Taking the transformer with Y0，d11 connection mode as an example， this paper analyzes and compares the sensitivity of four transformer differential protection schemes with different compensation methods， which are widely used in the practice. The theoretical analysis and simulation results show that， the sensitivity of differential protection is influenced by the compensation method. The differential protection sensitivity of different principles is determined by the reservation of sequence component in differential current. The sensitivity of winding differential protection and phase current minus zero-sequence current differential protection are higher because they fully reserve the sequence component of fault current. The phase-to-phase current differential protection and phase current minus zero-sequence（self-produced） current differential protection widely used are of lower sensitivity because they have no relation with the zero-sequence current of inner fault current. However they have no difference in the essence and can be compensated with proper settings.