Engine Combustion Noise Analysis: Testing Methods, Standards, and Applications

Engine combustion noise analysis is a specialized NVH testing discipline that quantifies the structure-borne noise generated by in-cylinder pressure rise and transmitted through the engine structure to the surrounding environment. Accredited third-party laboratories perform combustion noise measurements according to recognized standards such as ISO 6798, ISO 9614, and GB/T 1859 to support powertrain development, acoustic refinement, and regulatory compliance. This article explains the generation mechanisms, testing scope, measurement methods, and practical applications of combustion noise analysis for gasoline, diesel, and alternative-fuel engines.

Technical Overview

Combustion noise is one of the dominant noise sources in reciprocating internal combustion engines. It originates from the rapid rise of in-cylinder pressure during the combustion process, particularly during the premixed burning phase in diesel engines or during knock events in spark-ignition engines. The sharp pressure excitation acts on the cylinder head, piston, liner, and crankcase, causing structural vibration that is subsequently radiated as airborne noise. Because the excitation travels through the engine structure before becoming audible noise, combustion noise is classified as a structure-borne noise source, and its severity depends both on the combustion process itself and on the acoustic transfer behavior of the engine structure.

The key metric used to evaluate combustion noise is the Combustion Noise Level (CNL), typically expressed in dB(A) and derived from the cylinder pressure spectrum. The difference between the Cylinder Pressure Level (CPL) and the resulting Combustion Noise Level is defined as the structural attenuation, which usually ranges from 40 to 60 dB in the frequency band between approximately 500 Hz and 5 kHz where combustion noise dominates. Diesel engines are generally more sensitive to combustion noise due to high rates of pressure rise, while modern gasoline direct-injection engines exhibit combustion-related noise concerns such as knock, rumble, and cold-start roughness. Factors influencing combustion noise include injection timing and pressure, fuel quality and cetane number, exhaust gas recirculation rates, charge temperature, compression ratio, combustion phasing, engine speed and load, and structural stiffness of the cylinder block and head.

Testing Samples

Compression-Ignition Diesel Engines

Passenger car diesel engines, light-duty commercial engines, and heavy-duty truck engines are the most common test subjects, as diesel combustion noise is a primary NVH concern. Both naturally aspirated and turbocharged configurations, including common-rail systems with multiple injection events, can be evaluated.

Spark-Ignition Gasoline Engines

Port fuel injection (PFI) and gasoline direct injection (GDI) engines are tested for knock intensity, combustion roughness, idle irregularity, and high-load combustion noise. Downsized, turbocharged, and Miller-cycle gasoline engines are frequent candidates for analysis.

Gaseous and Alternative Fuel Engines

Engines operating on compressed natural gas (CNG), liquefied natural gas (LNG), hydrogen blends, methanol, and other alternative fuels require combustion noise characterization because fuel properties significantly alter pressure rise rates and heat release patterns. Research engines operating advanced combustion modes such as HCCI, RCCI, and low-temperature combustion also fall within this scope.

Hybrid Powertrain Engines and Research Units

Dedicated hybrid engines designed for Atkinson-cycle operation, single-cylinder research engines, and fully instrumented multi-cylinder prototypes are tested during development programs. Complete engine assemblies on dynamometer stands, as well as isolated components such as cylinder heads, blocks, pistons, and injection system hardware, may be submitted for evaluation.

Testing Items

• Combustion Noise Level (CNL) measurement across the engine operating map
• Cylinder Pressure Level (CPL) spectral analysis in 1/3-octave and narrow bands
• Structural attenuation characterization between cylinder pressure and radiated noise
• Rate of pressure rise (dP/dθ) and heat release rate analysis
• Knock detection, knock intensity quantification, and knock frequency identification
• Cold-start, warm-up, and idle combustion noise evaluation
• Sound power level determination of the complete engine assembly
• Engine order analysis and frequency-domain noise source ranking
• Combustion stability assessment through coefficient of variation of IMEP (COV_IMEP)
• Comparative noise studies of fuels, injection strategies, and calibration parameters
• Premium versus regular fuel combustion noise differentiation
• Statistical separation of combustion noise from total engine noise

Testing Methods

In-Cylinder Pressure Indication

Piezoelectric cylinder pressure transducers are installed in the combustion chambers, either via adapted glow plugs, spark plugs, or dedicated drilling locations. Pressure signals are sampled with crank-angle resolution of 0.1 to 1.0 degree to capture the rapid pressure transients responsible for high-frequency noise. Thermodynamic post-processing yields indicated mean effective pressure, heat release rate, and pressure rise gradient, all of which correlate directly with noise generation.

