The purpose of my thesis was to study the LabCar system, which was produced by Etas for Harware-in-the-Loop (HiL) tests at Robert Bosch Ltd., with a special focus on measuring uncertainty of the tower.
In my dissertation, firstly, I am presenting the increasingly spreading embedded systems in the automotive industry, the development standard and the development model corresponding to international standards. In addition, I present HiL testing and the hardware and software architecture of LabCar used to test power transmission controller units (TCUs). Finally, as an introduction to my measurements I give an overview of the theoretical background of measurement uncertainty.
My basic assumption was at the beginning of the measurement that all non-ideal electrical components that are powered by a potential source of heat are therefore the most critical factor for the tower. To test this, I have set up measurement models and, during several measurements, I have observed the behavior of the measuring tower right after a cold start, as at constant load the tower is warmed to the operating temperature by sticking thermocouples to the critical points. Apart from the temperature dependence, I also studied the calibration and loss of the long signal paths.
Summing up and evaluating my measurement results, I found that although many times the temperature and the impact of the signal are present, their effect is so small (and only in a few cases are counted) that in the ordinary measurement exercise they are negligible. In the few cases, it is worth considering that a correction tag can even further refine the measurement model.