Challenges Fall 2009

Please read through this list before ranking your preferred teams. Don't worry if you're not sure what everything means. Returning team members, Monroe, Julie, Matt, and Heather will be more than happy to get you up to speed. Just use this as a guide to see generally what kinds of things the teams are working on. Check out each team's page on the wiki for a more basic description of what they do and to see what work they've already accomplished.

Automated Design Tool

Team Leader: Heather Reed

Number of team members needed: 9-10

Important team member skills:

• CEE 3310, or equivalent Fluid Dynamics course
• CEE 4540 co-requisite
• Students must be comfortable with coding
• Students should be familiar with the AguaClara design
• AutoCAD and/or MathCAD knowledge is a plus
• We are willing to train new members

Challenges

Design support for the APP/AguaClara team in Honduras

Review the Agalteca design and make sure that design improvements created since May of 2009 are incorporated in the design

• Eliminate floc hopper
• Add weirs at the end of the inlet and exit channels
• Add a drain line to the exit channel (it won't be possible to drain the exit channel with the weir in place unless we add a new drain line)
• Add a removable drain port in the bottom of the sed tank to facilitate cleaning
• Consider making the bottom of the sed tank have a steeper slope to facilitate cleaning of the distribution tunnels
• Consider making the sludge drain have fewer orifices that are then larger in diameter (perhaps eliminate half of the orifices)
• Set the maximum energy dissipation rate at the end of the flocculator to match the maximum energy dissipation rate at the end of the Marcala flocculator. Refer to the CEE 4540 notes on Flocculation

Download Design Tool (ADT)

• Check that all pieces are scaling properly
• Identify design errors and work with the team in Honduras to develop improved design algorithms
• Update the list of variables that are returned to the user to ensure that all relevant parameters are returned.
• Eliminate variables in the Variable Naming Guide that aren't used.
• Develop a protocol to release new versions of the ADT. Begin by developing a method to create a stable release of the MathCAD files. Also develop a testing protocol that will use an independent check to verify that the plant dimensions are correct before releasing a new version.

Chemical Doser

We are currently switching from the linear flow orifice meter (LFOM) and laminar flow controller to a submerged orifice flow meter and orifice based flow controller.

• Need to develop AutoCAD code to draw this part of the plant
• Need to develop the design algorithms for the orifice based dose controller

Entrance tank

• A first draft of this code has been created, but it has not yet been reviewed.
• The Rapid mix and Chemical Doser needs to be drawn in the entrance tank.

Rapid mix components

The rapid mix design algorithm has been updated in the past year and consists of macroscale mixing followed by microscale mixing. The macroscale mixing could be a simple inline static mixer. The alum must be injected (through a submerged injector) immediately upstream from the inline static mixer. The microscale mixer is a submerged orifice at the bottom of the dividing wall between the entrance tank and the flocculator. This orifice is designed to have approximately 50 cm of head loss since it is this head loss that will also be used to measure the plant flow rate and drive the chemical dose controller. A PowerPoint presentation on the rapid mix process is available from the CEE 4540 syllabus.

This task should begin with a manual drawing showing how the rapid mix unit, entrance tank, flocculator, and chemical dose controller would fit together. After that drawing is reviewed by Monroe and by Agua Para el Pueblo staff the MathCAD to AutoCAD (MtA) code can be written.

Chemical storage tanks

• These tanks need to be drawn. These may be an item that the onsite civil engineer will relocate to fit site conditions, but the design tool should show them at the correct elevation and in a reasonable location.

Floc Hopper Drain Valves

• The floc hopper can be put on hold for this semester because we haven't been able to create a floc blanket in a full scale AguaClara plant. This task can wait until we develop a method to create a floc blanket.

Horizontal Flocculator

Continue coding the flocculator to include the option for horizontal flocculation for large plants and determine the flow rate for the transition from vertical to horizontal flow.