Combustion Noise Meter Method

The measured cylinder pressure spectrum is converted into an estimated A-weighted combustion noise level using standardized attenuation reference curves. This method enables rapid, cycle-resolved evaluation of combustion excitation independent of the acoustic environment, making it suitable for dynamometer test cells and on-board diagnostics research.

Radiated Noise Measurement

Airborne noise radiated by the engine is measured according to ISO 6798 for reciprocating internal combustion engine noise measurement. Complementary sound power determinations follow ISO 3744 and ISO 3745 using the enveloping surface method, while ISO 9614-1 and ISO 9614-2 sound intensity techniques are applied in acoustically challenging environments. National standards such as GB/T 1859 and SAE J1074 provide equivalent measurement frameworks for engine sound level evaluation. Measurements are performed in hemi-anechoic or reverberant chambers under controlled speed and load conditions.

Structural Response and Transfer Path Analysis

Accelerometers mounted on the cylinder block, head, and main bearing areas capture surface vibration responses. Structural transfer functions between cylinder pressure excitation and surface velocity are derived to quantify attenuation characteristics. Coherence analysis between cylinder pressure and microphone signals is used to identify frequency regions dominated by combustion excitation.

Noise Source Separation Techniques

To isolate combustion noise from mechanical sources such as piston slap, timing chain, and injector ticking, laboratories apply motored-versus-fired comparisons, selective cylinder deactivation, statistical regression across operating points, and cylinder-by-cylinder coherence analysis. These techniques enable accurate source ranking and targeted refinement of the dominant contributor.

Applications

Combustion noise analysis serves multiple purposes across the engine development lifecycle. During the design phase, it guides the optimization of combustion bowl geometry, injection strategy, and calibration maps to balance thermal efficiency against acoustic comfort. During validation, it supports benchmarking against competitor engines, verification of noise compliance targets, and pre-certification screening before formal type-approval noise testing.

• Powertrain NVH development and target-setting for new engine platforms
• Combustion calibration optimization for injection timing, rail pressure, and pilot injection strategies
• Fuel quality studies, including cetane number sensitivity and gasoline octane rating effects
• Assessment of alternative fuels and advanced combustion concepts
• End-of-line quality control and noise-based production screening
• Warranty investigation and root-cause diagnosis of field noise complaints
• Academic and industrial research into low-noise combustion control strategies
• Regulatory compliance support and technical documentation for market access

FAQ

What is the difference between combustion noise and mechanical noise?

Combustion noise is excited by the rapid in-cylinder pressure rise during burning and is transmitted through the block and head as structure-borne vibration. Mechanical noise originates from impacts and relative motion of components such as pistons, bearings, chains, and injectors. Laboratories separate the two using motored testing, coherence analysis, and cylinder-by-cylinder isolation techniques.

Which engines need combustion noise analysis?

All reciprocating engines benefit from evaluation. Diesel engines are the most sensitive due to aggressive premixed combustion, GDI gasoline engines require knock and rumble assessment, and alternative-fuel engines need re-characterization whenever fuel properties change pressure rise behavior.

What standards govern combustion noise testing?

Measurement procedures typically reference ISO 6798 for engine noise measurement, ISO 3744 and ISO 3745 for sound power determination, ISO 9614-1 and ISO 9614-2 for sound intensity methods, together with GB/T 1859 and SAE J1074 as complementary national and industry standards.

How is combustion noise quantified from cylinder pressure data?

The cylinder pressure trace is converted into a frequency spectrum, weighted with reference attenuation curves, and integrated to yield a single-number Combustion Noise Level in dB(A). Comparing the cylinder pressure spectrum with the measured radiated noise reveals the structural attenuation of the engine.

Can testing be performed on a vehicle instead of an engine bench?

Yes. While engine dynamometer test cells provide the best control over speed, load, and temperature, measurements can also be conducted in chassis configurations or semi-anechoic vehicle environments when bench testing is not practical, using adapted instrumentation.

How long does a typical combustion noise analysis program take?

A focused single-condition assessment may be completed within a few days, while a full operating-map characterization with calibration sweeps and structural attenuation studies typically requires one to several weeks, depending on instrumentation complexity and the number of test variants.