The Design and Development of a Water Tunnel To be Utilized for by UCYk0WY


									                                                                           The Design of a Water Tunnel
                                                                             To be Utilized for Flow Visualization
                                                                                        Jennifer Panik: Senior Project 2003-2004
                                                                                                 Advisor: Ann Anderson
Abstract:                                                                                                                      Desired Design Specifications:
    Flow visualization is an important technique used to understand fluid-                                                      Geometric                                        Transport
 mechanical processes. It allows one to see flow patterns over objects or models.                                                 -Must fit on a table top                         -System to be moved on table
 The goal of my project was to design a water tunnel to be used by Union College                                                  -Test section 12x12x20 inches
 for flow visualization. This poster presents the results of that study and includes                                                                                             Ergonomic
 my analytical and numerical results for flow in the water tunnel as well as                                                    Performance                                        -Models easily loaded & unloaded
 performance results for my visualization system.                                                                                 -Flow rate must allow for Re = 1
                                                                                                                                  -Flow rate must be adjustable                  Safety
    The final design includes a gravity reservoir that drives the flow in the test
                                                                                                                                  -Visibility of test section is required          -Tracers must be safe
 section and utilizes a pump to return the water from the dump tank to the
                                                                                                                                  -Tests must occur in a feasible time             -No electrical-water interactions
 reservoir. The height is approximately 6 feet, the length is approximately 6.25
 feet, and the width at the widest point is 2 feet. My visualization system consists
                                                                                                                                Cost                                             Manufacturing
 of two models with internal tubing systems that releases dye into the water flow.
                                                                                                                                  -Must be under $800                             -To be made in machine shop
 The flow around the models was at a Reynolds number of approximately 12,000.

                                                               Water Tunnel                                                                            Flow Visualization
Design Options:                                                                                                                Flow Visualization Models:
    To start my design process I brainstormed three general water tunnel designs.                                                  I created two models to be tested and because my tunnel has not been
 These designs were then analyzed to determine which one should used for the                                                    manufactured, I used the Civil Engineering Department’s tilting water flume (seen
 final design. Design 1 requires a pump to drive the flow in the test section. While                                            below) to run my tests. I used street hockey balls for my models and designed an
 Designs 2 & 3 have gravity reservoirs driving the flow in the test section.                                                    internal tubing system to allow for two different colors to be ejected out of four
 Excluded in the pictures below is the piping and pump necessary to transport the                                               1mm holes in each model. The positioning of the holes in the two models were
 water from the dump tank up to the reservoir.                                                                                  varied in order to visualize different flow.

                                                                                                                                                                        Dye Ejection

                              Design 1                                     Design 2                      Design 3                                     Model 1                                Model 2

Flow and Pressure Drop Analysis:
    I calculated the flow rate necessary to get a Reynolds number of 1 in a test
 section with a 144in2 cross sectional area and estimated the head loss in each
 system taking into account entrance and exit losses, losses for valves, and losses
 from the piping. For the first design the total loss was used to find the pressure
 drop, which can be used for pump sizing. For the second and third designs the                                                               Internal structure of models                Tilting Water Flume
 losses were used to find the pressure drop which was then used to find the
          Reynolds Number vs. Time
                                   necessary Δ z (height) required to maintain a                                               Testing Results:
                                   Reynolds number of 1 in the test section. I plotted
                                                    2 cm diameter
                                                    6 cm diameter

                                   the Reynolds number versus time for different
                                                    10 cm diameter
                                                                                                                                   The still photos shown below are from the videos of my final tests on each
  Time (hours)


                                   object sizes and determined testing at Re=1 is not                                           model. The flow around the modes has a Reynolds number of approximately
                                   feasible due to the length of time it would take to                                          12000. As can be seen from the picture of Model 1, there was separation of flow
                                   view 12 inches of flow.
                                  Reynolds Number
                                                                     100                                                        at the back of the model. From the picture of Model 2 it can be seen that there is
                                                                                                                                not only separation at the back of the model (the red stream) but also a mixing of
 Figure 1. Plot showing that at Re=1 it takes 2-10 hours for the dye to travel 12 inches of flow,
 which is not acceptable                                                                                                        the two streams.

Star-CD Analysis:                                                                                                                Direction
                                                                                                                                 of water
    I used Star-CD to optimize the geometry of my tunnel. A comparison of the                                                      flow
 velocity flow field through two inlet designs can be seen below, along with the
 final design. It is important to note the uniformity of the flow through the test
 section for the optimal design as compared with the recirculation in the initial

                                                                                                                                                        Model 1                             Model 2

                                                                                                                               Conclusions and Recommendations:
                                                                                                                                  In conclusion the design meets all specifications except allowing for flow at a
                    Abrupt inlet leads to recirculation of flow                   Optimal inlet design leads to uniform flow    Reynolds number of 1 in my tunnel. My final design was designed to run tests at
                                                                                                                                Reynolds number of 100. With my final design complete and discussions with the
                                                                                                                                machine shop already underway, the fabrication of this tunnel is not far off. Once
                                                                                                                                a pump is found that can handle the flow through the system, my design with the
                                                                                                                                given dimensions can be fabricated at the machine shop.

                                                                                                                                Prof. Ann Anderson            Prof. Brad Bruno           Prof. Tom Jewell
                                                                                                                                Stan Gorski                   Frank Coluccio             IEF - for funding

                            Star-CD analysis of final design                      Dimensioned Drawing of final design

To top