When an engine component participating in the combustion process of an internal combustion piston engine, malfunctions, this malfunction may be reflected in the ensuing cylinder pressure traces, acoustic emission and vibration signals. In this paper, we explore the idea of exploiting information detected b y pressure, vibration and acoustic emission sensors in order to develop fault diagnostic classifiers. It is shown that, following training on examples of normal operation of a diesel engine and four fault conditions, Artificial Neural Nets based on data from anyone of these three sensors can be used to identify the fault condition. In addition, a system consisting of an ensemble of three nets, each of which is based on a different sensor, can be assembled. The advantages of such a system in terms of protection against sensor failure are discussed.