• Document and describe the solution algorithm for the vertical flow flocculator
• Design a horizontal flow flocculator and develop a clear algorithm for the solution process (see Flocculator PowerPoint from CEE 4540)
• Create the equation or system of equations that will determine whether a design will have a horizontal or vertical flow flocculator. The minimum flow for a vertical flow flocculator will be related to the minimum channel width given a channel width that is about 3 x the baffle spacing and the requirement that the baffle spacing be at least 45 cm for constructability.
• Develop a method to design both horizontal and vertical flow flocculators.
• Code the necessary drawing algorithms. Make the flocculator code as generic as possible to be able to handle both vertical and horizontal flocculators.
• Design and draw the drain system for horizontal flocculators.
• Determine if the client should be able to choose the flocculator depth (as independent from the sed tank depth) or if the client should be able to choose a design for either a vertical or horizontal flow flocculator.
  • If the client chooses horizontal or vertical flow, then the ADT would design for a depth equal to the sed tank when possible, but would design shallower floc tanks for small flows if horizontal flow is selected.
  • If the client chooses the flocculator depth then the ADT should default to horizontal flocculator design whenever possible to eliminate the problem with the drains.

Vertical Flocculator

• Need to ensure that the size of the ports between channels does not exceed the size available on the floc tank wall. Also, ensure that the spacing of the baffles in the last channel does not force them into the exit channel space (this may be taken care of by switching to horizontal flow).
• The drain system for vertical flow flocculators that are large enough that the baffles are ferrocement needs special attention. Given that the Marcala plant performs better than any of the other AguaClara plants and given that the Marcala plant uses ferrocement baffles with small ports at the bottom of the lower baffles, it seems reasonable to adopt the use of ports in the baffles for draining. This will require the use of the MathCAD code for draining tanks in series to choose the port size. See "Vertical Flocculator baffle port drain design.xmcd" in the sourceforge repository. Code and draw the flocculator drain system. This drain system should only be drawn when the flocculator baffles are rigid and the flocculator is vertical. We need a method to determine if the flocculator baffles are rigid. One possible method would be to set a criterion based on the thickness of the baffles that the client specified.

Sedimentation Inlet and Exit Tanks

• The tanks at the end of the inlet and exit channels that hold the pipes leading to the distribution tank and to waste no longer need to be as tall as they are currently drawn. Rather they can just cantilever off the side of the sedimentation tank.
• The PVC pipe component that is embedded in concrete needs to be drawn as a PVC coupling
• The removable pipes that are connected above the coupling should be long enough that they reach the same elevation as the top of the sedimentation tank walls (include freeboard).
• The waste pipe leaving the inlet tank should be drawn as a Tee that connects to the waste line from the exit tank.

Sedimentation Tank Control Pieces

• The pieces that allow the sedimentation tanks to be shut off need to be drawn (caps for the inlet channels, and exit channels). In addition, the caps should have a 1" PVC pipe that extends above the surface of the water in order to prevent air being trapped in the pipes delivering flocculated water to the sedimentation tank, and also to serve as a handle for removing the caps.
• Design and draw the caps for the effluent manifolds that are used for refilling the sedimentation tanks with clean water after the tank is cleaned. The caps for this purpose have an orifice that is sized to fill the sed tank in as short a period of time as possible without using all of the plant flow.

Plant drain system

Most of the plant waste flows enter an open channel that runs along the end of the tanks corresponding to the end with the sedimentation tank entrance channel. The following items must be designed and then drawn.

• Drain channel built of concrete
• Drain valves for all the tanks (many are already sized, but none are drawn)
• Concrete covers for the drain channel with access ports

Materials List

This list has been started and needs to be edited with input from the engineers in Honduras. Calculate the following

• Total wall area
• Total tank plan view areas
•  

Documentation videos

Create an AutoCAD video that will show the assembly of the plant with descriptions of each part. This video could serve as the documentation that gets sent to a client who designs a plant with the Download Design Tool (ADT) that explains the technology behind each piece. This video will also be used for training and teaching. Ideally the video should be made using a series of commands based on the plant dimensions so that the video can be created automatically for different plant designs.

Scale model

Create a scale model of an AguaClara plant that can be disassembled to show maintenance operations and to show the water path. This scale model should be should be small enough that it can be easily transported.

Research

Laminar Tube Floc

The development of FReTA and the data processing methods used to analyze its measurements has given AguaClara a powerful research tool. The investigation by Ian Tse into fluid shear influences on hydraulic flocculation was the first of many studies that could be performed with the Laminar Tube Floc/FReTA apparatus.

