Increasing accuracy and repeatability of fuel consumption measurement in chassis dynamometer testing

Abstract

The aim of this paper is to identify and investigate the effect of small changes in test conditions when quantifying fuel consumption. Twelve test set-up variables were identified and intentionally perturbed from a standard condition, including the effect of removing the power-assisted steering pump. Initially a design-of-experiments (DoE) approach was adopted and the results showed that most of the tested parameters had significant effects on fuel consumption. Most of these effects were greater than the effect of typical technology changes assessed on chassis dynamometer facilities. For example, an increase of 8.7 per cent in fuel consumption was observed following a 90min battery discharge from vehicle headlamps. Similarly an increase of 5.5 per cent was observed when the rig was run 3km/h faster over a drive cycle, and 2.6 per cent when using tyres deflated by 0.5 bar. As a consequence, statistical tolerancing was used to suggest typical tolerances for test rig set-up variables. For example it was recommended that the tyre pressure be controlled to within 0.1 bar and the test rig speed to 0.3km/h. Further investigations were conducted into the effect of battery discharge, coast-down time, and engine cooling. These highlighted the need for rigorous battery charge management as the battery voltage was found not to be an appropriate measure of the variation in the alternator loading. Coast-down time was found to be a good control measure for a number of set-up variables affecting the rolling resistance of the vehicle. Finally the variations in the engine cooling were quantified using a cumulative engine temperature over a drive cycle. This was found to correlate well with fuel consumption. For each of these subsequent investigations, results were compared with the DoE predictions and found to agree well when considering the relatively low number of tests compared with the number of factors.