METHOD FOR CHARACTERISING THE KNOCK-RESISTANCE OF FUELS

Abstract

The invention relates to a method for characterising the knock-resistance of a fuel using a test engine, wherein the curve against time of the cylinder pressure of the test engine during combustion of the fuel in the test engine is determined and the determined pressure signals are compared with the corresponding pressure signals of a least one standard fuel of known knock-resistance.

Description

The invention refers to a method for characterizing the knock resistance of fuels with the help of a test engine.

Standard DIN EN 228 stipulates the characteristic values and properties for non-leaded types of gasoline as minimum requirements. Table 1 shows an extract of the essential characteristic values for fuel.

	Requirements according to DIN EN 228
Characteristic value	Unit	SuperPlus	Super	Normal
Density at 15° C.	Kg/m3		720-775
Knock resistance
R.O.N.		min. 98	min. 95	min. 91
M.O.N.		min. 88	min. 85	min. 82.5
Lead content	mg/L		max. 5
Distillation range *	% (V/V)
Evaporated quantity (Class A)
at 70° C., E70		20-48
at 100° C., E100		46-71
at 150° C., E150		min. 75
Evaporated quantity (Class D/D1)
at 70° C., E70		22-50
at 100° C., E100		46-71
at 150° C., E150		min. 75
Final boiling point (FBP) (Class A/D/D1)	° C.	max. 210
Volatility indicator VLI **
(VLI = 10 × VP + 7/E70)
Class D1	Index	max. 1150
Distillation residue	% (V/V)	max. 2
Vapor pressure (DVPE)	kPa
Class A		45.0-60.0
Class D/D1		60.0-90.0
Evaporation residue	mg/100 mL	max. 5
Benzene content	% (V/V)	max. 1
Sulphur content	mg/kg	max. 150
Oxidation stability	min	max. 360
Copper corrosion	Extent of	max. 1
	corrosion
* Class A: May 1-September 30 (summer)
Class D: November 16-March 15 (winter)
Class D1: March 16-April 30 & October 1-November 15 (transition)
** Vapor Lock Index

Especially important in all of this is knock resistance, which is described with two characteristic numbers, the motor octane number (M.O.N.) and the research octane number (R.O.N.). Briefly explained, it can be said that knocking combustion can occur in any gasoline engine and cause extensive engine damage if intensive enough. For this reason, engine developers are required to prevent the non-uniform combustion occurrence of the knocking—in other words, to integrate knocking control systems in engine controls and to constructively prepare the engine for the fuel's knock resistance. High octane numbers permit higher performance with the simultaneous higher degree of effectiveness of the engine and therefore lower consumption. For these reasons, higher prices can also be fetched with higher-octane fuels even tough they have almost the same energy content (fuel value).

The worldwide determination of octane numbers is nowadays carried out empirically and according to standardized processes in the fuel producers' laboratories. Special one-cylinder test engines with a compression ratio as variable that can be adjusted to the respective fuel quality are used for this purpose. The objective is to compare the knocking intensity of the fuel to be tested with fuels of known octane number and to determine its number, if needed, by interpolating the octane numbers. The standard arbitrarily assigned octane number for isooctane is 100 and for n-heptane is 0. By mixing these components, a fuel can be produced that will have the same knocking intensity as the fuel to be tested. The octane number that is searched for will then correspond to the volumetric share of isooctane in the fuel mixture. Testing conditions differentiate between M.O.N. and R.O.N. if all other process steps agree and the same measuring technique and test engine were used.

The degree of knocking intensity is generated with an electric sensor (electronic detonation meter) screwed into the engine's combustion chamber (FIG. 1) and an indicator (knock meter) displays the result. The knocking intensity and frequency calculations are not performed with the measuring data, however. Our own research has shown us that fuels with equal octane numbers can actually have a different knocking behavior regarding intensity and frequency. If this method is professionally applied with the corresponding experience, an accuracy of no more than +0.2 octane numbers can be achieved. The operation is done manually and takes between 20 and 30 minutes per octane number, There have been—and still are—many attempts to determine the octane number with calculations or another instrument—i.e., outside the engine. Unfortunately, so far it has not been possible to achieve this with the desired accuracy. If the fuel composition is known, a gas chromatography analysis or infrared spectroscopy can give reasonably accurate results, but these methods are only used in refineries because they know exactly the composition of their fuel. A fundamental improvement of the device and method has not been found.

The following problems have been detected in assessing the valid processes for determining the octane number;

• • The accuracy of the process can be improved upon
  • The process is time-consuming; it cannot be automated and allows no online indication.
  • In the process, the knocking intensity is indicated directly after the analog processing of the measuring signal. No exact calculation of the knocking intensity and frequency takes place.
  • Whether the determined octane numbers actually reflect the knocking resistance of the fuels that current combustion engines require is called into question.

It is therefore the task of the invention to suggest a process that will make a fast and reliable characterization of the knocking resistance of fuels possible.

The task is solved by determining the chronological sequence of the testing engine's cylinder pressure during the combustion of the fuel in the test engine so the determined pressure signals can be compared with the corresponding pressure signals of at least one standard fuel of known knocking resistance.

The utilization of modern measuring and analytical techniques allows the exact characterization of the various parameters of the knocking combustion processes that occur in the test engine, such as knocking intensity, knocking pressure amplitude of the peak pressure, speed of pressure increase, frequency and the frequency distributions of these parameters. The approach taken and the necessary measuring techniques are described below.

In this case, a piezoelectric pressure sensor (as typically used by engine developers in their R&D tasks) is advantageously built in instead of the electronic detonation meter. Thus, the signal is amplified and converted to a voltage signal proportionate to the cylinder pressure (p) (see FIG. 3). The voltage signal is fed to a fast data capture unit, where it is digitalized, processed further and stored.

