In the oil and gas industry, vertical storage tanks need to be checked regularly using non-destructive technology to detect defects caused by corrosion and cracking. Due to the extremely large size of the tanks, the detection equipment should have high speed and should be automated. MFL storage tank wall-climbing robots are the best fit for these requirements. In MFL technology, the signal magnitude depends on many factors such as the size and configuration of the magnetizer, the thickness of the steel object to be tested, defect depth, sensor lift-off, etc. In this work, by means of finite element simulation (FEM) and by experiments using a prototype, measured on samples with artificial defects, we investigated most of the important factors affecting signal magnitude caused by defects. The results obtained help us optimize the design and manufacture of equipment, ensuring that the equipment has a minimum weight to be able to climb the tank vertically while ensuring full detection.