• Extend the experimental range evaluated by Ian Tse by increasing the length of the tube flocculator. One of the goals of this research is to measure the effect of collision potential on the residual turbidity. We have preliminary evidence that the residual turbidity continues to increase as the collision potential is increased. The challenge is to determine how long that trend continues. Increase the total flocculator length by a factor of 2 and then by a factor of 3.
• Investigate tapered flocculation designs. Hydraulic flocculators in AguaClara plants are currently designed such that the energy dissipation rates incrementally decrease over the length of the flocculator. Begin with an experimental setup that has two energy dissipation rates and determine if there is any benefit. The energy dissipation rate in the second section of the flocculator can be independently controlled by using a peristaltic pump to remove a fraction of the flow. Thus the same size tubing can be used for both sections of the flocculator.

• Investigate the influence of microscale mixing. For these tests use a well designed flocculator that produces very low residual turbidity. It is unclear what influence rapid mix parameters have on plant performance. Compare performance of systems with no rapid mix to systems that have a high energy dissipation rate (of at least 1 W/kg). Measure the effects of changing the energy dissipation rate (perhaps 0.1 W/kg to 10 W/kg) and the residence time (1 s with a single orifice mixer to 100 s with a long small diameter coil) in the rapid mix unit.

• Measure the potential impact of poor macroscale mixing by adding approximately 10% of the turbidity AFTER the rapid mix process.

• measure the effects of the influent synthetic water compositions (particle type, particle concentration, introduction of organic acids, pH, alkalinity, etc.). For example, performing experiments with different initial turbidities can provide insight into how particle concentration affects floc strength and turbidity removal efficiencies. Likewise, varying the pH of the influent may help elucidate changes in floc strength as a function of pH. It is possible that floc strength (as measured by floc size) is well correlated with optimal alum dose and pH.

• A laboratory scale hydraulic flocculator that operates under turbulent conditions that are relatively homogeneous and easy to characterize could go a long way into understanding turbulent flocculation. Comparison of residual turbidity and floc sedimentation velocity from turbulent and laminar flow flocculators could be used to validate flocculation models. Design a turbulent tube flow flocculator by using a larger diameter tube. Determine the required flow rates and assess the capabilities of the temperature controlled water source and the peristaltic pumps to deliver the required flows. Design an upgrade to the experimental apparatus to deliver the higher flow rate if needed.

Floc Blanket Research

Subteam Leader: Matt Hurst

Challenges

• Test effect of floc blanket with NOM

• Test effect of pH on floc blanket performance (if possible)

• Conduct tests with low turbidity and NOM

• Model particle removal mechanisms in the floc blanket

Bibliography

Deliverables

• Documented effects of organic interaction with clay and alum system

• Complete model of particle removal mechanisms in the floc blanket with the approval of Dr. Weber-Shirk and Prof Lion

• Draft of paper investigating effects of NOM, pH and low turbidity water on floc blanket performance

Plate Settler Spacing

Current Team Leader: Rachel Philipson

Number of team members needed: 3-4

Important team member skills:

• CEE 3310 or equivalent Fluid Dynamics course

New Challenges related to sedimentation tank geometry

These challenges have been added to this team because the apparatus can be easily modified to rapidly conduct these experiments.

Preliminary Challenge

• A hypothesis of the Floating Floc team is that air bubbles will come out of solution in supersaturated water. To confirm this, supersaturated water will be sent through the plate settler spacing apparatus. The goal is to observe visually and from performance data whether air bubbles are coming out of solution. It is possible in a subsequent experiment that we will test this hypothesis adding natural organic matter (NOM).

Challenges related to Plate Settler Failure

• Use the ramp function in process controller software to change the velocity gradient and flow rate through the tube settler to test for floc failure roll-up in the tube settler.
• Improve the Theoretical Analysis of the Velocity Gradient by including the Reynolds Number dependence on velocity and then solving for the sedimentation velocity. Use the velocity gradient to obtain an equation for velocity at the inner surface of a floc. Set those two equations for velocity equal to eliminate the sedimentation velocity (but keep the velocity term that is the average velocity in the tube) and then solve for the floc diameter as a function of the average flow velocity and the tube diameter.

