The turbo engine has been around us for many years. It has always been associated with powerful and expensive sports cars. But as development has progressed and the price of turbo technology has fallen, the turbo has found its way into everyday motoring. This is not only to raise the status of the cars, but also because the turbo actually provides a more efficient engine. Turbocharger uses waste heat in the exhaust to drive a compressor which in its turn is forcing more air into the combustion chamber. Higher density of air in the combustion chamber gives that more fuel can be added, which results in a more powerful combustion. The turbo has some drawbacks. For example, the turbocharger works in a register that is narrower than the engine register, which means that it can only be used within a small portion of the engine's RPM range. When the turbo is not within its effective registry it acts as a restrictor for the engine, leading to reduced efficiency. These problems have led to many different solutions. Diesel engines have long used the variable geometry turbo (VGT). It uses small vanes in the turbine housing to control exhaust flow against the turbine, to get the best performance out of the turbo. For several reasons, the VGT turbocharger had bad impact on gasoline engines and in this project, the problem of controlling the vanes will be investigated by using microprocessor technology. The microcontroller will regulate the pressure, and therefore two pressure sensors will be used as input parameters. One will be used for the intake pressure, and the other one for the manifold pressure. An electric servo is used to control the vanes in the turbocharger. The microcontroller reads the signals from the pressure sensors several times a second. Then a “look-up table” compare those two values against each other and send a given pulse width to the servo, which sets the correct angle. Many tests had to be made to obtain the best pulse width that gave the best vane angle at different speeds and load conditions. The final code gave good results with indication of a more efficient engine. But because of the project boundaries the system was not totally finished and tested for real in a dyno for a more detailed evaluation of function and impact of efficiency.