Home | Contents | 日本語

CHAPTER 4

Measuring Breathing and Leaks

A PAPR for the general public must do more than supply air: the device itself needs to know whether it is worn correctly and whether it is leaking. We developed ways to estimate this from a differential-pressure sensor and the pump control signal alone, without an expensive flow meter.

Estimating respiratory states and controlling pressure

Estimation of Respiratory States Based on a Measurement Model of Airflow Characteristics in PAPR Using Differential Pressure and PWM Control Signals: In the Development of a Public-Oriented PAPR as an Alternative to Lockdown Measures
Y. Fujii, A. Takita, S. Hashimoto, K. Amagai — Sensors, Vol.25, No.9, 2939, 2025.
https://doi.org/10.3390/s25092939

Role: Respiratory-state estimation

Models the flow characteristics of a low-cost PAPR supply unit (nonwoven/HEPA filter plus pump) and its exhaust filter, and estimates the breathing state from differential pressure and the PWM control signal. Validated with a manually driven piston breathing simulator.

  • Supply flow is modelled as a function of differential pressure and PWM duty; exhaust flow as a function of differential pressure; respiratory flow is estimated as their difference.
  • Effect of pressure control: adjusting the duty value reduced the standard deviation of differential pressure from 9.0 Pa to 4.9 Pa (about 45%). The correlation (R²) between breathing and pressure fluctuation fell from 0.86 to 0.22.
  • Towards comfort: outlines pressure control that depressurises during exhalation and pressurises during inhalation.

Components: Sensirion SDP810 differential-pressure sensor; ESP32 microcontroller.

Real-time flow and leak estimation with one sensor

The Real-Time Estimation of Respiratory Flow and Mask Leakage in a PAPR Using a Single Differential-Pressure Sensor and Microcontroller-Based Smartphone Interface in the Development of a Public-Oriented Powered Air-Purifying Respirator as an Alternative to Lockdown Measures
Y. Fujii — Sensors, Vol.25, No.17, 5340, 2025.
https://doi.org/10.3390/s25175340

Role: Real-time estimation of respiratory flow and leakage

Estimates respiratory flow and detects mask leakage with a single differential-pressure sensor and a microcontroller, and shows the result on a smartphone.

  • Only two inputs: differential pressure and battery voltage. Microcontroller: Arduino UNO R4 WiFi.
  • Accuracy: against a reference flow meter, R² = 0.952 and mean absolute error 15.9 L/min.
  • Leak detection: with a deliberate gap, a leak of about 4.6 L/min (about 1.9% of supply flow) was estimated and the app changed from "no leak" to "leak".
  • Smartphone link: Bluetooth Low Energy to an Android app. White = not worn, blue = no leak, red = leak.
  • Prototype: about 656 g; about 5 h 20 min of continuous operation.

Validation used a mannequin and a manually driven breathing simulator; tests with people are still to come. The decision thresholds are provisional.

← Previous: 3. Feasibility and Requirements | Next: 5. Time-Managed Use →

Contents

Measures for the Next Pandemic: engineering and social infrastructure for controlling infection without stopping society | Yusaku Fujii (Gunma University)
This site presents research-stage work; figures are model calculations or prototype measurements.