High power efficiency, small size and reduced heat dissipation are highly desirable for both the low power and high power devices and the switched-mode Class-D amplifiers can readily satisfy these requirements. However, full adoption of the Class-D technology has been somewhat limited due to concerns such as signal distortion and poor power supply noise rejection. In a well-designed Class-D amplifier, the intrinsic distortions dominate the linearity performance when the input signal is large and at high frequency. In the first part of this research, a systematic analysis on the intrinsic distortion of the closed-loop Class-D amplifier is investigated based on a large-signal time-domain modeling technique. The audible output components of closed-loop Class-D amplifiers with either AD- or BD-modulated 2nd-order designs are derived. The expressions are simple, accurate, and clearly reflect the relationship among the input signal, carrier frequency, and the distortion components. In the second part of this research, a novel dual-feedforward carrier modulation topology is proposed for a 2nd-order Class-D amplifier. The proposed design can be considered a combination of two solutions/steps—a quadratic carrier generator plus two feed-forward signals. It is demonstrated that the THD of the Class-D amplifier using the proposed topology has improved by ten times over a wide range of input signal frequency and magnitude. The adopted low switching frequency also ensures a low switching loss and hence high power efficiency.