The third pillar of the keyword—"Simulation"—is where theory meets reality. Delta-Sigma converters are notoriously difficult to simulate because they require a mix of transient analysis and spectral analysis.
Standard SPICE transient simulations are often too slow for Delta-Sigma converters. Because the modulator is oversampling, one must simulate thousands of clock cycles to gather enough data points for a meaningful FFT (Fast Fourier Transform).
by Richard Schreier and Gabor C. Temes (Wiley-IEEE Press)
Finite opamp gain (e.g., 60 dB) causes the integrator pole to move from DC → small offset. This degrades NTF notch depth, increasing in-band noise. Requirement: Opamp DC gain >> OSR for high resolution.
[ P_n,in = \int_-f_B^f_B \frac\Delta^212 f_s \cdot |NTF(f)|^2 df \approx \frac\pi^236 \cdot \frac\Delta^2OSR^3 ]
The third pillar of the keyword—"Simulation"—is where theory meets reality. Delta-Sigma converters are notoriously difficult to simulate because they require a mix of transient analysis and spectral analysis.
Standard SPICE transient simulations are often too slow for Delta-Sigma converters. Because the modulator is oversampling, one must simulate thousands of clock cycles to gather enough data points for a meaningful FFT (Fast Fourier Transform). delta-sigma data converters theory design and simulation pdf
by Richard Schreier and Gabor C. Temes (Wiley-IEEE Press) OSR for high resolution.
[ P_n
Finite opamp gain (e.g., 60 dB) causes the integrator pole to move from DC → small offset. This degrades NTF notch depth, increasing in-band noise. Requirement: Opamp DC gain >> OSR for high resolution. delta-sigma data converters theory design and simulation pdf
[ P_n,in = \int_-f_B^f_B \frac\Delta^212 f_s \cdot |NTF(f)|^2 df \approx \frac\pi^236 \cdot \frac\Delta^2OSR^3 ]