Search Results

Cylinder pressure analysis in internal combustion engines involves measuring very large quantities of data, typically 2 Mbytes per test point. Derived results, such as peak cylinder pressure, IMEP and burn angles can be calculated either in real-time or by post-processing of the pressure data and the need for long term saving of the raw data after processing is then optional. Despite the many advantages of retaining the raw data, this is often not done because of the necessity for large capacity storage systems and the time required to transfer the large files across networks. The ideal situation would be to save the derived results and statistical parameters plus the raw pressure data in a much more efficient format which offers substantially smaller data files without significant loss of accuracy and functionality.

Accurate heat release analysis of cylinder pressure data is a powerful tool used in the development of diesel engines. However, significant errors in the calculated heat release values can occur due to shortcomings in both the experimental measurements and in the heat release model and this can produce misleading results. This paper shows the effect of such common errors on the calculated gross heat release data obtained when analysing simulated and experimental direct injection diesel engine pressure diagrams using a traditional single-zone First Law heat release model. The work reveals that the greatest uncertainty in most cases will be caused by assuming the wrong rate of heat transfer between the cylinder charge and combustion chamber walls. To overcome this limitation, an alternative heat release model is proposed and shown to give very good results over a wide range of operating conditions.

This paper addresses issues associated with the accurate determination of absolute cylinder pressure in internal combustion engines. Pressure referencing errors are shown to produce large errors in derived parameters such as polytropic index, mass fraction burned and charge temperature. Two alternative pressure correction methods, namely inlet manifold pressure and polytropic index referencing are investigated in detail. Sources of errors and algorithm improvements are investigated and discussed. Comparison between the two pressure referencing techniques is made using measured cylinder pressure data obtained from a gasoline engine operating over a wide range of speeds and loads. The analysis shows that both of the methods should be capable of referencing typical experimental pressure data to within +-100 mbar. The work has demonstrated that accurate absolute pressure referencing can only be achieved if common pressure measurement errors are minimised.

This paper addresses issues associated with the accurate determination of mass fraction burned (MFB) in gasoline engines. Items covered include an evaluation of the accuracy of alternative MFB models and the effects of errors in the absolute pressure referencing, crank angle phasing and assigned compression ratio. The implications of using crank angle averaged pressure data and varying the crank angle resolution and number of engine cycles are also covered. The well known Rassweiler and Withrow MFB model was found to produce the best results in comparative tests with simulated and experimental pressure data. Absolute pressure referencing offset caused the largest error in the calculated MFB and burn angles, particularly at low engine load. Calculated data at the extreme ends of the MFB curve were shown to be most sensitive to measurement errors and noise.

This paper addresses issues associated with the accurate determination of indicated mean effective pressure (imep) for internal combustion engines. Items covered include a comparison of alternative imep equations and the effects of crank angle resolution, signal noise, thermal shock, crank angle phasing and connecting rod length errors. The best imep equation to use has been identified and the magnitude of errors under a wide range of conditions quantified. The results show that inaccuracy in the calculated imep will mainly be caused by thermal shock and errors in the crank angle phasing and transducer sensitivity. In contrast, the effects of coarse crank angle resolution, incorrectly specified connecting rod length, signal noise and integration period error should be relatively small.