Flocculator Setup
The experimental setup, including equipment, chemicals, and computer software, remained nearly identical to what has been used in past semesters. The main features of the flocculator setup include one 11 gallon bucket containing an alum concentration of 1.5 g/L (originally at 3g/L, but was diluted because flow rates were too low) and another 11 gallon bucket containing a kaolin clay-water mixture of 750 NTU, or roughly 2.02 g/L of clay per liter of water. The alum stock bucket was connected to a Masterflex L/S peristaltic pump manufactured by Cole-Parmer Instrument Company by a size 13 flexible rubber tube, whereas the clay stock bucket was connected to a peristaltic pump by a size 14 flexible rubber tube. Raw water from a temperature controlled constant head tank connects to a peristaltic pump with two pump heads by two size 17 flexible rubber tubes. All three pump rates are controlled independently based on the experiment and controlled by the computer program Process Controller. The clay solution and raw water streams are joined and the two mix before entering the influent turbidimeter. One MicroTol 2 Turbidimeter manufactured by HF Scientific, Inc. is connected to the tube leading into and one out of the tube flocculator. The alum solution connects with the rest of the influent solution after the influent turbidimeter, as alum should not be figured into the influent turbidimeter reading, as it is added for the purpose of cleaning the water. A short length of tubing coiled with a 3 inch diameter and a tubing size of a ¼ inch inner diameter - smaller than the flocculator tubing - is placed in between the influent turbidimeter and the start of the tube flocculator, acting as a rapid mix unit. After rapid mix, the solution flows through a clear plastic tube of 6mm inner diameter of varying length (depending on the specifications of a particular experimental run) wrapped around a large, hollow plastic cylinder. This part is the tube flocculator, where particles will have a chance to react with the alum to form flocs. After flocculation, the flocs will flow through the effluent turbidimeter, which consists of a long, clear plastic tube placed through the center of the turbidimeter sensor in order to allow better flow of flocculated material through the sensor and to facilitate clarification once the flow is stopped. The effluent turbidimeter also provides visual confirmation that flocs have formed and settle out at differing rates depending on the size of the floc.
Over the Fall 2007 semester, we have found better ways to manage our experimental setup. Changes have been made only as regular maintenance or if the change increased flexibility in the experiment. We regularly replaced older pumping tubes with new tubes to prevent leaks due to wearing from the constant pump moving over the tube. The lab setup now includes flocculator tubing lengths of 10 feet, 25 feet, and 50 feet. These lengths are used individually during an experiment and can simply be connected by the user. The different lengths of the flocculator tube dictate various flocculator volumes, which ultimately changes the residence time of the flocculator when coupled with certain flow rates. Floc formation and density partly depend on the flocculator residence time. With more time spent in the flocculator, the flocs have more interactions with other flocs and the floc size increases. At shorter flocculator tube lengths, particles have less time to interact, therefore, flocs are smaller.
Additionally, the team replaced the old flocculator tubes with a set of clear plastic tubes that are more flexible and transparent than the former. Since the new tubes would be crushed under pressure because of the flexibility of the material, it required the flocculator to be balanced on a PVC pipe platform. This platform helps prevent the soft flocculator tubing located on the bottom of the flocculator cylinder to lose shape under the weight of the cylinder.
The rapidly declining temperature due to the changing seasons required attention to be given to the dissolved oxygen concentration in the tap water. The decrease in temperature had caused the formation of air bubbles in the settling column during the quiescent settling state. The bubbles were primarily forming along the wall of the column. The presence of the bubbles may affect the ability and accuracy of the turbidimeter in determining the actual turbidity of the fluid it is sampling. In order to prevent the air bubbles from forming inside of the settling column, the drainage pipe at the effluent of the entire experimental setup was raised 1 meter above its previous level to raise the overall system pressure. The higher system pressure should compensate for the higher raw water dissolved oxygen concentration by keeping the gases in solution throughout the system. Air bubble formation inside the settling column was no longer visible after raising the system pressure; therefore, it can be assumed that an extra one meter of head was able to compensate for the effect of the colder water temperature coming into the laboratory.
A LABVIEW based program called Process Controller, a computer program which controls the experimental setup by managing alum, clay, and raw water flows, also dictates the different cycles of the experiment. Different Process Controller programs can be created to run different experiments, changing parameters such as flow rate, flocculator volume, or the clay stock concentration. The clay stock concentration, the desired alum, and the clay dose, which is equivalent to the desired influent turbidity, are user defined inputs in the Process Controller program. Both the alum and clay flow rates are functions of the flocculator flow rate, the user defined stock concentration and the desired dose. A pump control function then translates the calculated flow rate into a scaled value recognizable by the pump. One of the major challenges of working on this research project is learning about how to control Process Controller to have it compute and run a specific experiment. During this semester, we developed several programs for various experiments, including an increment function to vary the turbidity of the influent water, and a second increment function to vary the G, or flow rate of the flocculator.
Process Controller automatically collects data from the two turbidimeters at a user defined interval. For most of the semester, this interval remained at 5 sec (12Hz) until the last few experiments were changed to a 1 sec (60Hz) data collection interval in order to retrieve more data. Also, the turbidimeters display a real-time turbidity reading every second. In order to retrieve this real-time data, the data collection interval in Process Controller had to be 1 s too. After the data is collected and stored, it must be analyzed to extract conclusions. The MathCAD data processor functions extract data given the folder in which the datalog(s) and statelog(s) are stored, the date(s) the experiment occurred, and the state to start, which are all user defined. The data processor will extract the number of runs and states the experiment cycled through. Additionally, the data processor has a function that can graph any of the variables of the experiment, such as incremented influent turbidity, effluent settling turbidity, and flocculator flow rate on a 3-D plot. MathCAD is also another program that takes time to learn and understand how to control in order to retrieve the desired results. After running an experiment with varied flow rates, it was observed that the data processor was copying over information from one run to the next when the array sizes did not match in order to maintain a constant array size. This array malfunction was corrected with a line of code in the MathCAD data processor.
Regular maintenance and cleaning of the apparatus was critical in ensuring accurate data collection. The regular cleaning of the influent turbidimeter vial and the effluent turbidimeter settling column were of particular importance, because impurities and residues collected on the walls of these two instruments interfere with the turbidimeter's ability to accurately determine the turbidity of the water. Before each run, the turbidimeters are powered down and the vial and column are carefully removed for cleaning. After powering down the instruments, all water must be drained out of the chambers and the instrument must be kept dry. For the influent turbidimeter, the glass vial can be removed from the head cap by twisting, but the influent and effluent tubes must be pinched shut to stop the flow of water in and out of the glass vial. For the effluent turbidimeter, the tube leading out of the bottom of the settling column must be disconnected and the manual valve connected to the top of the column must be manually opened to drain the water inside the column. Once all the water is drained, the metal connectors connected to the top and bottom of the column must be removed before the glass column can be pulled out from the top. Once carefully removed from the turbidimeter, a soft foam brush should be used to dislodge and remove all the residue and contaminants on the inner wall of the column. When clean, insert the column from the top down and secure all connectors tightly and reconnect all tubes to prevent leakage. Once the apparatus is reconnected, run the state that cleans the effluent turbidimeter with the manual valve open to remove all the air from the system. Once all of the air has been removed, the apparatus is ready to be used again.
Back to Tube Floc Research