Jar Testing: Getting started on a low budget by David Pask NDWC Technical Services Coordinator By now, almost everyone realizes that to get optimum performance from your filtration plant, it is necessary to do what are commonly known as "jar tests." Good operators may make treatment adjustments as they notice changes in raw water quality or temperature, judging their results by the look of the floc in the clarifier or filter. However, to ensure maximum efficiency, you should at least conduct jar tests. A jar test is simply a pilot-scale test of the treatment you are using in your plant, and is used to determine if you're using the right amount of chemicals. I can hear some of you saying, "But I don't have all that fancy lab gearŅor the time to use it." However, jar testing need not be difficult or expensive, in time or equipment. You can get started with substitute or make-do equipment; then buy the proper laboratory equipment and supplies as the need arises or as funds are made available. One should always acknowledge that tests to control a process do not need to be as precise or follow the exact specification as analyses that are used to verify compliance with regulations. For instance, I can look at a sample of water through a 24-inch-long tube and be reasonably certain that the sample is less than one Neph-lometric Turbidity Units (NTUs), but, if it were possible, I would prefer to have a turbidimeter on my bench (at a cost of around $850) to be sure. Why perform jar tests? By performing jar tests, you can: - try alternative treatment doses and strategies without altering the performance of the full-scale treatment plant, and - easily compare the results of several different chemical treatments for time of formation, floc size, settleability, and perhaps filtration characteristics. One cannot make such comparisons with the full plant's treatment. Therefore, when the quality or temperature of your raw water changes, do a jar test before you change the chemical dosing pumps, and you will likely get the right results the first time: no turbidity breakthrough, no unusual bacterial counts, and no complaints from your customers. How are they done? Jar tests simply entail adding treatment chemicals in the right amounts and sequence to a sample of raw water, which is in a jar or beaker. The sample is then stirred, so that the formation, development, and settlement of floc can be watched, just as it would be in the full-scale treatment plant. A series of tests are then performed to compare the effects of different amounts of flocculating agents at different pH values to achieve the right size floc for the requirements of your particular plant. What equipment is needed? The basic equipment requirements include: a) a stirring apparatus for 2 to 6 jars or beakers; b) a test kit or meter to measure pH and alkalinity; c) stock solutions of treatment chemicals; d) measuring cylinders and pipettes to measure raw water and chemicals; e) a thermometer; f) a clock or timer; g) a measuring tape, calculator, and notebook (used to calculate, to record results, and for future reference). If you are using a new coagulant or filter aid, you will need either pre-weighed samples or a small chemical balance and weights to enable you to make up accurate solutions. Of course, if your system can afford one, a turbidimeter is another useful addition to your equipment supply. A new six-gang stirrer will cost about $850, but one can perform useful work with two magnetic stirrers, which cost about $250. For occasional use, I have used two-gang stirrers made from workshop scraps, but for this article I made a set from parts available at national hardware and electronics suppliers for a total cost of $25 (excluding labor). This handmade set of stirrers is shown in the photograph on page 4; a complete parts list is available by calling me at (800) 624-8301. The only awkward part of building the stirrers is reaming out the fitting cross, so that it will slide easily onto the 3/4-inch pipe. To do this, I used a piece of hacksaw blade and set it into a slot that was cut in the end of a section of pipe (see graphic in Figure 1). Also, the 1.5-volt motors are not really designed to operate at such a low speed; however, they will work if each is controlled with a 25 ohm, two watt, variable resistor. One alkaline "D" cell battery will provide at least 10 hours of continuous stirring. (Of course, using batteries will ease the minds of your safety and insurance officials, who prefer that you do not have any non-certified mains voltage equipment around the plant.) There also are practical alternatives to using standard laboratory equipment for some measurements. I particularly recommend using plastic disposable medical syringes (without the needles), instead of pipettes. The syringes are accurate enough, and if you have one for each chemical, they can be preloaded for rapid dosing into the jar. I have found the 3-milliliter (ml) and 20-ml size syringes to be suitable (1 ml = 1 cc). The "square" mason jar has advantages over the standard laboratory beaker when used for flocculation testing. With the mason jar, the turbulence is increased and the overall rate of rotation is reduced, making observation easier. I still use a readjusted cheap jewelry balance (which costs about $20) with a set of lab weights (about $25) for reagents. A basic laboratory balance weighing to 0.01 gram can be obtained for about $100. I have a 50-ml lab cylinder, but also use a 500-ml kitchen measuring jug (which I calibrated at 502 ml). These substitutes are about as accurate as your supply meter and should not, therefore, provide misleading results. What is the proper procedure? Unless this is the start-up of a new plant, you will already have the chemical dosing rates that provide acceptable treatment results. Our objective with jar testing is to tune the plant for optimum performance; that is, for the longest possible filter runs with the least usage of chemicals to provide consistent, good quality water that is within required standards. Before you begin jar testing, however, take an hour or two to check all of your equipment. - Is the inlet flow meter accurate? Shut off your supply pumps and measure the rise in level of your clear well against the readings of the meter. - Shut off the filters and do the same test on the supply pumps. - Check that the chemical dosing pumps are delivering the correct rate by measuring the rate of fall in the daily supply tanks. - Check that there is no accumulation of solids in the settling tanks and that the filter sand is clear of any "mudballing" and is properly graded and up to level. - Check also that the backwash pumps are operating at the set flow rate by timing the rise in level at the start of the backwash cycle. - Correct, or at least make a note of, any deficiencies. The most awkward part of your jar tests will be the calculations to determine the correct amount of chemicals to add to the raw water in your jar. I find this much easier if you work in metric (SI) units, and then convert back to your own equipment units to transfer your results into practice. See the examples listed in the box on page 5. What now? This article only "scratches the surface" of the chemical treatment process for flocculation and filtration. Much more can be learned from a good manual and from the training sessions conducted by your local Rural Water Association and American Water Works Association. (For these organizations' telephone numbers, call 1-800-624-8301.) What I have tried to illustrate is that you can achieve useful results with a modest outlay of time and money. The improvement in your plant performance may then help to persuade your manager or community to invest in better laboratory equipment, which you will undoubtedly need in the future.