FAQ Clarification - AguaClara/SWOT GitHub Wiki

If the mechanized clarifier design is so inefficient, why is it so widely used?

Possibly a mix of marketing, lack of transparency in the water treatment sector, the use of foreign aid funds that require that a high percentage of the donated funds be spent on hardware from the donor country, and the lack of clearly superior water treatment technologies.

How do you deal with communities after failures?

AguaClara maintains long term connections with the communities with AguaClara water treatment plants. The failures that we have had to date have not prevented the facilities from providing water that met the national water quality standards. Thus when there was a failure where the technologies didn't work as well as we had expected, we continued to work on the technical challenge and in most cases the fix was low cost. Given that we don't have a donor base that is contributing to implementation of the technologies we have not been able to finance upgrades and instead we have had to rely on the ability of the communities to finance the upgrades. Some communities have placed a high value on providing the best service possible and those communities have consistently upgraded their plants as new technologies have been developed. Other communities have not taken this initiative and those communities have working plants that don't perform as well as the latest technologies. A prime example of this is that only one of the early communities has invested in adding StaRS filters to their treatment train.

How do you quantify floc dimensions and optimal structure?

We don't yet know how to do this. We do know that floc size increases with coagulant dose and decreases with shear.

Plate settlers

Why is the head loss through the plate settlers so small and how was this proven?
The head loss is so small because the velocity is so low and the flow passages (the spaces between the plates) are relatively large. We proved this by using a free body diagram and then calculating the fluid shear on the plates.

Why might there be nonuniform flow distribution between the plate settlers?

Any differences in piezometric head or inertia (caused by large scale flow circulation for example) will cause nonuniform flow distribution between the plate settlers.

What happens in the plate settlers?

The particles continue to be under the influence of gravity and they continue to have a terminal velocity downward. That small velocity causes some of the particles to reach the plate that is below them. Particles accumulate on the plate and then eventually an avalanche forms and the particles slide back down into the floc filter.

How do you optimize the plate settler design?

We don't yet have enough understanding of the processes to optimize the plate settler design. At a minimum we need to have a better understanding of the controlling particle aggregation mechanisms in the floc filter and an estimate of the maximum stable upflow velocity in the sedimentation tank before we can begin to optimize the design of the plate settlers.

Since plate settlers reduce sed tank area, is there a way this saved space can be used increase the flow capacity?

Puzzle: the velocity relationship in the plate settlers

The plate settlers have complex flow because there is an entrance region where the flow profile is being established. The plates are generally long enough that the parabolic laminar flow velocity profile is established (assuming there aren't any density currents). We used that to estimate the head loss in the plate settlers. The actual head loss is slightly higher than our estimate because the entrance region has higher velocity gradients at the plate surfaces.

Diffusers

What is the role of the diffusers? Why do the diffuser pipes have to be rectangular shaped?
The goal is to create a continuous jet that extends from one end of the sed tank to the other. The diffuser rectangles are right next to each other and thus almost form a continuous jet. When the rectangular diffuser jets flow through the jet reverser the jets merge into a single continuous line jet.

How are the spacings between diffusers designed?

The spacings are set to be just a tiny bit larger than the actual length required to ensure that there aren't problems fitting the diffusers into the spaces during fabrication.

Is the rectangular shape diffuser common in the design of sedimentation tanks?

We invented the rectangular diffuser shape, the line jet, and the jet reverser geometry for stable operation of the floc filter. We don't know if this geometry has ever been used by others.

How do you calculate the diffuser velocity?

Each diffuser provides water to a length of the sedimentation tank that corresponds to the distance between centers of the diffusers. The flow rate through that area of the sed tank must equal the flow rate through the diffuser. (V_ff_up * B_diffuser * W_clarifier = V_diffuser * S_diffuser * W_Diffuser).

Do the diffusers get clogged?

Diffusers do get clogged if particles larger than W_diffuser are allowed to pass through the screen in the entrance tank. Thus the entrance tank screen must have smaller openings than W_diffuser. Yes, we learned this by making the mistake.

Floc Filter

How does the particle size distribution in a floc filter cause such a dramatic shift in turbidity removal? How does the floc filter reduce the settled water turbidity?
If the flocculator effluent contains many flocs and particles that have sedimentation velocities that are lower than the capture velocity, then those particles would not be captured by the sed tank. If those particles have additional opportunities for collisions and aggregate while in the floc filter, then there can be a dramatic reduction in the number of particles that escape the plate settlers. We don't yet understand the floc filter mechanisms well enough to have a descriptive mathematical model of the process. That means we have much to learn yet!

What is the measure of the floc filter efficiency?

The most useful parameter might be how much does the floc filter reduce the settled water turbidity.

Floc hopper

What determines the height of the weir for the floc hopper?
Presumably this would be set by some optimal design for the depth of the floc filter. This would be based on the economic analysis of the cost of making deeper sedimentation tanks vs the value of the improved performance or the value of being able to use less coagulant. We don't yet have enough understanding of the floc filter to optimize this.

How do the flocs flow into the floc hopper?

