...
This semester I have worked on a new Invent team, the Turbidimeter team. I worked on the design of a simple, inexpensive, reliable way of measuring the turbidity of a water sample. The design involves filling a meter-length tube of 2 in. diameter with a volume of water that is to be tested for approximate turbidity. An LED light attached to a lowering component of slightly longer length than the tube itself is then lowered into the volume of turbid water. Using a power law relationship that correlates the turbidity of the water sample to the depth at which a pattern on the LED light becomes indistinguishable, an approximate turbidity measurement can be made.
I worked on the experimental design used to obtain the power law relationship associated with the use of different LED visual patterns as well as different LED light bulb battery powers. Experiments have involved keeping a constant volume of water in the turbidimeter tube and incrementally adding a mass of kaolin clay to the water to obtain steady increases in turbidity. LED depth and turbidity measurements are made after each addition of clay and a plot of LED Depth (cm) vs. Turbidity (NTU) is constructed for each data set. Regression analysis is then used to fit a power - law curve to the data points. The R^2 value of each data fit helps us decide which LED visual pattern and battery power will allow the most reliable and accurate power-law relationship between LED depth and water turbidity. This will lead us to standardizing the components of our turbidimeter to include a turbidity scale on the LED lowering component itself, allowing any operator to easily read off the turbidity of a water sample at the water surface level of the turbidimeter tube.