Request Further Information
|
The figures on the following pages, Figures 5, 6, 7, 8, are the test results on waters from various sources. They show the kind of important and revelent information that can be obtained from bench scale jar testing in the laboratory. Once the dosages are determined, including dilution and application sequence, etc., the testing for flocculation mixing time and energy input begins. This has such a wide range of possibilities that experience is very helpful in arriving at an optimum reasonably fast. In coming up with some trial alternatives, an idea of the plant flocculation system is important. If the job is to improve an existing plant, the system is there on the site to analyze. In all probability, an existing plant would have vertical paddle, horizontal reel, oscillating paddle, or walking beam type mixers. Some of these systems are readily adapted to the tapered regime of energy input in the flocculation and some are not. Obviously, in new plants the designer can specify a system that will give the desired results. In an existing system, every effort is necessary to utilize the existing equipment and structures. For example, if 4, 6, or 8 compartments are already available, the bench scale work should be carried out with mixing regimes which are compatible. Theoretically, for 20 minutes of flocculation in four compartments, the flow would allow 5 minutes in each. The testing work then could be oriented accordingly. Similarly, for 8 compartments, there would be a 2.5 minutes for each. In a horizontal reel system, the possibilities are much greater because it can be made into a continuous channel with G values and time varying as the water flows through the system. Experience in sorting out this complex matter is fairly consistent for clarification of colored and turbid waters. A short period of high energy agitation followed by a relatively long period of low agitation either with or without a polymer has been very effective. This is only a general observation and could be wrong in any specific case. This is, however, a profitable place to start since a beginner could easily get lost in the maze of possibilities. An example of a beginning then could be the optimum coagulant dose with the addition of 0.1 mg/1 of polymer added just prior to the speed reduction at 5 minutes then continue for another 15 minutes at 20 RPM. This would be a polymer for floc bridging and building, probably high molecular weight, non-ionic. Total flocculation time would be 20 minutes. The sampling is the same in that the objective is to construct a settling velocity distribution curve which will guide the improvement of the existing or the design of new flocculation and settling basins. The settled water samples, therefore, should be taken after 1, 2, 3, 5, and 10 minutes in order to obtain sufficient data. Turbidity of each sample is read and recorded. |
The settling velocity curves plot turbidity remaining in the sample on the vertical axis (log scale) against settling velocities in cm/min on the horizontal axis (arithmetic scale). This flocculation cycle is repeated until the optimum mixing regime is determined. At this point in the bench scale work the coagulant and polymer dose along with mixing regime have been determined. Settling velocity distribution curves have been constructed from the turbidity samples taken from the test run on mixing regimes. Obviously the testing for optimum mixing regime involves many trials to arrive at the optimum. The optimum mixing time is the next variable to be determined. This is done by using the optimum mixing regime and stirring the jars for different times from 10 to 50 minutes or more. The optimum coagulant and polymer dosages are used. In other words, the only variable is mixing time. For example:
There is no practical reason to go beyond 80 minutes of flocculation. The jar No. 1 would be removed or the stirring paddle stopped after 10 minutes and turbidity samples taken at 1, 2, 3, 5, and 10 minutes. The procedure would be the same for each succeeding jar until all were completed. Since there are 8 jars, two runs of 4 jars each would be required. The data are recorded but in this test flocculation time in minutes is plotted on the horizontal scale while turbidity is plotted on the vertical scale. Both are arithmetic scales. Typically the clarification improves very rapidly during the first 10 to 20 minutes. After this time the curves flatten out with relatively little improvement, then after 30 to 40 minutes the turbidities begin to increase indicating the beginning of floc deterioration due to excessive mixing. The weaker the floc the more rapid will be the deterioration. The accompanying figures are typical of the clarification relative to time (see Figures 9 and 10). Floc weakens more quickly in cold water and deteriorates faster. The other variables which need to be determined are coagulant dilution, effect of and amount of sludge return, and sequence of chemical addition. The optimums for these variables are determined in the same manner as that described for determining flocculation mixing regime and the data are plotted in the same way. This laboratory jar testing program results in the accumulation of a large body of information upon which design and operation decisions can be made. These include the design of physical structures as well as the process. It should be understood that the data represent the experience in treating the water under the conditions at the time of the testing work. If the raw water changes in turbidity, temperature, alkalinity, or in other ways throughout the year, the testing should be done under these changed conditions. For this reason, it is important that operators be trained to do this work the year-round. From the information obtained in the testing, work decisions can be made on treatment process design which would include:
The decisions on physical process would include:
One other variable should be tested and that is the effect of pH adjustments in clarification. This is done by fixing all variables except the pH which is varied by adding lime. The pH range is usually varied from 3.5 to about 12 and titration curves plotted with lime dosage against pH. Figures I I and 12 show two different titration curves. It should be noted how different each water reacts and therefore how important it is for the operator to know as much as possible about his specific water. |