PSS Fall 2009 Ramp State Experiment

Overview of Methods

The critical velocity is when the velocity on the outer edge of the floc particle is equal to the floc settling velocity. Any velocity exceeding the critical velocity is when floc roll up begins.

In order to experimentally determine the critical velocity at which floc roll up begins, flow rates through the tube settler were increased incrementally utilizing a ramp state function in process control software.

By incrementally increasing the flow rate through the tube settler, we can compare the effluent turbidity performance over time. A critical velocity could be identified based upon effluent performance and compared to our theoretical model.

Using the same experimental apparatus as was used in Summer 2009 and Spring 2009, and the ramp state process controller function, we hope to understand if our theoretical model of floc roll up behavior describes system behavior. Ultimately, we hope to minimize the floc roll up in the plate settlers and further reduce the effluent turbidity. Also, we want to potentially understand how to create flocs that will experience less roll-up and have better performance.

Results and Discussion

By running the ramp state function on different tube settler diameters, we hope to further develop the relationship between plate settler size and floc roll up.

Experiment 1: Ramp State with 9.5 mm Plate Settler Tube Diameter

This experiment started with a flow rate of 6 mL/min and over the course of 24 hours, gradually increased to a flow rate of 50 mL/min. This flow rate range corresponds to a capture velocity range of approximately 0.127 mm/s to 1.053 mm/s.

Figure 1: Effluent Turbidity vs. Flow Rate

Experiment 1 Conclusions

The very clear spike in effluent turbidity observed at a flow rate of approximately 18 mL/min represents the point at which the floc particles began to roll up the tube settler, which was confirmed visually in the experimental apparatus. The velocity represents the critical velocity. At a certain velocity, the turbidity stabilizes, and stops increasing. This is because, at a certain point, the number of flocs rolling up in the settler cannot increase anymore and thus the turbidity cannot increase anymore.

These experimental results can be compared with the expected results of our theoretical floc rollup calculations. Theoretical calculations for a 9.5 mm diameter plate settler tube predict that floc rollup should start to occur at a flow rate of 15.693 mL/min. Comparing this theoretical value with the observed floc rollup flow rate of approximately 18 mL/min above, we see that the experimental observations support the theoretical calculations quite well, within experimental error.

Experiment 2: Ramp State Function with 15.3mm Tube Settler Diameter

This experiment starts with a flow rate of 6 mL/min and gradually increases to a flow rate of 140 mL/min over the course of 24 hours. This flow rate range corresponds to a capture velocity range of approximately 0.127 mm/s to 2.963 mm/s.

Figure 2: Effluent Turbidity vs. Flow Rate

Experiment 2 Conclusions

Unlike the results for the 9.5mm tube, there is no obvious sharp increase in effluent turbidity. Although the turbidity of the effluent water does increase as the flow rate is increased, this change in turbidity is minimal compared to the 9.5 mm tube. The effluent tubidity is slightly higher than our standard of 1 NTU, but this is not a significant enough difference to assume that floc rollup has occured. Furthermore, theoretical calculations for a 15.3 mm diameter plate settler tube predict that floc rollup should start to occur at a flow rate of 65.557 mL/min. Analyzing the results plotted in Figure 2 above with this in mind, we see no visual confirmation of this in the form of a sharp peak in turbidity in the data around the predicted flow rate. Therefore, we have concluded that there is no clear evidence that floc rollup has occured. Due to the fact that the predicted values from our theoretical model calculations matched the observed results for the 9.5 mm tube quite well, we propose running more experiments with the 15.3 mm tube to verify whether the above discrepancy was due to experimental error or a problem with the theoretical model in predicting floc rollup in larger diameter tubes.

Troubleshooting