File - An Economical and Open Source Particle Image

Download Report

Transcript File - An Economical and Open Source Particle Image

A step-by-step guide to calibrating a custom PIV
system using PIVlab 1.2.
Contents
 Equipment
 Configuration
 Physical Procedure
 PIVlab 1.2 Analysis Procedure
 Results
 Sources of Error
 Conclusions
 Future Directions
Equipment
 Harvard Syringe Pump Model 33
 Two, plastic 60ml syringes with inside diameter (ID)of
of 26mm
 2m of clear rubber hose with an ID of 4.5mm
 T-valve
 Clear, glass or plastic tubing with of ID of 6.25mm and
outside diameter (OD) of 7.7mm
 Buret Clamp
 Silicone
 PIV system
Configuration
Syringe Pump Procedure
 1. Accurately mark the glass tube at 32.6mm




(calculated with the ID using πr²h) above the
opening. This is 1ml in the glass tube.
2. Silicone all tubing joints, as not to allow any air to
enter or exit the tube system.
3. After silicone is cured, fill the PIV tank with water to
desired level.
4. Insert the glass tube into the water, slightly under
the water’s surface.
5. Using a level, plumb the glass tube to make it
vertical.
Syringe Pump Procedure cont.
 6. Program the diameter of the syringes and set pump
to Continuous run
 7. Program the syringe pump to 50ml/hr., which
calculates to 0.026m/s (which is the maximum
pumping rate for these syringes)
 8. Turn on PIV system
 9. Press “Run/Stop” on syringe pump
 10. Record vortex
 11. After experimentation, film a known distance for a
calibration image
PIVlab 1.2 Analysis Procedure: load
images with sequencing style 1-2, 2-3,
3-4,…
Select region of interest (ROI) just
below the glass pipe opening
Image pre-processing, enable contrastlimited adaptive histogram equalization
(CLACHE) and highpass filtering (highpass)
 CLACHE enhances contrast in the image
 Highpass sharpens the image and removes
background signals
PIV settings: set interrogation area ([px] in both
dimensions) to 24 and Step ([px] spacing between
the centers of interrogation area) in both
dimensions to 12
Analysis: analyze all frames
Post processing: set standard deviation
(stdev) filter threshhold [n*stdev] to 8.
Apply to all frames
Calibration: load calibration image and
select reference distance along with
time step
Plot: set display parameter to velocity
magnitude, smooth data to 20% and
apply to all frames
A few velocity calculation
options…
-0- frame (represented as -0-) is the moment the water is
completely expelled from the tube into the aquarium
Velocity is calculated by programming the flow rate (i.e.
.833ml/s)*distance fluid traveled (i.e. 0.0326m) =
0.026m/s² . Record the maximum velocity/frame
 1. A running average: take 20 slides after -0- and
average the velocity calculations (smoothing data)
 2. Only look at 1 frame and average the same frame for
each run
Extractions: polyline analysis
 A polyline is a line segment that calculates each individual





vector’s velocity. This is represented in a graph with the Yaxis as velocity and the X-axis as the distance on the line.
The polyline can be drawn through the regions of highest
velocity denoted by the color bar.
Set parameter to Velocity magnitude and Type to Polyline
Draw the line across the region with the highest velocity
determined by the color-bar
Plot data
Record the velocity for that frame
Repeat this process to take a running average
Extractions: polyline analysis
Torus volume
 Volume of a torus is
2π²(D/2)r².
 Mass is calculated using the
density of H2O at room
temp.
 Smooth the data to 100%.
Torus picture courtesy of: http://blog.teachersource.com/2012/01/
Measure distance and angle:
measure radius (D/2) from core to
core
To insure accurate measurement of radius r, take
6 measurements from the same side of the torus
half from the core to the outside wall and
average them: radius 1
Radius 2…
Radius 3… repeat 3 more times
Results

Results cont.
1ml High
Velocity
(0.026m/s)
Run
1ml Low
Velocity
(0.016m/s)
1ml High
1ml Low
High Velocity High
Velocity Torus Velocity Torus Acceleration Velocity
Volume (ml) Volume (ml) (m/s²)
Force (N)
High
Velocity
Work (J)
Low
Velocity
Low
Low
Acceleratio Velocity Velocity
n (m/s²)
Force (N) Work (J)
1
0.027
0.015
1.029
0.978
5.21
0.00534 3.20E-05
3.25
0.00316 1.90E-05
2
0.026
0.021
0.986
1.015
5.22
0.00513 3.07E-05
3.24
0.00327 1.96E-05
3
0.020
0.023
0.995
1.021
5.24
0.00520 3.12E-05
3.25
0.00330 1.98E-05
4
0.027
0.019
1.028
1.023
5.22
0.00535 3.21E-05
3.26
0.00332 1.99E-05
5
0.025
0.017
1.040
0.998
5.23
0.00542 3.25E-05
3.25
0.00323 1.94E-05
0.025
0.019
1.016
1.001
SE
0.0011
0.0013
0.0105
0.0085
t
-0.434
2.549
1.519
0.861
p
<0.05
<0.05
<0.05
<0.05
Ave.
Ave. Accel.
(m/s²)
Ave.
Force (N)
Ave.
Work (J)
5.22
3.25
0.00529
0.00326
3.17E-05
1.95-05
Possible Sources of Error
 The Harvard Apparatus Twin Syringe Pump Model 33
has an accuracy within ±0.35% and reproducibility
within ±0.1%
 The clear rubber hose may be expanding and
contracting due to thermal fluctuations
 Convection currents with the experimental system
Conclusions
 All measurements using this 2D PIV system are not
statistically different from the expected values
suggested by the syringe pump apparatus
 Although the Reynolds number (Re) for the syringe
pump experiment is 162 and 100 for 0.026m/s and
0.016m/s, respectively, and the Re estimate for aquatic
leaping in Rana catesbeiana is 16100, this method can
be used to calculate force, work, and power of
amphibian propulsion in an aquatic medium
Future Directions
 To use a static glass tube from the syringes to keep the
water pressure constant
 To use larger syringes with a faster syringe pump will
allow vortices to be produced at higher Re ranges,
possibly within the frog aquatic leap velocities
 To produce a wider range of flow velocities
 Conduct experiment in a more temperature stable area
Acknowledgements
 Undergraduate Students






Duane Barbano
Erik Dillingham
Antonia Tallante
Nicholas Gengler
Russell Nelson
Maxwell Wheeler
 Graduate Students
 Krysta Powers, M.S. student
 Sang Hoon Yeo, Ph.D. student
 Kari Taylor, M.S. student
 Post-docs
 Cinnamon Pace, Ph. D.
 Jenna Monroy, Ph.D.
Faculty
Kiisa Nishikawa, Physiology, NAU
Stan Lindtsted, Physiology, NAU
Alice Gibb, Physiology, NAU
Ted Uyeno, Biomechanics, VSU
Brent Nelson, Mechanical Engineering, NAU
David Lee, Biomechanics, UNLV
Funding
NAU VPR
NAU GSG
Sigma Xi GIAR Grant G20111015158992
Personal
Mr. Pippiens