Appendix A
Formula Sheet
The key equations of the course, chapter by chapter, with the equation numbers used in the text.
Contents
- A.1 Why RF, Why Open Source
- A.2 Simulating RF Circuits
- A.3 Transmission Lines, the Smith Chart and S-Parameters
- A.4 Noise and Linearity
- A.5 Modulation and Transceiver Architectures
- A.6 Transistors at Radio Frequencies
- A.7 On-Chip Passives
- A.8 Impedance Matching
- A.9 Low-Noise Amplifiers
- A.10 Mixers
- A.11 LC Oscillators and Phase Noise
- A.12 PLL Dynamics and Phase Noise
- A.13 Frequency Dividers
- A.14 Fractional-N Synthesizers
- A.15 Digital PLLs and Clock Recovery
- A.16 Power Amplifiers
- A.17 Millimetre-Wave Design: a 50 GHz Power Amplifier
- A.18 RF Layout and Electromagnetic Verification
- A.19 The Complete Transceiver
A.1 Why RF, Why Open Source
Free-space wavelength, and power in dBm:
\[\lambda_0 = \frac{c}{f} \tag{1.1}\] \[P_{dBm} = 10\log_{10}\frac{P}{1\ \mathrm{mW}} \tag{1.2}\]Power of a sinusoid of peak amplitude \(V_p\) in a resistance \(R\) (0 dBm in 50 Ω is 0.316 V peak):
\[P = \frac{V_p^2}{2R} = \frac{V_{rms}^2}{R} \tag{1.3}\]Thermal noise available from a resistor at \(T_0\) = 290 K, in a bandwidth \(B\):
\[P_n = kT_0B \;\Rightarrow\; P_{n,dBm} = -174\ \mathrm{dBm/Hz} + 10\log_{10}\frac{B}{1\ \mathrm{Hz}} \tag{1.4}\]Wavelength in a dielectric of relative permittivity \(\varepsilon_r\) and electrical length of a line of length \(\ell\) (lumped if \(\ell < \lambda/10\), i.e. \(\theta < 36°\)):
\[\lambda = \frac{c}{f\sqrt{\varepsilon_r}}, \qquad \theta = 360°\,\frac{\ell}{\lambda} \tag{1.5}\]Maximum stable and maximum available gain, with the Rollett factor \(K\):
\[\mathrm{MSG} = \left\vert\frac{S_{21}}{S_{12}}\right\vert, \qquad \mathrm{MAG} = \mathrm{MSG}\,\left(K - \sqrt{K^2 - 1}\right)\ \ (K > 1) \tag{1.6}\]Transit and maximum oscillation frequencies extrapolated from a measurement at \(f\) (gains falling at 20 dB/decade):
\[f_T \approx f\,\vert h_{21}(f)\vert, \qquad f_{max} \approx f\,\sqrt{U(f)} \tag{1.7}\]A.2 Simulating RF Circuits
S21 from an AC analysis with a source of EMF \(v_s\) behind \(Z_0\) and a load \(Z_0\) (an ngspice port with ac 1):
Noise factor from ngspice’s .noise, whose port resistance is noiseless, so that the input-referred density \(\overline{v_{n,in}^2}\) is the added noise only (\(R_S\) = 50 Ω, \(T\) the simulation temperature):
ngspice refers the noise figure to the simulation temperature \(T\); the IEEE value refers it to 290 K:
\[F_{290} = 1 + (F_T - 1)\,\frac{T}{290\ \mathrm{K}} \tag{2.3}\]Peak EMF of a sine source behind \(R_S\) with available power \(P\):
\[V_A = \sqrt{8R_S\,P} \tag{2.4}\]FFT bin spacing, and the record length for coherent sampling (\(M\) whole periods of \(f_0\)):
