Research 

Home Teams Resources Documents Photos Trips Join Search

Research | System Regulation | La34 Design | Language & Culture

Research and Development

Laboratory Scale Tube Flocculator

  1. Determine the minimum turbidity reading that the column detectors can measure and compare with the values that are obtained using the standard sample vials. The goal is to accurately measure down to at least 1 NTU. Make sure that ambient light is not influencing the turbidity readings.
  2. Determine the optimal alum dose as a function of turbidity. Pernitsky and Edzwald (2006) in their paper on Selection of alum and polyaluminum coagulants give some guidance on alum dose. I suggest testing alum doses that range from 0 mg/L to 40 mg/L while varying the turbidity between 0 and 100 NTU. We should include an alum dose of 0 mg/L for each turbidity level so we have a control showing what removal we get without any alum. The tests for determining the optimal alum dose should be conducted with the velocity gradient at approximately 20/s to ensure that we can build large flocs.
  3. After obtaining the alum dose as a function of kaolin turbidity the relationship should be turned into an equation and that equation can be used to set the alum dose for future experiments
  4. Measure flocculation extent** and efficiency* as a function of Gq (flocculator length), G (flow rate), and influent NTU. The goal of these experiments is to determine the maximum value that G can have as a function of Gq. The results will make it possible to intelligently design a flocculator with variable shear.
  5. Develop data analysis tools that return two parameters characterizing extent and efficiency from each flocculation run. Extent could be obtained from the initial slope of the turbidity normalized by the influent turbidity and multiplied by the distance to the turbidimeter sensor from the top of the detector column to give a value with dimensions of velocity (sedimentation velocity of the flocs). Efficiency can be measured as the final turbidity at a time that corresponds to a sedimentation velocity of 10 to 30 m/day.

A very important deliverable for the Tube flocculator research team is a manuscript that is ready for submission in a scientific journal (perhaps Water Research or Environmental Science and Technology). The manuscript preparation including a thorough literature review needs to begin before the middle of the semester.

*Flocculation efficiency is measured by the turbidity of the water between the flocs (measured after the flocs settle).

**Flocculation extent would be measured by the increase in floc size. This is related to the size of the flocs and hence to their sedimentation velocity. The sedimentation velocity is related to the rate of decrease of turbidity during the sedimentation phase.

Options for additional research using the Lab Floc Apparatus

Vertical Flow Hydraulic Flocculator (VFHF)

The VFHF research has multiple goals

  1. Gain experience operating a water treatment plant with a real surface water.
  2. Measure the settled turbidity as a function of Gq (sample position) in the flocculator and as function of G (flow rate).
  3. Develop a plan to create more homogeneous shear levels within baffled sections (see FLUENT) and to vary the shear level as a function of position or Gq in the reactor.

Flow Control Module (FCM)

During the summer of 07 the AguaClara team developed a FCM that is currently being tested at Ojojona Honduras. The FCM is not completely reliable and has had several failures. We need to identify each of the failure modes and develop strategies to increase the reliability of the unit. If we can develop methods to ensure reliable operation we will have a device that can be used in millions of locations for controlling chemical feeds for community water supplies.

The FCM should have the following characteristics

Demo Plant

Measure the effluent turbidity to see how well this plant performs. Compare results with the measured values from the tube flocculator at similar Gq values. Access whether increasing Gq would improve performance. Experiment with the addition of head loss elements lowered into the flocculator to increase the velocity gradient and improve the effluent turbidity.

Write an article for publication in a research journal such as Environmental Science & Technology or Water Research describing the demo plant in sufficient detail that others could build the plant.

Corrugated plastic supplier was GEpolymer shapes in Rochester (585) 254-5800

Potential improvements in the Demo Plant

  1. Modify the plant so it is possible to drain the sludge from the sedimentation tank.
  2. Change the design of the way the chemical dosing is varied so that the solutions don't leak out of the holes. An improvement in this technique could be applied to the full scale water treatment plants as well.
  3. Add a feed system for chlorination.

FLUENT

Check out the FLUENT website.

