...
A linear dosing meter is currently being utilized in current AguaClara plants designed for Honduras to control the flow of alum, based on varying plant flow rates. In this system, the alum flows from the stock tank to the constant head tank through a small 0.23 cm orifice. This orifice is becoming becomes clogged with an a chemical precipitate that is probably a combination of calcium and aluminum hydroxide or carbonate unknown substance, preventing the flow of alum into the entrance tank and consequently reducing the effectiveness of flocculation. The constant head tank orifice has to be cleaned multiple times per day by the operator.
A more ideal system would clog at most once per day. The objective of this experiment is to determine the cause of the clogging in Honduras (precipitant or foreign material) as well as materials) (I'm not sure what else it could be? Hair?) and estimate the frequency of clogging in the new non-linear dosing system. If the new orifice (0.1 cm diameter) clogs frequently with lab grade alum, the system will need to be redesigned to incorporate a larger orifice. If the source of the clogging is determined to be foreign material (little or no clogging occurs with lab grade alum) a strainer will be included in the overall system design to remove the foreign material.
...
The experimental setup accurately mimics the conditions in the non-linear dose controller without incorporating the additional hassles of a large stock tank, constant head tank and entrance tank. As seen in Figure 1, the water circulates through a peristaltic pump to a small one liter reservoir intended to steady the pulsing input from the pump. The reservoir is connected to a tee, which connects a pressure sensor to the system just before the orifice. Water then drips through the smallest orifice, 0.1 cm diameter, into another one liter reservoir. This reservoir is connected to the pump, creating a closed system. The pump is set to discharge water at the same rate as the water flows through the orifice to maintain steady-state operation. The experiment was first be run with deionized water as a control. Then it was run with the maximum concentration of alum to be used in a plant, 125 g/L. As the experiment runs, Process Controller records the pressure directly upstream of the orifice. A safety sensor is connected to the second reservoir to provide further validation of the presence of a clog. As the pressure at sensor 1 increases due to a clog, the pressure at sensor 2 decreases due to decreased inflow from the orifice.
Results
First, a A control experiment, using deionized water, was performed to depict normal pressure readings without the possibility of a clog at the orifice. These pressure readings serve as a basis for comparison to the data obtained when alum is run through the system. The results from pressure sensor 1 can be seen in the graph below.
...
Next, alum was used in order to determine the time it would take for a clog to form in our 0.1 cm orifice. The experiment was run for four days but the most conclusive data was seen within the first hours. The pressure readings, from pressure sensor 1, are shown in Graph 2 below.
...
The elevated pressure, shown in Graph 2, seems indicative of a clog, but the spike was not repeated at all for the remainder of the experiment. If the clog was "blown out" due to the increase in pressure then it should simply reform after another 4 hour time span, forming a clogging cycle. (Precipitation and blow out could possibly be more intermittent than you would periodically predict.) Yet this cycle of perpetual clogging and clearing of the orifice was not observed. This was unexpected and might indicate erroneous pressure readings for the spike.
Conclusions/Future Work
The 0.1 cm orifice appeared to clog after 4 hours, indicating that a larger dosing orifice might need to be used. Further testing will need to be performed in order to verify these results, and determine the clog time for any larger orifice we select. An ideal system would only require cleaning once a day. (Already stated this earlier)
Since lab grade alum was utilized in this experiment and a clog was still present, the presence of foreign material, might not be the main cause of the clogs present in the Honduran plants. An inline strainer would still be recommended for future dosing systems (Is the source water utilized to mix the alum really that dirty that we would need to strain it?), yet it should be noted that this would not eliminate the occurrence of clogs since they may also appear to be caused by alum precipitatethe precipitation of aluminum and calcium species in the water.