Challenges related to raw water quality effects on floc characteristics

• Investigate the effect of influent parameters such as natural organic matter, pH, and alum dose on floc blanket and plate settler performance. Work with tube floc team to analyze and interpret these results.

Other potential pending challanges

Although we reliably build and operate floc blankets at laboratory scale, we have not been able to build a floc blanket in full scale AguaClara plants. We hypothesize that the geometry in the bottom of the sedimentation tank directly determines the feasibility of forming a floc blanket. We hypothesize that any surfaces with an angle that is shallower than the angle of repose will accumulate flocs until the flocs reach the angle of repose. The covers to the distribution tunnels in the sedimentation tank at Cuatro Comunidades were set at 60 degrees. It is quite likely that the angle of repose is actually steeper than 60 degrees. Also the small flat bottom at the center may provide a location for significant accumulation of flocs.
The task for this team is to quantify the angle of repose at the bottom of a floc blanket and to design a floc blanket reactor that has minimal or no lag time in building a floc blanket. Normally the lag time is caused by sedimentation of flocs until the angle of repose is reached.

• Remove the cone from the bottom of the sedimentation column and build a floc blanket.
• Build a floc blanket and measure the time required to build the floc blanket to 50 cm and the floc blanket growth rate.
• Measure the height of the deposited sludge and use that height to calculate the angle of repose
• Add a cone that has the same angle as the angle of repose to the bottom of the sedimentation tank
• Build a floc blanket and measure the time required to build the floc blanket to 50 cm and the floc blanket growth rate.

Chemical Dose Controller

Team Leader:

Number of team members needed: 4

Important team member skills:

• MathCAD
• Process Controller
• Fluid Dynamics

Challenges

Non-Linear Chemical Dose Controller

This task is the centerpiece of a grant that we received from the EPA P3 program. This team requires construction and testing of a full scale prototype. A team of students will be presenting the prototype and the AguaClara project on the National Mall in Washington D.C. near the end of the spring semester as part of a competition to receive phase II funding for $75,000. This team should review the proposals that won phase II funding last year to do our very best to produce a winning entry. This design is also needed asap for the water treatment plant that is being built for Agalteca.

• Write the algorithms for the microscale mixing orifice that connects the rapid mix chamber to the flocculator for optimum energy dissipation, head loss, and float sizing.

• Design and construct a full scale fully functional non-linear CDC. Prepare to go through several iterations of this design to come up with a elegant, easy to use, simple to fabricate. See the Rapid Mix for equations to size the orifice. See the Flow Control and Measurement lecture from CEE 4540 for background on the nonlinear dose controller including design of the dose scale.
• Conduct experiments to evaluate design and compare experimental results with theoretical analysis.
• Evaluate the optimal head loss used for plant flow measurement and its effect on the accuracy of the chemical dosing especially at low dosages.
• Determine if more than one stock concentration of alum is needed or if two different sized orifices could be used to accurately provide the full range of alum dosages.

For additional challenges, see the suggested challenges our team did not address this semester.

Floating Flocs

Team Leader: Tanya or anyone with previous knowledge of floating flocs problem

Number of team members needed 3-4 members

Important team member skills:

• Fluid mechanics
• Process Controller (can be trained)
• Microsoft Excel
• MathCAD

Challenges

Details of the experiments that have been planned can be found on the Floating Flocs Fall 2009 Tentative Experiments page.

• Possibly redesign the experimental setup to better simulate the water quality at actual AguaClara plants by adding a clay source or surfactant source.
• Confirm the cause of the floating floc problem by running experiments with the Plate Settler Team's apparatus. Two experiments have been planned in order to test whether the cause of the floating floc problem is bubbles adhering to flocs or whether bubbles form on flocs and lift them to the surface. It is also possible that both of these mechanisms are responsible for floating floc.

