The Ultimate Technical Guide to Industrial Acoustic Cameras: Principles, Parameters, and Applications

In the world of predictive maintenance, the ability to "see" sound has revolutionized how industrial facilities detect faults. Acoustic cameras transform invisible ultrasonic emissions into visual images, enabling engineers to pinpoint partial discharge, gas leaks, and mechanical anomalies with unprecedented precision.

But not all acoustic cameras are created equal. Understanding the technology behind them—from microphone array design to frequency range and AI-powered diagnostics—is essential for making informed purchasing decisions.

This technical guide draws from Hertzinno's extensive training materials to provide a comprehensive overview of acoustic camera technology, evaluation criteria, and application-specific considerations.

Part I: What is an Acoustic Camera?

An acoustic camera is an imaging device used to locate sound sources and characterize them. Unlike traditional microphones that merely measure sound pressure levels, acoustic cameras create a visual representation of sound, overlaying color-coded intensity maps onto optical images.

Core Components

Every acoustic camera consists of four fundamental elements:

The Physics Behind Acoustic Imaging

The relationship between frequency and wavelength is fundamental to acoustic camera design:

λ = v / f

Where:

• λ = wavelength (meters)

• v = speed of sound in air (≈343 m/s at room temperature)

• f = frequency (Hz)

For example, at 1000 Hz:
λ = 343 / 1000 = 0.343 meters

This relationship determines the physical constraints of microphone arrays. Lower frequencies have longer wavelengths and require larger array diameters to achieve accurate localization. Conversely, higher frequencies with shorter wavelengths can be detected by more compact arrays.

This same principle applies to sonar technology—a hull-mounted sonar operating in water (where sound speed is approximately 1500 m/s) requires different array dimensions than an air-coupled acoustic camera.

Key insight: With a typical array diameter of 165mm, it is physically impossible to accurately measure frequencies below approximately 1 kHz due to wavelength constraints.

Part II: How to Evaluate an Acoustic Camera

When comparing acoustic cameras, several critical parameters determine real-world performance.

1. Microphone Number: More is Not Always Better

While it's intuitive to assume that more microphones equal better performance, the reality is more nuanced.

Theoretical Gain:

• Doubling the number of array elements increases array gain by 3 dB

• Example: Increasing from 64 to 128 elements yields +3 dB gain

• Increasing from 128 to 256 elements yields another +3 dB gain

Practical Limitations:

• Single MEMS microphones generate self-noise that affects weak signal measurement

• Increasing microphone count improves SNR and reduces self-noise by 3 dB per doubling

• However, with a fixed array diameter, increasing element density increases mutual coupling between elements

• Excessive mutual coupling introduces interference that reduces array gain

Lateral Spatial Resolution:
This critical parameter measures the minimum distance between two sound sources that can be resolved perpendicular to the camera axis. Smaller values indicate better performance—the ability to distinguish between closely spaced sources.

The Bottom Line: Optimal microphone count balances gain improvement against mutual coupling interference. Hertzinno's engineering team carefully optimizes array geometry for each application scenario.

2. Frequency Range: The Most Important Feature

Frequency range fundamentally determines what an acoustic camera can detect.

Why Frequency Matters:

Different industrial faults emit energy in characteristic frequency bands:

The 20 kHz – 40 kHz Sweet Spot:

Factory environments are acoustically chaotic. Most machinery noise peaks below 10 kHz and diminishes beyond 60 kHz. Gas leaks and partial discharges, however, manifest most clearly in the 20 kHz to 40 kHz range—above most background noise but low enough to travel reasonable distances.

Distance Considerations:

Higher frequency waves are absorbed more rapidly by air, limiting their travel distance. For:

• Long-distance detection (outdoor substations, pipelines): Lower frequencies (10–30 kHz) travel farther

• Close-range inspection (compressed air systems, vacuum leaks): Higher frequencies (up to 65 kHz) provide better sensitivity

The Ideal Acoustic Camera should offer adjustable frequency ranges, allowing operators to tune the device for specific tasks.

