Finding reason for increase in pressure in flocculator.

[Roswell-Lo]

Today we started an experiment running only water through the flocculator/sed tank.

Simple control setup to see if there is an increase in pressure throughout this trial. If there isn't we can do a trial using only clay or only coagulant.

[MehrinSelimgir]

The above data indicates a small, gradual increase in pressure with just running water through.

[MehrinSelimgir]

Today (2/27) at 4:30pm we begun a trail to analyze pressure difference when adding in just coagulant. Our next experiment will be to run an experiment with clay & coagulant and run at full pump speed for 30 seconds to see if we can diminish pressure difference.

[Kanha-Matai]

After conducting a trial with just water with coagulant, we've found even higher rate of head loss. After running for 3 hours, our ProCoDa had reset leaving us with just 3 hours of data. However, the head loss is far more significant--20cm in 3 hours with just coagulant versus 40cm in 13 hours with both coagulant and clay.

[monroews]

This is so exciting! This suggests that my hypothesis that clay was attaching to the walls and causing the head loss is incorrect. Instead this suggests that coagulant nanoparticles are responsible for the head loss. The reason that feeding coagulant without clay results in higher head loss is because none of the coagulant was able to attach to clay and thus more attached to the walls.

• Create a simple head loss model based on the assumption that each coagulant nanoparticle is a little sphere experiencing fluid drag that can be calculated based on the velocity at the center of the sphere.
  • Estimate the number of coagulant nanoparticles per tubing area that would be required to achieve the observed head loss.
  • Then calculate the fraction of the coagulant nanoparticles that are attaching to the tubing walls rather than staying in suspension
• Begin exploring what fluid velocity or shear is required to remove these nanoparticles.
  • Try high speed pumping with a peristaltic pump
  • If that doesn't dislodge the particles, then try a high speed syringe injection.

[Kanha-Matai]

After conducting that experiment, out of curiosity we decided to re-examine the hydrophobic tubing to see whether there was a big difference. Below is the overall data we got out of the trial- there was no issue with faulty readings after midnight allowing us to get a conclusive trial. We have also fixed our sensor to obtain positive values as opposed to negative values for pressure.

Looking at the data, we can see a continuous build-up of pressure. To fairly compare the results from the hydrophobic test and the regular tubing, below I have attached the graphs for the first three hours of both experiments as we only got a 3-hour long trial for regular tubing. I have matched the axes to easily notice the difference between the two. From the experiment, the headloss with the change with hydrophobic tubing was almost half the change with regular tubing (15cm versus 25cm).

Additionally, another very interesting observation from this trial is the sudden drop in pressure around the 12th hour- this is when the coagulant ran out. This potentially proves that running water for a short amount of time without coagulant can actually clear out the coagulant. Below I have attached a more focused graph at the 12th hour (note: the x-axis has been changed to minutes to get a more precise and understandable reading). This would be extremely beneficial as we were able to lose almost all of the built up pressure within 2 minutes, ramping up water flow to 4x that of the trial could mean removing all of the built-up pressure in 30 seconds.

[monroews]

This is awesome! These are great results. @WPennock The rapid drop in head loss when the coagulant stopped was likely caused by the air bubbles that were injected into the flocculator when the coagulant ran out. These air bubbles cause a dramatic increase in shear because they force the water to flow like a plug rather than with a parabolic velocity profile. It is possible that all that is required to strip the coagulant nanoparticles off of the tubing wall is a single air bubble.

Note that the head loss dropped off in a time that was comparable to the residence time in the flocculator. What exactly was the hydraulic residence time for your flocculator and how does that compare with the graph above?

This suggest that one option is to inject a bubble of air on a regular basis. Air bubbles are terrible in floc blankets and tube settlers, so the air bubbles would need to be removed at the end of the flocculator. A peristaltic pump could cycle air through a flocculator by injecting air at the front of the flocculator and capturing the air bubbles at the downstream end of the flocculator.

The other option is a short burst of high velocity water. We need the data on the shear required to move the nanoparticles.

[WPennock]

I love serendipitous findings! This is so cool! I hope to hear more about how this progresses!!