In this context, it can be advantageous if the pressure signals are filtered with the help of a band pass filter (especially within the 3 to 15 kHz range and/or with high pass from 3 kHz) before comparing them with the corresponding signals of one or several standard fuels.

At the same time, it is also advantageous if the chronological sequence of the crank angle (α in FIG. 3) of the test engine is determined during combustion with the help of a crank angle sensor. With it, the determined pressure signals can be analyzed either as a function of time or of the crank angle—and with it depending on the ignition timing, for example. FIG. 4 shows a representation of the cylinder pressure and the respective, time-dependent knocking pressure amplitudes and therefore of the crank angle as well.

Knocking intensity or the other parameters mentioned above that can be derived from the cylinder pressure sequence and their distribution as a histogram, cumulative frequency or individual characteristic number can be displayed for describing the knocking (FIG. 5). Even a comparison of the individual parameters is possible, as important knowledge about the knocking behavior of the examined fuel can be gained. In this respect, it must be pointed out that the respective operating conditions of the tested engine must be considered when interpreting the captured data to prevent false resulting parameters.

A comparison of the corresponding results with those from typical standard fuels such as isooctane and/or n-heptane obtained in an analog way leads to one or several characteristic numbers that describe exactly the knocking resistance of the fuel to be tested.

In this case, the respective characteristic numbers based on the chronological sequence of the test engine's cylinder pressure signal that are obtained can be measured either during the individual test engine cycle or obtained from the statistical evaluation of the pressure signals of several cycles.

In addition, it is also advantageous if the compression ratio of the test engine is calculated based on the cylinder pressure in a defined crank angle, whereby the crank angle serves as measure of the ignition timing. Since the knocking behavior of a fuel depends, among other things, from the test engine's compression ratio, it is advantageous to take this parameter into account when comparing the pressure signals of the fuel tested with those of the standard fuel.

Furthermore, it is also extremely advantageous if the measured cylinder pressure signals and the corresponding additional known data from the tested fuel—such as, for example, its exact chemical composition—can be stored in a database. After all, these measured values are the basis for future evaluations of new or not yet analyzed fuels. In this case (and especially while complying with standardized process steps), fast and reliable conclusions about the knocking behavior of a fuel to be tested can be obtained. With the corresponding data stock, an analysis of standard fuels is therefore no longer necessary in every case.

Examples of the process steps are listed below:

• • Time-based measurement of the cylinder pressure signal (scanning frequency: 100 kHz)
  • Band pass filtration of the measured pressure signal (3 to 15 kHz)
  • Determination of the maximum amplitude height of the filtered pressure signal per cycle (best results with a signal processor)
  • Evaluation of the amplitude height and its frequency in an adjustable time window (300 cycles, intensity distribution or cumulative frequency)
  • Correlation of results with the known standard fuels or their mixtures
  • Display and storage of the correlation indices as measure for knock resistance

The important advantages of the device and the process include:

• • Higher accuracy, especially through the use of precision technology and statistical analysis
  • More exact evaluation of the knocking processes through the analysis of knocking intensity and frequency
  • Possibility of automation
  • Online display of the characteristic number of the knocking resistance
  • Automatic or half-automatic process
  • Measurement with standard fuels no longer needed as soon as the results have been made available in a database
  • The compression ratio can be exactly calculated from the pressure sequence occurring before the time of ignition

Modifications of the invention are easily possible within the framework of the patent claims, in which case it is expressly mentioned that all individual characteristics published in the patent claims, in the description and in the figures can become reality in any combination thereof as far as this is possible and makes sense. Thus, it can be quite advantageous if the pressure and/or temperature of the combustion mixture and its dwelling time (especially in form of a characteristic number calculated from this) can be taken into account. It can also be advantageous if the ignition timing and the start of knocking of the combustion mixture can be recorded (especially with the help of a sensor) and the intermediate time difference and/or the difference of the respective crank angles of the test engine are considered. In this way, it is possible to determine very accurately the time period to be analyzed, which results in an especially reliable implementation of the test. In addition, it is also possible to determine the burning period (heat input through burning) of an individual or several test engine cycles for obtaining there from the ignition delay, the preliminary ignition, the maximum burning speed and/or the residual amount of the combustion mixture from fuel and combustion gas, especially combustion air, when the knocking starts. The burning period of an individual or several cycles of the test engine can also be determined (and at least a pre-defined value of the burning period such as the start of burning) with the help of at least one sensor, especially one for measuring the ionic current inside the test engine, a sensor for measuring the structure-borne noise of the test engine and/or an optical sensor. This method allows the very precise determination of the start of the burning so that the period for the corresponding cylinder pressure measurement can be determined with pretty good reliability. It is likewise extremely advantageous for the characterization of knocking resistance to include a statistical analysis of the readings (especially those of the pressure signals), in which case this statistical analysis should encompass the recording of the readings (especially of the pressure signals) of one or several cycles (between 200 and 500, for example), of the test engine, particularly in a defined operating point of the test engine. It would also be advantageous for the statistical analysis to encompass the calculation of the mean values of the readings so that possible measurement fluctuations should only have a minimal effect on the characterization of knocking resistance,

Claims

1. Process for characterizing the knocking resistance of a fuel with the help of a test engine, characterized in that the chronological sequence of the test engine cylinder pressure is determined while the fuel burns in the test engine and the determined pressure signals are compared with the corresponding pressure signals of at least one standard fuel of known knocking resistance.

2-27. (canceled)