The fluidized suspension of flocs flows over the weir and drops into the floc hopper. This is a density difference driven flow that is just like the flow of water over a weir. The velocity of the floc suspension is low because the density difference between the fluidized floc suspension and the overlying water is small.

What does the floc hopper do and how does it function?

The floc hopper provides a place for the floc suspension to consolidate and thus it reduces the volume of water that must be wasted. The flocs have a sedimentation velocity equal to the upflow velocity in the floc filter and thus that velocity sets the rate of consolidation. We don't yet have design equations for the floc hopper because we don't have data on the floc filter concentration in plants operating with real surface water. We expect the floc filter concentration to be reduced when the dissolved organic concentration is higher. A lower concentration floc filter will require a larger horizontal surface area for consolidation. It is possible that the floc hopper would benefit from having a set of plate settlers installed in the floc hopper to provide more horizontal surface area for consolidation. We don't have good information on failure modes for floc hoppers. Clearly the floc hopper must have steeply sloped bottom so that all of the sludge is directed toward the drain. At very high influent turbidities the floc hopper could fail by not being able to consolidate the flocs fast enough and this would require wasting a more dilute floc suspension at a higher rate to prevent the floc filter from growing up into the plate settlers.

Clarifier

How to make sure the velocities are above 0.15 m/s to deliver flocs to the sed tank? Q=V*A. Set A to be small enough. In the sedimentation tank inlet channel we use an upward sloped floor to keep the velocity closer to constant and thus minimize floc sedimentation. This design objective is incompatible with designing the sed tank inlet channel to be able to handle a water treatment plant expansion to handle increased demand in the future.

How can we get perfectly even flow distribution?

"Perfectly" is a lofty goal. We attempt to get flow distributions that are close enough to uniform so that we don't expect to see significant deterioration in performance. We are guessing at these constraints.

How often are clarifiers cleaned?

Moroceli plant operators clean the sed tank every 3 months. They clean the sed tank because the plate settlers become increasingly filled with flocs and perhaps bacteria growth. This would be a good use of a pressure washer to quickly clean the plate settlers in situ.

What is the timeline for the clarifier design improvements since the first design?

We have been working on improvements to the clarifiers from the very beginning and it continues to be priority research for us because the clarifiers are the largest unit process and remove the vast majority of the contaminants.

In what direction do you see the clarifier improving in the coming years?

I expect that we will learn how to increase the upflow velocity and how to optimize performance of the floc filter. Once we've optimized the floc filter, then perhaps we can move on to optimizing the design of the plate settlers.

What constrains the width of the sed tank?

We are currently using the width of the plastic sheeting that we use to build the plate settlers. We've built one sed tank that was roughly 2 m wide and used two sets of plate settlers. That works, but the plant operators find it a little more difficult to disassemble that clarifier because they use the dividing walls to walk on. Another constraint is the width of the clarifier bay that is served by one inlet manifold. As the width increases the depth of the triangular bottom section of the clarifier bay increases and the amount of fill required to build the sloped sections increases. There are clearly many different geometric options here and more exploration is needed especially as we consider building larger clarifiers for cities.

Why will flow circulation cause poor performance?

If all of the flow goes through 10% of the plate settlers, then the capture velocity will be 10 times larger than the design.

Is there ever ever a situation where it would be advantageous to have a clarifier with a horizontal bottom?

A flat bottom will have sludge accumulation, will not have a stable floc filter, and will not perform as well. The only advantage to the flat bottom is that it is cheaper to construct.

What are we the least happy about in the clarifier?

The poor performance during positive temperature gradients and the relatively large footprint required for high performance are the two issues that we would like to improve.

Where do we have the most obvious room for improvement?

The floc filter provides the most obvious room for improvement because we don't yet understand what controls the floc filter performance.

Still a little confused about the geometry of the clarifier.

Check out the Onshape model of an AguaClara clarifier.

How much of the flocs are removed by the clarifier?

Clarification reduces the turbidity to approximately 1 NTU.

What slope is required in the bottom of the clarifier to prevent solids from accumulating?

We've been using 50 degrees. There isn't a single right answer here because the required steepness increases with coagulant dose.

If you have multiple clarifier bays, why not just use several inlets in one tank (and eliminate the dividing walls)?

This is a good idea. We've done this at San Nicolas. The disadvantages are that it is then a bigger tank to take off line for maintenance and it is more difficult to install the plate settlers. Those disadvantages might not be significant if designing a water treatment plant that provides hundreds of liters per second.

What is the path of flocs through the different parts of the clarifier?

Big flocs hang out in the floc filter and at some point they fall over the floc weir. Little flocs rise through the floc filter with the water, enter the plate settlers and settle under the influence of gravity to the lower plate. Primary particles and small flocs are carried up through the plate settlers and out of the sed tank toward the filters. The flocs that fall on the plate settler eventually slide back down into the floc filter. The flocs aggregate as they slide down the plates. It is possible that some big flocs hang out in the bottom of the floc filter and never exit. If that is happening, then there is a need to occasionally dump water from the bottom of the floc filter. This is another area requiring more research.