Varying Influent Turbidity
In order to determine the influent ranges with which a polyurethane foam filter will produce acceptable effluent conditions, a variety of influent turbidities (5, 10, 20 and 50 NTU) were tested using all three foam pore sizes available: 30, 60 and 90 ppi. The experimental set up is the same used for other experimental trials. One important note is that the alum dose changes with the changing influent turbidity. Previous work was done using a 1.5 mg/L alum dose for a 5 NTU influent. It was assumed that there was a linear relationship between influent turbidity and alum dose, thus the alum dose for various influent turbidities could easily be calculated.
After determining the influent turbidities that the filter will produce water under U.S. EPA standards of 0.3 NTU for at least 24 hours, we will be able to provide clear instructions for appropriate use of the filter. These instructions are important because while running the filter under high influent turbidities provides lower effluent turbidities, it also makes foam collapse happen much more quickly. Foam collapse is a failure mechanism that permanently alters the structure of the foam, and its ability to effectively filter water.
We first began experimenting with 30ppi polyurethane foam. Our experimental setup consisted of a column with 10 layers of stacked pieces of 2.54cm thick foam. We then ran 5, 20, and 50 NTU water with a residual alum concentration of 1.5 mg/L through the filter for 24 hours using an approach velocity of 4 mm/s.
Figure 1: 30ppi foam, 50 NTU influent
Figure 2: 30ppi foam, 50 NTU influent
Figure 3: 30ppi foam, 20 NTU influent
Figure 4: 30ppi foam, 20 NTU influent
Figure 5: 30ppi foam, 5 NTU influent
Figure 6: 30ppi foam, 5 NTU influent
We also want to ensure that we would get a consistent and predictable performance from 30ppi foam. In order to verify this, we ran the experiment with a 50 NTU influent 2 additional times and compared the results from the 3 trials.
Figure 7: 30ppi foam, 50 NTU influent experiment comparison
These results indicate that 30ppi foam can produce a maximum pC* of 0.769 (Figure 4) which is well below that of typical sand filtration units. Even at the lowest tested influent turbidity, the 30 ppi foam could only reduce the effluent turbidity of the water to 1.32 NTU(Figure 5). Since our goal is to reduce the effluent turbidity of our filtered water to below the US EPA drinking water standard of 0.3 NTU, we determined that 30ppi foam does not perform well enough to be used as a filter media for our design. The fact that we were not able to reproduce any of the results from our 50 NTU influent experiment with 30 ppi foam is also of great concern since we want to be able to predict the performance of our filter.