Demo 1 — Initial Prototype

Overview

Our initial prototype demonstrates the core proof of concept for the adaptive water purifier. At this stage, we have a working turbidity sensor connected to the Arduino, a peristaltic pump that responds to control logic, an ultrasonic sensor circuit assembled for future water-level monitoring, and an initial 3-D printed tank / housing concept. This milestone shows that the design is progressing from concept into a functional prototype that targets the minimum design specifications.

The current prototype is centered around turbidity-based feedback control. If the water remains too turbid, the pump continues operating so water can be recirculated. Once turbidity decreases below a threshold, the system is intended to stop or redirect flow. We also identified an important hardware issue during testing: the current filter / mesh configuration is too large, so the next iteration will use a smaller mesh to better retain fine particulate matter and make the turbidity response more meaningful.

Prototype Functionality

Working: 3 Partial / Pending: 3 Not Built Yet: 3

1.Turbidity sensor — reads and logs values; calibration curve established for initial material (milk baseline).
2.Peristaltic pump — switches on/off on command to drive flow through the system.
3.Filter housing — 3-D printed enclosure fabricated and test-fitted.
4.Sensor calibration — needs full redo for diatomaceous earth (DE) suspension; milk-based curve no longer valid after material switch.
5.Filter cartridge / mesh size — the ordered unit is oversized for the housing, and the current filter / mesh configuration is too large for the particulate sizes being tested. A smaller replacement is needed so the system can seal properly and better capture fine suspended contaminants.
6.Serial monitor during pump operation — USB serial is unavailable while the pump is active due to a shared power / interrupt conflict, though the Arduino sketch continues to execute. A data-logging workaround is still needed.
7.Valve / flow-diversion system — directional valves to redirect flow once effluent is deemed clean are not yet installed.
8.Integrated single housing — pump, filter, and electronics are still separate; full enclosure and plumbing integration are pending.
9.pH sensor — final-output water quality sensor has not yet been sourced or wired.

Application of ChemE Principles

Solid–Liquid Separation (Filtration)

Suspended particulates such as diatomaceous earth, roughly 5–50 µm, are removed by depth or surface filtration. The pressure drop across the filter bed, described by Darcy’s law, governs flux. A clogged, oversized, or poorly fitted cartridge severely limits throughput and can create bypass flow.

Mass Transfer — Turbidity as Surrogate

Turbidity correlates with suspended-solids concentration through light scattering and attenuation. The turbidity sensor measures transmitted or scattered light intensity, and calibration maps voltage to a turbidity metric so the controller can infer water quality in real time.

Residence Time and Flow Rate

Effective clarification requires enough contact time inside the filter. Residence time is approximated by τ = V / Q, where the peristaltic pump sets Q. Filter volume, pressure differential, and flow resistance together determine whether residence time is long enough for target particle removal.

Feedback Control Loop

The control strategy reads turbidity, compares it to a setpoint, and uses that signal to decide whether water should continue recirculating or be allowed to move downstream. This is an on/off feedback loop analogous to a recycle process in chemical engineering.

Ultrasonic Level Detection

The HC-SR04 ultrasonic sensor is being integrated to measure the distance from the top of the final tank to the water surface. This will allow the system to detect when the final tank is full and notify the user to empty it and restart the purification cycle.

Current Performance Limitation The oversized filter cartridge means the housing does not seal properly, which can allow bypass flow around the media. In addition, the current mesh is too large for the particulates being tested. Together, these issues inflate apparent purification performance and make calibration and filtration data unreliable until the correctly sized cartridge and smaller filter mesh are used.

Test Material: Milk to Diatomaceous Earth

Why we switched Milk particle size is too small to meaningfully test the filter, and milk also goes bad over time. Diatomaceous earth produces a more reproducible inorganic particulate suspension with better-defined settling behavior, making it a more realistic and stable turbidity test material.

Settling test: DE suspensions were photographed to show visible differences in turbidity and settling over time.
Implication: A new turbidity calibration curve must be generated using DE standards before closed-loop testing can proceed.

