Joseph Theiss' Portfolio

Changing the angle of inlets in a t-junction was found to have a weak interaction with mixing.
Two buckets, one with red dye, were filled with water and hung above all other equipment such
that there was no pressure difference. A tube was drawn from each bucket to a ball joint which
controlled flow to a self-made clear poly-vinyl t-junction. The t-junctions tested were constructed
with various angles. The flow was imaged with a standard camera and was recomposed with only a
red channel. The tubing was isolated from the background, and a histogram of the color intensity
was prepared. After removing the outlier, six of the seven histograms had their standard deviation
measured. They were graphed on a plot of standard deviation vs. angle and found to have a weak
linear regression with a slope of (15 ± 80%) which shows the weakness of this analytical method.
This trend shows that as angle increase so should mixing and vice versa.

Measuring Fluid Flow in a T-Junction Pipe with Colored Dye

PCS was used on small polystyrene spheres of radius (53.5 ± 2)nm in a water solution, to measure
the sphere’s radius experimentally. A He-Ne laser was used and shone through a vial containing the
spheres. A PMT was at an angle of 90 to catch light scattered off of the spheres. Due to the small
size of the spheres when compared to the wavelength generated by the He-Ne laser (632.58 nm),
Rayleigh Scattering was used. A computer program recorded the intensity values from the PMT
and saved them for later review. Autocorrelation was used on this stochastic data set to relate it to
the underlying mechanism. After fitting an exponential to multiple trails of varying concentration,
an average error of 1.8% was found with a minimum error of 0.5% and a maximum error of 3.1%.
No measured values were outside one standard deviation from the known value of the spheres.

Measuring the hydrodynamic radius of polystyrene spheres in water

It was found that Stokes’ model for drag fit only on smooth objects or non-smooth objects at
low speeds. The standard deviation from the predicted values was an average of 1.10e-05 in the
linear looking ranges. Newton’s model was also fit to linear ranges and had deviations of an average
4.49e-06. Unlike Stokes’ model, Newton’s predicted the movement of fast non-smooth objects.
Objects had transition periods where they would go from Newton’s model to Stokes’ as the objects
would slow down. This transition period was largely ignored and should be investigated in further
experiments.

Laminar vs. Turbulent flow around a rotating sphere