We need to use FLUENT to assess the variability in shear levels in the VFHF. Specifically we need to know the relationship between the spatial average shear level (as calculated from the mechanical energy losses) and the spatial maximum shear level. The ratio of the spatial maximum to the spatial average shear is expected to be a weak function of the Reynolds number. The velocity gradient ratio will be useful in determining how much higher the average velocity gradient could be set in a uniform velocity gradient flow field. Of course, we will not be able to design a flocculator with a completely uniform velocity gradient, but we may be able to reduce the maximum to average velocity gradient ratio significantly. By reducing the ratio we will be able to increase the average velocity gradient and thus decrease the residence time in the flocculator while still achieving the target value for Gq.

It is likely that more uniform velocity gradients produce more uniform flocs that are also less prone to disruption. In heterogeneous velocity gradient environments it is possible for flocs to grow large in regions of low velocity gradients. When those flocs move into regions with higher velocity gradients they are torn apart and likely leave behind some sub particles that high a low sedimentation velocity and that contribute to residual turbidity after sedimentation.

Spinning Bottle Test

Begin with a careful review of the theory of the spinning bottle test. Theory seems to suggest that smaller diameter bottles will more quickly reach the steady state velocity gradients during constant angular acceleration. Explore the possibility of using turbidimeter sample vials. Determine if it would be possible to allow the flocs to settle and then measure the turbidity by placing the vial in the turbidimeter. Explore all options for building a simple constant angular acceleration device including the fall weight system used currently as well as battery powered electric motors. Keep the goal of being able to use this system at AguaClara water treatment plants in Honduras.

Design

Detailed Design for Moroceli delivered to APP by October30

The deliverables include:

Design Specifications for Marcala delivered by mid Sept.

The Marcala plant is currently under construction with supervision by Diana Calix and Fred Stottlemyer. They need specifications on flocculator design (baffle spacing) and design of the channel that distributes water to the sedimentation tanks. We have hard copy documentation of the plant including drawings and photos.

Automated Design Software

The design software is currently able to draw the sedimentation and flocculation tanks with their baffles. Now we need to take this to the next level and add all of the piping and the sedimentation tank inlet channel as well as all of the items that go on the platform. The goal is to create a design tool in MathCAD that can take a few design parameters as inputs and create a detailed 3-D AutoCAD drawing. This design tool should be posted on our website and setup so that anyone can easily obtain a detailed design.

Structural Design

Research tank design and learn about techniques to create waterproof tanks. Explore alternatives including plastic tank inserts, Ferro cement, and brick with plaster.

Additional Research Areas for Future Study

POU Slow Sand Filtration

The AguaClara Municipal water treatment plants won't work in very small communities especially where the distance between houses is large and a piped distribution system isn't yet feasible. Point of Use treatment may be the only alternative in those settings. Slow Sand Filtration (SSF) has advantages over many other filtration technologies because it is cheap, is cleanable, has low head loss, and doesn't require any chemicals. SSF does have the disadvantage of only working well if the turbidity is normally below about 30 NTU. Like many POU treatment systems SSF will perform best if it operates at a constant flow rate.

Rapid Sand Filtration

Implementation of rapid sand filtration with backwash is technically challenging for small communities. The backwash requirement either requires many filters with deep filter boxes or a backwash pump.  Our assessment is that this technology is not yet developed to be sustained by small communities. Our recommendation is that filtration technologies be considered a second stage process if a community desires even higher quality water after successful implementation and operation of flocculation-sedimentation systems.

Pretreatment

Currently pending is the development of a method to reduce the influent water turbidity, especially during high turbidity events. Our project team plans to explore the possibility of using plate settlers before flocculation.

We need to learn why settled water turbidity tends to increase when the raw water turbidity increases. Theory predicts that flocculation should perform better when the particle concentration increases. We need to determine if the increased turbidity is due to inadequate Gq that leaves a small residual of primary (not flocculated) particles or if the problem is with sedimentation. It is possible that large flocs colliding release fragments. As the number of large flocs increases the release of fragments increases.

Rapid Mix

An economical hydraulic mixer has been fully developed, showing opportunities for small improvements. The unit is an inline static mixer, constructed of PVC pipe fittings. The unit could be improved by switching to tab-induced vortex mixing.

Sedimentation

Preliminary testing with upflow sedimentation tanks using sludge blanket flocculation indicated that performance exceeded that of plate settler sedimentation.  Additional studies will be done with well-designed comparisons (traditional horizontal flow, plate settler, and sludge blanket upflow) to optimize particle transport to the bottom of the tank.




Webmaster: J. Capurso & M. Weber-Shirk
Copyright © 2007 AguaClara | All Rights Reserved