Functional Description and Troubleshooting of FReTA Components

Clay Stock Preparation, Clay Stock Valve, and Raw Water:

Generally we use a clay stock concentration of 10 g/L for high NTU tests (100 NTU or greater), and concentrations of 3.33 g/L for low NTU tests (less than 100 NTU). It is not necessary for the clay stock concentration to be precise. The clay stock valve which meters in clay to the raw water is controlled by a feedback loop in Process Controller. If the raw water turbidity drops too low, as measured by the influent turbidimeter, the clay stock valve will open every few seconds to allow more clay into the raw water until it has reached the target turbidity.

Common Problems: If you find that the influent turbidity is too high or too low (or has large fluctuations) you may need to adjust the clay stock valve. First verify that it is not clogged and that it is opening only when the turbidity is too low at the correct intervals. If the valve is functioning properly, you may need to adjust the settings in Process Controller. The setpoint 'off time' controls the amount of time after the valve has closed before it will open again when the turbidity is below the target. It can be either increased or decreased. For mid to high NTU experiments, an off time of around 10 seconds generally functions well. For experiments as low as 5 NTU, the off time may need to be as great as 20 or more seconds. Additionally, the flow can be reduced by partially closing the manual valve before the clay stock valve. Currently we are working on putting in new tubing to the influent turbidimeter with a larger diameter; we wish to increase the flow rate to the turbidimeter and therefore reduce residence time in the loop to reduce the response time in the feedback loop.

Alum Stock Bucket and Pumps:

We usually use a stock alum concentration of 2.5 g/L. With this stock concentration, the pumps can handle alum dosages ranging from about 5-160 mg/L (for a plant flowrate of 5 mL/s). For lower alum dosages, we use a stock concentration of 700 mg/L. This allows for approximate dosages ranging from 1.5mg/L to 40 mg/L. Be sure to verify that the dosages you want can actually be delivered to the plant by the alum pump; higher or lower dosages may require adjustments to the stock alum concentration depending on plant flowrate. Additionally, fresh alum should be used at the start of each experiment.

Common Problems: Make sure that the alum pump is functioning; it is possible that the yellow cable connecting it to the stamp box may have come loose. Additionally, it is not uncommon for bubbles to form in the tubing leading to the alum pump when it is first connected to the stock bucket. If you find that there are large bubbles, they could restrict the flow and you may need to flush them out. Additionally, if you find that the pump is not delivering the correct dosage, it may need to be recalibrated in Process Controller.

Tube Flocculator:

The flocculator can be run at one of three lengths, either 2796 cm, 5592 cm, or 8388 cm. Occasionally it may need to be cleaned by running a piece of sponge through it during backwash.

Common Problems: It is important to clear the flocculator of large bubbles before beginning an experiment because they can cause floc breakup and skew results. Bubbles can easily be removed during backwash by opening the release valve on the influent tubing when they reach the beginning of the flocculator. Additionally, tubing that is not in use but has water in it should be plugged to prevent it from dripping on the equipment.

Influent turbidimeter:

The influent turbidimeter is programmed in to a feedback loop monitoring the raw water turbidity.

Common Problems: Generally the influent turbidimeter has few problems. After high turbidity tests, the cuvette may need to be rinsed out as some of the clay may settle out on the sides of the cuvette, interfering with readings. It is good to verify that the cuvette is not dirty, and clean it if needed before beginning an experiment.

Effluent Turbidimeter, Ball Valve, and Settling Column:

The effluent turbidimeter records the turbidity over time during the settling state as flocs settle down into the settling column. While this happens, the ball valve is closed, sealing off the column from the rest of the flocculator. This is when most of the useful data is recorded; the turbidity vs. time plot is used to contruct a turbidity vs. settling velocity plot in the data processor.

Common Problems: Occasionally you may notice that during a backwash, the effluent turbidimeter is reading turbidities above 0.3 NTU. This is higher turbidity than clean water from the tap, so it indicates that the settling column needs to be cleaned. In order to do this, turn off FReTA. Next, in Process Controller, open the valve labeled 'effluent to column.' Also, be sure to disconnect the extra loop of tubing at the end of the drainage tube so that the water can drain down into the sink. Finally, open the manual valve built in above the ball valve before the flocculator (careful of any water that may come out initially). You should see the water level begin to drop in the column. After the column has drained, remove brass fitting connecting the drainage tubing to the bottom of the column. Two special wrenches for opening these fittings can be found in Ian's drawer. Be careful not to apply torque on the glass column when unscrewing the fittings; use both wrenches to hold the fasteners. Next, use the large pipe cleaner found in Ians drawer to clean out the column, being careful not to scratch the glass. Finally, reattach bottom tubing, screwing on the brass fastener first by hand, and then finally with the wrenches. When reattaching the drainage tubing, be sure that the end of the column passes completely through the white Teflon ring in the fastener; otherwise it may leak.

If for some reason the effluent turbidimeter needs to be removed, drain the column as described above and disconnect the drainage tubing. The crews holding on the turbidmeter can then be loosened and slid out of their position. Finally, the turbidimeter can be lowered down off the end of the column.

The most critical type of failure that can occur in the settling column is a leak at the upper end of the column. When this happens, water will run down the settling column and can flow into the turbidimeter casing. In order to prevent this, we have added a latex cover around the column that will direct water over the outside of the turbidimeter casing. Additionally, Velcro straps have been added to prevent the column from sliding down out of place entirely. If there is a leak at the top of the column, stop any experiment and drain the column. You may need to remove the turbidimeter to adjust the connection. Like the wtih bottom fastener, the top of the column must pass entirely through the Teflon ring in the connector to provide a good seal. Tighten using the same wrenches.