The electric potential drop technique is an effective crack monitoring method, especially in harsh environment conditions such as high-temperature and high-radiation, etc. In this method, the calibration curve - the relationship between potential drop across the crack and the crack sizes - is needed. This work employed a cell-centered finite volume discretization with unstructured quadrilateral mesh to model the potential distribution within cracked specimens. The numerical model is implemented via a C++ program. The program was used to calculate calibration curves of the single edge cracked and central cracked specimen with the crack length to specimen width ratios from 0.1 to 0.8 at 0.1 intervals. The numerical calibration curve conforms well with Johnson’s analytical solutions with the maximum error of 1.15% in the single edge crack cases and 1.69% in the central cracked cases. This thesis also uses the obtained model to determine the inclined edge crack shape with crack length to specimen width ratios varies from 0.1 to 0.8 and cracked angle from 7.5° to 45° at every 7.5° interval. The adjacent potential measured positions are added to establish another calibration curve of the adjacent potential ratio. With those two curves, the inclined crack shape can be identified. This method is verified by 2 case studies. The first specimen has an inclined crack with crack length to specimen width ratio of 0.55 and inclined angle of 40°. The predicted crack length error is 0.11% and cracked angle error is 0.25%. The second specimen has an inclined crack with crack length to specimen width ratio of 0.35 and inclined angle of 10°. The predicted crack length error is 0.34% and cracked angle error is 4.5%.