AbstractObjectivesProceduresTotal Suspended Solids (TSS) Protocol:Results and DiscussionConclusionSuggestions and CommentsAbstractThe goals of this research were twofold: to utilize a Honeywell turbidity sensor to automatically control the settling state in a wastewater treatment plant and to utilize the sensor to maintain a set concentration of total suspended solids within the reactor. We were able to accomplish our first goal within the scope of our research- the settling time was effectively reduced by 95% thorough use of the turbidity sensor. We were also able to gain a better understanding about how to modify the sensor by adding six layers of Scotch tape to accomplish our second goal. A more permanent modification would be needed if the sensor were to be used for long-term monitoring. However, the most significant conclusion is that turbidity sensors can be effectively employed for these two purposes, especially if specifically designed for wastewater treatment plants.ObjectivesThe primary goals of this laboratory research project were to determine if it was feasible to employ a turbidity sensor to 1) reduce the settling time for the 4 L single batchreactor wastewater treatment plant (WWTP) and 2) control the concentration of total suspended solids within the reactor. Each of these objectives would increase the efficiency and fine-tuning control of the treatment process; the use of a turbidity sensor would enable the computer software running the plant to automatically optimize the time spent in the settling state and maintain the suspended solids at the optimal steady-state concentration (2-3 g/L).ProceduresThe experimental WWTP was setup as a single batch reactor with five consecutive and cyclical states: fill with waste, fill with water, aerate, settle, and drain. The plant was either manually or automatically controlled through use of a LabVIEW program. The time spent in the fill with waste, fill with water, and drain states was volumetrically controlled with a pressure sensors in the reactor. The time spent in the aerate state was temporally controlled, and initially there were two air flow rates from thediffuser: 2000 mol/s or 0 mol/s. Aeration was later controlled through a PID algorithm, using a much larger accumulator with a blow-off tube. The settle state was originally temporally controlled as well until we changed the LabVIEW code to use the turbidity sensor to control the settling state.The sensor we used was a Honeywell turbidity sensor. It was mounted and sealedwithin a cylindrical PVC reducer and the wire came out the top through a vertical cylindrical piece of PVC (Figure 11). The sensor readings were easily skewed by ambient light, so we covered the reactor with either a box or a black garbage bag during data acquisition to prevent any external light from reaching the detector. The turbidity sensor had a voltage range of 0.5 volts to 5.0 volts. Our first step was to test the unmodified turbidity sensor by monitoring the voltage readings for a wide range of total suspended solids (TSS) concentrations. We started with a concentrated solution of activated sludge (approximately 1.28 g/L) and then, step-by-step, we diluted the solution by replacing a certain volume of the solution with distilled water until the TSS concentration was approximately 0.02 g/L. Afterlogging voltage readings and taking 50 samples over the course of the dilution, we were able to take some of the samples and analyze them for TSS according to the following procedure, as described in the CEE 453 Laboratory Manual:Total Suspended Solids (TSS) Protocol:1) Using filter tweezers, transfer a prewashed and dried filter (Whatman GF/orequivalent) from a desiccator to balance and determine tare weight to nearest 0.1mg.2) Place in a Millipore® filtration apparatus, or equivalent, and apply suction.3) Pour through 50-mL sample (or other suitable volume).4) Wash sample container, then the filter holder and filter, with two 10-mL portionsof distilled water.5) Carefully remove filter with tweezers.6) Dry 1 hr at 103-105°C (use an aluminum dish to hold filter in drying oven and toprovide identification of filter).7) Cool1 in desiccator and reweigh to nearest 0.1 mg.Compute mg/L TSS as (mass of residue)/(volume of initial sample)This first experiment yielded a curve that showed the relationship between voltage readings from the turbidity sensor and the TSS concentration within the reactor (Figure 1). This relationship showed that at high concentrations the voltage reading was less than 5.0 volts. As the concentration was diluted, the voltage increased until it leveledoff at 5 volts (the maximum voltage sensitivity for the sensor) for about five sample readings. Eventually the voltage readings began to decrease until reaching 0.5 volts. From this experiment we were able to understand that at high TSS concentrations,particles tend to interfere with the photons of light traveling towards the detector, thereby blocking the detector. This explained why the voltage readings were low at high concentrations. This posed our first challenge to using the turbidity sensor as we had intended because we were hoping to us it to measure high TSS concentrations in the reactor.We then attempted to modify the sensor using a cuvette. According to Beer’s Law, if we shortened the path length there would be less solution for the light to travel through, thereby increasing the sensitivity of the detector. This would have enabled us to use the sensor to monitor high TSS concentrations. Yet, as described later in the results and discussion, this did not work as planned.We then tried new modifications including adding aluminum foil to hopefully reflect more light towards the detector since we could not directly intensify the light source. We also tried blocking the detector using layers of Scotch tape. Then we tried both of these modifications simultaneously. The specific results of each modification are described in the next section and Table 1. Eventually we ended up with a satisfactory modification using six layers of Scotch tape over the detector.This then enabled us to redevelop our initial curve with the new sensor modification (Figure 7 a & b). For this experiment, we decided to use a continuously stirred tank reactor (CSTR) setup. Again we started with a concentrated solution of activated sludge and then took samples over the course of the dilution. This time