The early detection of the malfunctions and faults is crucial for performance and reliability of automobiles. In the last millennium, the existing automobile on-board diagnosis systems were incapable of diagnosing some faults such as loss of engine compression, plug-not-firing and carburettor. This set of faults is very important to be detected, and more so that, automobile systems of the future will be autonomous and a self driven system with little or no human interventions. A new condition monitoring approach that uses the sense of smell was investigated to diagnose the faults of plug-not-firing, loss of compression and carburettor fault. An electronic nose based condition monitoring hardware and software was developed to obtain a smell print of the exhaust fumes of an automobile gasoline engine in different operating conditions. This work presents results of experiments based on the data obtained (1400 x 10 data samples each for ten (10) fault classes) from the developed condition monitoring scheme for gasoline fuelled automobile engine using four different algorithms.