VESPA motorbikes are two-wheeled transportation that are widely used to support daily activities. VESPA engine has a carburetor that serves to regulate the entry of a mixture of air and fuel into the engine. Carburetor settings need to be done to get a mixture of fuel and air which is ideal so that performance and fuel consumption are optimal. Carburetor faults often occur because they are done mechanically so that they require mechanics with special skills. A method is needed for error detection of fuel mixtures. The error detection method that has been used is Principal Componenet Analysis (PCA). This study aims to apply the PCA method to detect fuel mixture errors on the 150cc 2 stroke VESPA Super engine. This study uses a carburetor sprint sport dellorto 20 20 type with three mixed variations, fuel-rich, fuel-poor and normal. Vespa engine is connected with NI-9234 type modules installed on the NI-DAQ-9174 chassis and Bruel & Kjaer accelerometer type 4507 B. 2200 rpm engine speed at each condition is measured using a tachometer. Retrieving script data using the NI-MAX application. Furthermore, extracting time domain data using Standard Deviation, Root Mean Square (RMS), Kurtosis, Skewness, Peak Value, Variance, and Crest Factor statistical parameters is processed using Matlab R2016a for PCA analysis. The results extraction from seven time domain parameters showed mixed results, so it was still not effective for detecting fuel mixture errors. Furthermore, PCA is used with PC1 and PC2, but has not yet obtained optimum results. Then PC3 is added, the results can calcify well between normal, rich and poor fuel mix conditions using time domain statistical parameters that produce a percentage value of 96.26% of the data variance.