Felix Yang, Lilly Mendoza, Ken Rivero-Rivera
The 1 Liter per Second (LPS) Plant testing team is continuing the work done by previous semester's Pre-Fabrication team by attaching the ESTaRS to the 1LPS plant and flocculator. By doing this AguaClara will have a complete 1 LPS Plant running in the lab so that experimental data can be gathered. This data will be used to optimize and improve current designs as well as further iterations of the technology.
Yuhao Du, Andrew Kang, Alaia Malaina
Two Stage Coagulant Addition team was designing and fabricating an apparatus that was used to experiment the benefit of having two stage coagulant addition, in opposition to one, while maintaining an efficient effluent turbidity. "Two stage" stood for adding clean coagulant after all previously added coagulant was fully covered by humic acid, which was the experimental substitution of NOM in the water. The research was based on Yingda's thesis, who studied the effect of humic acid on the coagulant dosing model. Based on the bonding mechanism, it could be hypothesized that two stage coagulant addition would increase the removal efficiency of particles. This semester, our group focused on comparing 1 stage addition with 2 stage addition and carrying on the experiment with different influent turbidities, as well as trying to improve Yingda's coagulant dosage model. By having a fixed HA/Clay concentration, we designed a ProCoDa increment system that will allow ProCoDa to increase the dosage of coagulants for separate experiments, without having to manually ending the experiment.
Lab process set-up
Ziwei (Vanessa Qi), Ye Rin (Erin) Kim, Wen Tien Dai
The Fall 2017 Humic Acid team was motivated to study the impact of the humic acid particles in water. Throughout the Fall 2017 semester, the team plans to explore the existence of optimal coagulant dosage that gives the lowest effluent turbidity at various humic acid concentrations. Then, the team seeks to set up a mathematical model that calculates the optimal coagulant dosage vs. humic acid concentration. A series of controlled experiments will be conducted with a computer software.
Lab Set-up
Lucinda Li, Liz Cantlebary, Lingzi Xia, Dylan Vu
Sand filters have historically been used to lower the turbidity of water, and continue to be used in many conventional water filtration systems. Dynamic modeling, as opposed to static modeling, of rapid stand filtration accounts for the buildup of particles over time in the filter, and this un- derstanding is needed for better filter design and operation. Past sub-teams found that head loss increases linearly with time. This research proposes the hypothesis that flocs are captured in rings created by filter grains, which on a larger scale implies an active filtration zone where empty pores become clogged by the flocs. This active zone moves throughout the bed until there is no remaining space for particles to clog. This research will examine major and minor head loss, along with efflu- ent turbidity, to find optimal filter performance based on varying flow rate, coagulant dosage, and influent turbidity in a 1.967m L/s sand filter. Based on this research, it is hypothesized that the sand bed can filter a certain amount of mass before failing.
Cheer Tsang, Yeonjin Yun, Ben Gassaway
The introduction of coagulant into turbid water causes collisions of suspended particles with coagulant nanoparticles, which promotes the growth of flocs. However, a large portion of the coagulant dose adheres to pipe walls rather than influent particles, requiring a higher than necessary coagulant dose to account for this effect. In order to minimize coagulant waste, an apparatus called the contact chamber was fabricated to increase collisions between influent particles and coagulant. The Fall 2017 Contact Chamber team analyzed the performance of the contact chamber by comparing influent and effluent turbidity in experiments with and without a contact chamber. After several trials, it was concluded that the contact chamber did not improve the effluent turbidity. In fact, the effluent turbidity with the contact chamber was significantly greater than the effluent turbidity without the contact chamber.
August Longo, Victoria Zhang, Michelle Cheng
The Fluoride subteam seeksto dvelop a sustainavble, inexpensive fluoride removal system for implementation in upcoming AguaClara plants located in India. After earning an EPA Phase II grant for the Spring 2016 fluoride remocal reactor, the subteam continued to improve fluoride purification by testing lab-scale systems consisting of either a single reactor or two reactors in series. During the experimentation process, clay was incorporated into the influent stream to abate PACl buildup at the bottom of the reactors. As the subteam seeks to transplant their system to India, it is currently working to optimize fluoride removal by minimizing use of resources. Thus, the team is currently trying to eliminate the use of clay and lower PACl dosages by increasing upflow velocity and further redesigning its reactor. Initial tests suggested insufficient fluoride removal for potable effluent, but the team is looking to repeat its previous experiments using the summer 2017 High Rate Sedimentation subteam’s reactor design
Roswell Lo, Tanvi Naidu, Luna Oiwa
The High G Flocculation team this semester designed an experimental set-up to test the effects of velocity gradient (G) in a flocculator and to determine the optimal G value based on flocculator performance in terms of effluent turbidity. The G value was varied in different trials by varying flocculator flow rate while controlling for coagulant dosage, influent turbidity, flocculation tube length, and upflow velocity through the sedimentation tank. G_theta was kept constant as around 20,000. The constant sedimentation tank upflow velocity was achieved using a waste stream between the flocculator outlet and sedimentation tank. It was found that for a standard coagulant dose, lower G values were associated with lower effluent turbidity, with 100 Hz being the lowest value tested. The same general relationship was observed for a higher coagulant dose, except that the lowest G values resulted in higher effluent turbidity due to floc blanket collapse. Data from this study will be used in the future to inform the geometry of the flocculator, i.e. the optimal distance between baffles in a full-scale water treatment plant.
Ji Young Kim, Steven Lopez, Fletcher Passow
AguaClara plants effectively remove turbidity and fecal indicator bacteria (FIB) from drinking water sources in Honduras. However, no study has documented the plants’ ability to remove specific diarrheal pathogen species. This study identified best practices for collecting and shipping pathogen DNA from sampling locations in Honduras back to Cornell University. A literature review identified options for filtration systems, chemical DNA preservation solutions, and shipping protocols. Tests of the clogging behavior of 5um and 0.1um pore size membrane filters demonstrated that this filtration method’s 60 h projected run time for a 10L samples outweighs its gains in simplicity.
Grace Zhang, Lois Lee, Srilekha Vangavolu
The Fall 2017 Sensor Development team worked on redesigning and calibrating the fluidized bed solids detector to be used for testing concentration of suspended particles in floc blankets as well as the submersible sensor designed to measure the height of the sludge blanket in the sedimentation tank.
Please note, no research report could be located for this team and the final presentation is linked instead
Priya Aggarwal, Will Lopez, Ana Ruess
The Fall 2017 Ram Pump subteam worked on mathematically modeling the ram pump’s mechanical behavior. Experiments conducted the previous semester proved that the ram pump does not operate as anticipated or desired. Ideally, modeling will explain this unpredicted behavior. With this knowledge, the team will be able to produce a more efficient and effective design. The team found a way to derive the forces involved in the pump, but more work needs to be done to determine what the optimal spring force is for the system.
