flume without the culvert in place.

TASK 3.  Analyze Data

Derive culvert entrance loss coefficients for the fixed bed tests, which simulate the pre-scour condition, and for the moveable bed tests, which simulate the scoured bed condition. We hypothesize that the headwater elevation will be lowered significantly for the scoured condition and that there will be a different set of design coefficients for that condition.

Compare the pre-scour velocity (V_R) assumptions made by CHANG and GKY with experimental data. Modify assumptions and derive new regression equations for the maximum depth of scour and for riprap design as necessary. We anticipate that the correlation coefficients will be significantly higher with improved velocity assumptions.

Evaluate the performance of cross vanes for scour reduction and for grade control.  Compare experimental results with numerical model results to be supplied by Xibing Dou if numerical results are available for analysis.  Compare experimental results with scour depths and local velocities that are generated by the numerical model.

TASK 4.  Final report

Submit a draft final lab report for review 60 days before end of Task Order. Revise and prepare final report, which will be published as a web document by FHWA. We will provide __ hard copies to the MD SHA

Return to top