If the cause is confirmed to be bubble formation on or adherence to flocs due to supersaturated water the team's tentative experimental plans include:

• Run experiments using mild surfactants to witness their effects on bubble formation. Currently, the team is considering using a dilute soap solution as a surfactant. Details as to the concentration of soap solution have yet to be determined. This will likely be based on a conversion of organic matter in the water of AguaClara plants.
• Run experiments using hydrophobic surfaces with or without surfactants to see the effects on bubble formation. The details of these experiments have yet to be determined. Currently, literature searches are being performed to find cost-effective hydrophobic surfaces to test. It has been found that plant surfaces may exhibit hydrophobic properties due to their waxy coating.
• Developing a new mechanism to either remove gas, stop bubble formation, or stop floating flocs themselves. Currently, the team is looking at a mechanism that involves using lamella or something similar to trap air pockets in the water. These air pockets would be knocked around by eddies and may collect dissolved gas in the process.

One possibility to consider is whether flocs that are floating can be redirected or captured so that they are not in the effluent water. Since achieving gas removal has been proven to be difficult, putting some time into considering other solutions may be worthwhile.

If the cause of floating flocs is shown to be something other than supersaturated water, the team may still focus on stopping flocs from floating; however, other approaches may be developed that are specifically tailored to actual cause of the problem.

CFD Simulation

Dr. Bhaskaran serves as an advisor to this team.
Wenqi Yi (mailto: yiwenqi@gmail.com) can help with training new members

Team Leader:

Number of team members needed: 2~3

Important team member skills:

• Strong background in fluid mechanics
• Basic idea of flocculation mechanisms (see Flocculation)
• Basic idea of computational fluid dynamics or numerical methods
• Previous experience with FLUENT is great, but not required
• Basic programming skills, C is great but not required (sometimes we use C to write user-defined-functions)

Flocculator Challenges

The long-term goal of the CFD team is to characterize the collision potential of hydraulic flocculators, improve our understanding of hydraulic flocculator, and suggest design changes that would improve the performance of hydraulic flocculators. The CFD team also provides the data that is used for the design of the AguaClara hydraulic flocculators.

• Develop method to obtain results that have identical inlet and outlet conditions for the flow space between two baffles. Potential methods include
  • simulations using many baffles in series to reach uniform flow
  • use a user defined function to set the inlet to a single baffle equal to the exit of the baffle at the end of each simulation
  • Use a smaller number of baffles in series, but set the inlet energy dissipation rate equal to values based on a simple model to reduce the number of baffles required to reach uniform flow
• Create an algorithm to characterize the ratio of the maximum energy dissipation rate to the average energy dissipation rate. Monroe performed a very crude analysis of this using the color plots produced during the spring of 2009. We need to formalize this calculation because it is an important design parameter. We need this parameter as a function of the baffle H/S ratio and also as a function of Re.
• Repeat the collision potential experiments at different Re to see if the results are general or if the Reynolds number has a significant impact. If the Reynolds number has a significant effect, then develop a simple model that characterizes the Reynolds number influence
• Create appropriate series of graphs and figures for a journal article.
• Extend to 3D model:
  • Improve convergence, to which energy dissipation rate is very sensitive
  • Configuration of parallel computing for 3D simulations
  • Modify the mesh: check the regional convergence of the mesh, coarsen the mesh in some region
• Validation
  • Sensitivity analysis of other parameters when necessary
  • Find experimental data to validate CFD results

Channel and Port Design Challenges

The channels and ports that carry the flocculated water to the sedimentation tank need to be designed to have the same maximum energy dissipation rate as at the end of the flocculator.
*Create the geometry of the transition from the flocculator to the inlet channel of the sedimentation tanks.
*Determine the required flow area of the channel that produces the same energy dissipation rate as the baffles at the end of the flocculator. The AguaClara design team is currently using the equation

 \[\varepsilon  \cong \frac{1}{{20W}}\left( {\frac{V}{{K_{vc} }}} \right)^3 \]

where V is the mean velocity, W is the dimension of flow that the vena contracta narrows further,

\[K_{vc} \]

is the area ratio of the vena contracta (0.6). The coefficient of 20 is a rough approximation and needs to be determined using CFD. The energy dissipation is the maximum energy dissipation rate and it must be defined using a similar approach as will be used for the flocculator baffles. After this analysis is complete we can discuss the merit of doing additional analysis for the ports that carry water into the sedimentation tanks or the ports that release water from the sedimentation distribution tunnels. The expectation is that the equation above will predict the energy dissipation rate reasonably well for different geometries as long as the flow paths take a 90 degree bend.

