logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-06302023-165635


Tipo di tesi
Tesi di laurea magistrale
Autore
LIPPI, ALESSIO
URN
etd-06302023-165635
Titolo
Knock detection and characterization in spark-ignition engines
Dipartimento
INGEGNERIA CIVILE E INDUSTRIALE
Corso di studi
INGEGNERIA DEI VEICOLI
Relatori
relatore Prof. Frigo, Stefano
relatore Prof. Ghandhi, Jaal
Parole chiave
  • ethanol
  • 4-stroke
  • detonation
  • knocking
  • spark-ignition
  • engine
  • knock
Data inizio appello
17/07/2023
Consultabilità
Non consultabile
Data di rilascio
17/07/2026
Riassunto
This study stems from the need to increase knowledge regarding the phenomenon of knocking, which has always limited the development of internal combustion engines, since the increase in volumetric compression ratio, the main parameter on which efficiency depends, is limited by the occurrence of this complex phenomenon.

The following thesis work was carried out at the Engine Research Center at the University of Wisconsin-Madison and begins with experimental data directly collected on a Kohler Aegis LH775, a spark-ignition engine with 90° V 2-cylinder configuration.

This engine was run on different fuels, specifically the percentages of ethanol within the gasoline were varied, and it was driven from incipient knock to moderate knock by varying the spark advance angle, all this under two different engine load conditions; a minimum of 10000 cycles was collected for each operating condition.

In-chamber pressure data were collected simultaneously by using two piezoelectric pressure transducers of different diameters, thus with different natural frequencies, mounted very close together to prevent different cavity resonance phenomena from altering the result. In order to investigate the nature of knocking combustion highly time-resolved measurements were performed. In addition, all parameters necessary to characterize the operation of the engine were collected.

All analysis and post-processing of the data was programmed using the Python language. The two pressure signals were processed both in the crank angle domain, to characterize the main thermodynamic quantities and compare the consistency of the data collected by the two sensors, and in the time domain for all analyses inherent to knock.

A new knock intensity index (Ring-down magnitude), based on the attenuation of pressure waves due to the knock phenomenon, was also developed and compared with the time-domain pressure-based method to characterize knock intensity, called Maximum Amplitude Of Pressure Oscillations (MAPO) as well as comparing the results obtained for both sensors.

Finally, the collected data were used to compare the mass of unburned fuel present in the combustion chamber at the time of knock onset, estimated by the method presented by A. J. Shahlari and J. B. Ghandhi in "Pressure-Based Knock Measurement Issues".
File