A 2 A Dual Loop LDO With Dynamic Negative Feedback Loop and G_m Boosting Error Amplifier for Off-Chip and Cap-Less Applications

This paper proposes a dual loop low-dropout regulator (LDO) for heavy load current (I<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {L}}$ </tex-math></inline-formula>). The proposed LDO features a fast load transient response and wide output capacitor (C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula>) range. The proposed dynamic negative feedback loop (DNFL) enhances the transient response by reducing the undershoot and widening the bandwidth. The transient response is improved further by the proposed dynamic G<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {m}}$ </tex-math></inline-formula> boosting error amplifier. The C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula> range gets wider thanks to the proposed DNFL, which introduces an additional zero in the feedback loop. The zero secures stability in a wide C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula> range. The prototype of the proposed LDO is fabricated in a 0.5-<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu $ </tex-math></inline-formula> m CMOS process. The input voltage (V<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {IN}}$ </tex-math></inline-formula>) ranges from 2.7 to 4.2 V, while the output voltage (V<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula>) ranges from 2.4 to 4.0 V. The maximum I<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {L}}$ </tex-math></inline-formula> is 2 A. The prototype provides enough phase margin across a C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula> range from 0 to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1~\mu $ </tex-math></inline-formula> F. The measured results show a 68-mV undershoot with a load current of 2.5 A/<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1~\mu $ </tex-math></inline-formula> s on a 0-<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu $ </tex-math></inline-formula> F C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula> capacitor. When the C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula> is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1~\mu $ </tex-math></inline-formula> F under the same conditions, the proposed LDO features an undershoot of 13 mV. The proposed LDO achieves the best FoM among the state-of-the-art LDOs without and with the off-chip C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula>. In cases where the off-chip C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula> is not employed, the achieved FoM is 5.228 ps<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\cdot V/\mu $ </tex-math></inline-formula> m2. When using the off-chip C<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathbf {O}}$ </tex-math></inline-formula>, the FoM is 0.624 ps/<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu $ </tex-math></inline-formula> m2.