PIV measurements in the Flocculator

Team Leader: Julia Schoen

Number of team members needed: 1

Important team member skills:

• CEE 3310, or equivalent Fluid Dynamics course
• Students must be comfortable with coding in MatLAB
• Preferably have taken CEE 437/637 or have familiarity with PIV

Challenges

Energy dissipation rates in AguaClara flocculators have been modeled by the CFD team using the software package FLUENT. The main goal of the team is to experimentally validate the CFD models. Particle image velocimetry (PIV) is technique to determine the velocity characteristics of the flow. A series of digital images are taken in rapid sequence in the flow. The positions of seed particles in this flow are used to determine the velocity at each point in the flow.

• Design and set up an appropriately scaled model flocculator.
  • The flocculator will mimic the design used in the CFD models and must work within the limitations of whatever flume is available for the experiment.
• Obtain high resolution velocity data at multiple places along the flocculator using PIV.
• Analyze data in Matlab
  • Images taken in during lab work will be used to characterize velocity and fluctuations in velocity in the flow. Energy dissipation can then be calculated using a structure function method or a spectral method. The two methods will be used and compared to the CFD data.

Outreach

Fundraising

Group Leader: N/A

Number of team members needed: * Minimum of 2-3

Important team member skills:

• Good written and oral skills
• Ability to work closely with P.R. when finding Fundraising Contacts
• Good analytical skills for choosing grants and reviewing previous budgets
• Adaptability to the different requests of different organizations when looking at Requests for Proposals (Grant Writing term meaning grant application)

Challenges

Agalteca plant

• Continue to raise funds for the Agalteca plant
• Send thank you letters to the recent donors who have supported the Agalteca project.

  Grant writing to Foundations

• Check out the Cornell�s Foundation website. Identify a list of foundations or companies that serve Latin America.
• Assess the feasibility and if there are no obstacles, write a proposal to http://nciia.org/grants/sustainablevision. The proposal is due by October 16, 2009. The final draft should be completed at least 2 weeks before the deadline.
• Finalize the Ford Foundation preproposal and send it to Monroe and then to Abby Westervelt (Director of Corporate and Foundation Relations, College of Enginenering).
• Submit the Ford Foundation Online Form

• Explore Conrad Hilton Foundation and Tinker Foundation
• Continue working on the Grant Text Modules Page.
• Update the Grant Short List as grants are submitted

  Corporate Sponsorships

• Create a sponsorship packet (see for example CUAUV or Cornell 100+ MPG )
• Use google group to request alumni for employer information.
• Update a list of AguaClara alumni and their employers alumni contact info
• Contact Matt Ulinski, Hansen Director of Instructional Labs, Mechanical and Aerospace Engineering for suggestions on running the business side of a student team project.

Public Relations

Group Leader: N/A

Number of team members needed: 4-6

Important team member skills:

• Adobe Photoshop to aid graphics design
• Microsoft Publisher to create brochures and newsletters
• Effective communication skills
• Leadership skills
• Knowledge of advertisement, business, media communications
• Creativity and ingenuity
• Spanish language proficiency is a plus

Challenges

• Get more team members involved with presentations
• Create a fact sheet for presenter. Include additional AguaClara info that is not already on the slides
• demo plant suite case maintenance
• Edit the Outreach Wiki to eliminate outdated information
• Update the brochure with new projects
• Keep the brochure rack outside the admission's office in Hollister supplied with brochures
• Write a fall edition of the newsletter and send to the AguaClara googlegroup before Thanksgiving break.
• Invite all of this semester's students from 2550/4550/4540/5051 to join the AguaClara Google group by sending them a Google group invitation.
• Submit team news updates to Cornell Chronicle, Engineering Information Update (mailto: engr_info_update@cornell.edu)
• Enforce the standardized Wiki format by constant updating
• Broaden the medium for pr. I.E. radio, newspapers, magazines, journals, pamphlets, brochures, banners, t-shirts, stickers, posters, etc.
• Establish a calendar of pr events in September that seeks to promote the project itself while incorporating pr events that surround fundraising, presentations, etc.
• Include students from business, entrepreneurial majors, marketing, communication, etc.