Filter Design Update Testing made it clear that the current filter / mesh opening is too large. The next design iteration will use a smaller mesh so that finer particles are retained more effectively, the filtration path is more realistic, and turbidity readings better reflect actual purification progress.

Hardware and Prototype Images

Pump and Turbidity Sensor Circuit

This image shows the Arduino, peristaltic pump, tubing, and inline turbidity sensor assembly currently being used for prototype testing.

Initial Tank CAD Model

This image shows the current cylindrical tank design being used to begin planning the physical reservoir, tubing routing, and future sensor placement.

Ultrasonic Sensor Circuit

The HC-SR04 ultrasonic sensor is wired as a separate subsystem for future final-tank water level monitoring and user notification.

Circuit Integration and Initial Assembly

The current prototype includes a working turbidity-sensor-to-pump control loop. The turbidity sensor is housed in the black inline assembly attached to the tubing and connected to the Arduino. The pump is controlled through the Arduino using threshold-based logic, so when turbidity readings indicate that water is still contaminated, the pump remains on and continues recirculating fluid. When the reading drops below the chosen threshold, the pump can stop or the system can eventually redirect flow.

The ultrasonic sensor was wired separately using a standard HC-SR04 setup and has been tested as a standalone subsystem. In future iterations, it will be mounted above the final tank so the system can detect when the water level reaches the sensor threshold and tell the user to empty the tank and restart purification.

The 3-D printed tank prototype provides an early physical basis for integrating the pump, tubing, sensors, and future pinch valves into a single flow path. At this stage, full tank-to-sensor-to-pump integration is still being developed.

Testing + Initial Performance Data

The turbidity sensor is integrated with the pump through Arduino-based threshold control. When turbidity readings indicate that the water is still contaminated, the pump remains active to recirculate fluid through the system. When turbidity falls below a defined threshold, the pump is intended to shut off. This demonstrates a basic working feedback loop between sensing and actuation, even though full closed-loop filtration testing is still pending hardware updates.

Test	Method	Result	Notes
Sensor calibration (milk)	Known dilutions to voltage reading	Curve obtained	Deprecated; DE calibration required
Pump on/off operation	Manual activation via Arduino	Functional	Pump switches on/off only; no flow rate modulation
DE settling rate	Visual plus sensor at timed intervals	Qualitative	Quantitative values to be logged once calibration is redone
Influent vs. effluent turbidity	Sensor before/after filter	Pending	Blocked until correct filter cartridge and smaller mesh are installed
TDS reduction	TDS meter	Pending	Will be measured once the system is fully sealed and running

Project Management + Next Steps

Action items before Demo 2, in rough priority order.

Blocker Wait for correctly-sized filter cartridge — the system cannot be fully sealed or performance-tested until the replacement arrives. Test fit should be performed as soon as it is received.
Hardware Reduce filter mesh size — the current mesh is too large for the fine particulate being tested. A smaller mesh is needed to improve contaminant retention and make turbidity-based control more meaningful.
Sensor Redo turbidity calibration with diatomaceous earth — prepare DE standards at known concentrations, record sensor output, generate a new calibration curve, and update firmware constants.
Software Fix serial monitor conflict during pump operation — investigate shared-ground or power-rail noise causing USB disconnect. Consider SD-card or alternate logging methods so data is captured while the pump runs.
Hardware Design and install valve / flow-diversion system — add pinch, solenoid, or servo valves to redirect flow to either recirculation or clean output once water quality reaches the target threshold.
Integration Consolidate all components into one housing — route tubing, mount pump and electronics, seal filter cartridge, and perform leak testing before additional sensor trials.
Sensor Integrate ultrasonic sensor into tank — mount the HC-SR04 above the final tank and use it to detect when the tank is full so the user can be notified to empty it and restart the purifier.
Sensor Source and integrate pH sensor — select a compatible pH probe, wire it to the Arduino, and include it in final output monitoring.

6) Timeline + Project Management (Gantt + Milestones)

Preliminary Design (report + block diagram) Initial Design (market research + SWOT + slides) Component testing (sensor + mesh + pump) Demo 1 build + proof of concept Demo 2 integration (valves + control + logging) Demo 3 enclosure + robustness Final report + final demo prep