3. Signal Characteristics: Understanding What You're Detecting

Both gas leaks and partial discharges generate broadband signals—energy spread across a wide frequency spectrum. However, they have distinct temporal characteristics:

This distinction is crucial for accurate diagnosis.

Part III: Applications of Acoustic Cameras

Industrial acoustic cameras excel in five primary applications:

1. Partial Discharge Detection

Challenge: High-voltage equipment insulation degradation emits ultrasonic energy, but not all ultrasonic signals indicate dangerous discharge.

The PRPD Advantage:

Frequency analysis tells you it "might" be discharge. Phase-Resolved Partial Discharge (PRPD) analysis tells you it "definitely" is discharge.

PRPD displays acoustic activity synchronized with the 50/60 Hz power frequency, revealing characteristic patterns for:

• Corona discharge

• Floating discharge

• Surface discharge

Hertzinno's AI Advantage: While many acoustic cameras can display PRPD patterns, Hertzinno's AI algorithms automatically classify PRPD types on-screen, helping engineers make maintenance decisions immediately—not after hours of data analysis.

2. Gas Leak Detection

Challenge: Locating pressurized gas leaks in noisy industrial environments.

Solution: Acoustic cameras filter out background noise and visualize the ultrasonic signature of escaping gas, even for:

• Natural gas (methane)

• Hydrogen

• Nitrogen

• Compressed air

3. Mechanical Fault Diagnosis

Rotating machinery emits characteristic acoustic signatures. Acoustic imaging detects:

• Bearing wear

• Gear misalignment

• Lubrication failure

• Cavitation in pumps

• Valve leakage

4. Vacuum Leak Detection

Even vacuum systems—which draw air inward rather than expelling it—generate detectable ultrasonic signatures at leak points.

5. High-Speed and Low-Speed Rotating Equipment

Acoustic cameras visualize noise from both:

• High-speed rotating machinery (turbines, compressors)

• Low-speed rotating equipment (large bearings, slow rollers)

Part IV: How to Choose an Acoustic Camera

Beyond specifications, several practical considerations influence real-world usability.

1. Thermal Management

Acoustic cameras perform massive parallel computations on streaming microphone data, generating significant heat. Poor thermal design leads to:

• Device overheating

• Operator discomfort or burns (casing temperature >60°C)

• Performance throttling

• Reduced component lifespan

What to look for: Advanced hardware design that balances processing power with thermal efficiency. Compare not just specifications, but ergonomics and operating temperature during extended use.

2. Application-Specific Optimization

No single acoustic camera excels at every application. The best choice depends on your primary use case.

3. Total Cost of Ownership

Consider:

• Battery life (impacting field productivity)

• Weight and ergonomics (affecting operator fatigue)

• Durability and IP rating (determining longevity in harsh environments)

• Software capabilities (AI analysis reduces interpretation time)

Part V: The Hertzinno Advantage

Hertzinno acoustic cameras are engineered with five core advantages:

1. High Performance

Optimized microphone array geometry and signal processing deliver exceptional sound source localization accuracy.

2. Wide Detection Bandwidth

Coverage from audible frequencies through 65 kHz ultrasonic enables detection of diverse fault types with a single device.

3. AI-Powered Diagnostics

Trained on millions of industrial acoustic samples, Hertzinno's algorithms:

• Automatically classify partial discharge types via PRPD analysis

• Distinguish between gas leaks and mechanical noise

• Provide real-time decision support on-screen

4. Multi-Scenario Product Portfolio

5. Multi-Sensor Fusion

Hertzinno integrates acoustic imaging with:

• Infrared thermography (640×512 resolution) for comprehensive fault diagnosis—acoustic for partial discharge, thermal for overheating

• Methane laser sensing for quantifying gas concentration

• PRPD analysis with AI classification for definitive partial discharge identification