\[\Delta f = \frac{1}{T_{rec}}, \qquad T_{rec} = \frac{M}{f_0} \tag{2.5}\]Harmonic balance: the unknowns, and the current balance at every harmonic \(k\):
\[v(t) = \sum_{k=-N}^{N} V_k\,e^{\,\mathrm{j}k\omega_0 t}, \qquad V_{-k} = V_k^* \tag{2.6}\] \[Y(k\omega_0)\,V_k + I_k + \mathrm{j}k\omega_0 Q_k - I_{s,k} = 0 \tag{2.7}\]Input-referred third-order intercept from one point of a two-tone test (valid only where the slopes are 1 and 3):
\[\mathrm{IIP3} = P_{in} + \tfrac{1}{2}\left(P_{fund} - P_{IM3}\right) \tag{2.8}\]rms value of an ngspice trnoise(NA NT ...) source for a one-sided white density \(S\):
FDTD time step (Courant limit) for cells \(\Delta x\), \(\Delta y\), \(\Delta z\):
\[\Delta t \le \frac{1}{c\sqrt{\Delta x^{-2} + \Delta y^{-2} + \Delta z^{-2}}} \tag{2.10}\]A.3 Transmission Lines, the Smith Chart and S-Parameters
Lines. Wavelength, propagation constant and characteristic impedance; lossless and low-loss limits:
\[\lambda = \frac{c}{f\sqrt{\varepsilon_{eff}}} \tag{3.1}\] \[\gamma = \alpha + j\beta = \sqrt{(R' + j\omega L')(G' + j\omega C')}, \qquad Z_c = \sqrt{\frac{R' + j\omega L'}{G' + j\omega C'}} \tag{3.3}\] \[Z_0 = \sqrt{L'/C'}, \qquad v_p = 1/\sqrt{L'C'} \tag{3.4}\] \[\alpha \approx \frac{R'}{2Z_0} + \frac{G'Z_0}{2} \tag{3.5}\]Reflection. Reflection coefficient, VSWR, return loss and mismatch loss:
\[\Gamma_L = \frac{Z_L - Z_0}{Z_L + Z_0} \tag{3.6}\] \[\mathrm{VSWR} = \frac{1 + \vert\Gamma\vert}{1 - \vert\Gamma\vert}, \quad \mathrm{RL} = -20\log_{10}\vert\Gamma\vert, \quad \mathrm{ML} = -10\log_{10}\left(1 - \vert\Gamma\vert^2\right) \tag{3.7}\]Terminated line of length \(d\), and the quarter-wave transformer:
\[\Gamma(d) = \Gamma_L e^{-2\gamma d}, \qquad Z_{in}(d) = Z_0\,\frac{Z_L + jZ_0\tan\beta d}{Z_0 + jZ_L\tan\beta d} \tag{3.9}\] \[Z_{in}^{sc} = jZ_0\tan\beta d, \qquad Z_{in}^{oc} = -jZ_0\cot\beta d \tag{3.10}\] \[Z_{in}(\lambda/4) = \frac{Z_1^2}{R_L} \;\Rightarrow\; Z_1 = \sqrt{Z_0 R_L} \tag{3.11}\]Smith chart: \(\Gamma = (z - 1)/(z + 1)\) with \(z = Z/Z_0\) (3.12). A line of length \(d\) rotates \(\Gamma\) clockwise by \(2\beta d\) (one turn per \(\lambda/2\)).
Two-ports. Power waves, and the conversions between S, Z and Y (reference impedance \(Z_0\)):
\[a = \frac{V + Z_0 I}{2\sqrt{Z_0}}, \quad b = \frac{V - Z_0 I}{2\sqrt{Z_0}}, \qquad \mathbf{b} = \mathbf{S}\,\mathbf{a} \tag{3.16}\] \[\begin{gathered} \mathbf{Z} = Z_0(\mathbf{I} + \mathbf{S})(\mathbf{I} - \mathbf{S})^{-1}, \quad \mathbf{Y} = \frac{1}{Z_0}(\mathbf{I} - \mathbf{S})(\mathbf{I} + \mathbf{S})^{-1}, \\ \mathbf{S} = (\mathbf{Z} - Z_0\mathbf{I})(\mathbf{Z} + Z_0\mathbf{I})^{-1} \end{gathered} \tag{3.17}\] \[\Gamma_{in} = S_{11} + \frac{S_{12}S_{21}\Gamma_L}{1 - S_{22}\Gamma_L}, \qquad \Gamma_{out} = S_{22} + \frac{S_{12}S_{21}\Gamma_S}{1 - S_{11}\Gamma_S} \tag{3.18}\]Power gains:
\[G_T = \frac{(1 - \vert\Gamma_S\vert^2)\,\vert S_{21}\vert^2\,(1 - \vert\Gamma_L\vert^2)}{\vert 1 - \Gamma_S\Gamma_{in}\vert^2\,\vert 1 - S_{22}\Gamma_L\vert^2} \tag{3.19}\] \[G_P = \frac{\vert S_{21}\vert^2 (1 - \vert\Gamma_L\vert^2)}{(1 - \vert\Gamma_{in}\vert^2)\,\vert 1 - S_{22}\Gamma_L\vert^2}, \qquad G_A = \frac{\vert S_{21}\vert^2 (1 - \vert\Gamma_S\vert^2)}{\vert 1 - S_{11}\Gamma_S\vert^2\,(1 - \vert\Gamma_{out}\vert^2)} \tag{3.20}\] \[G_{TU,max} = \frac{\vert S_{21}\vert^2}{(1 - \vert S_{11}\vert^2)(1 - \vert S_{22}\vert^2)}, \qquad u = \frac{\vert S_{11}S_{12}S_{21}S_{22}\vert}{(1 - \vert S_{11}\vert^2)(1 - \vert S_{22}\vert^2)} \tag{3.21}\] \[\mathrm{MSG} = \left\vert\frac{S_{21}}{S_{12}}\right\vert, \qquad \mathrm{MAG} = \left\vert\frac{S_{21}}{S_{12}}\right\vert\left(K - \sqrt{K^2 - 1}\right) \;\; (K > 1) \tag{3.22}\] \[U = \frac{\vert Y_{21} - Y_{12}\vert^2}{4\left(\mathrm{Re}\,Y_{11}\,\mathrm{Re}\,Y_{22} - \mathrm{Re}\,Y_{12}\,\mathrm{Re}\,Y_{21}\right)} \tag{3.23}\]Stability. Unconditionally stable if \(K > 1\) and \(\vert\Delta\vert < 1\), or if \(\mu > 1\):
\[\begin{gathered} K = \frac{1 - \vert S_{11}\vert^2 - \vert S_{22}\vert^2 + \vert\Delta\vert^2}{2\,\vert S_{12}S_{21}\vert}, \quad \Delta = S_{11}S_{22} - S_{12}S_{21}, \\ \mu = \frac{1 - \vert S_{11}\vert^2}{\vert S_{22} - \Delta S_{11}^*\vert + \vert S_{12}S_{21}\vert} \end{gathered} \tag{3.24}\] \[C_L = \frac{(S_{22} - \Delta S_{11}^*)^*}{\vert S_{22}\vert^2 - \vert\Delta\vert^2}, \quad r_L = \frac{\vert S_{12}S_{21}\vert}{\left\vert\,\vert S_{22}\vert^2 - \vert\Delta\vert^2\right\vert}; \qquad C_S, r_S: \text{ swap } S_{11} \leftrightarrow S_{22} \tag{3.25}\]Open–short de-embedding:
\[\mathbf{Y}_{DUT} = \left[(\mathbf{Y}_{meas} - \mathbf{Y}_{open})^{-1} - (\mathbf{Y}_{short} - \mathbf{Y}_{open})^{-1}\right]^{-1} \tag{3.26}\]A.4 Noise and Linearity
Thermal noise of a resistance, the available noise power, and shot noise:
\[\overline{v_n^2} = 4kTR\,\Delta f \tag{4.1}\] \[P_n = kTB, \qquad kT_0 = -174\ \mathrm{dBm/Hz}\ \ (T_0 = 290\ \mathrm{K}) \tag{4.2}\] \[\overline{i_n^2} = 2qI\,\Delta f \tag{4.3}\]Input-referred noise of a transistor as an equivalent noise resistance (MOSFET; bipolar transistor):
\[\overline{v_n^2} = 4kTR_n\,\Delta f, \qquad R_n = R_g + \frac{\gamma}{g_m}, \qquad R_n = r_b + r_e + \frac{1}{2g_m} \tag{4.6}\]Noise factor, noise figure and noise temperature (source at \(T_0\), \(G_A\) the available gain):
\[F = \frac{\mathrm{SNR}_{in}}{\mathrm{SNR}_{out}} = 1 + \frac{N_{added}}{kT_0BG_A}, \qquad \mathrm{NF} = 10\log_{10}F, \qquad T_e = (F-1)\,T_0 \tag{4.7–4.8}\]Resistively terminated common-source or common-emitter stage (\(R_P = R_S\), load resistor \(R_D\)):
\[F = 2 + \frac{4R_n}{R_S} + \frac{4}{g_m^2R_DR_S} \tag{4.9}\]Noise factor for a source reflection coefficient \(\Gamma_S\), from the noise parameters:
\[F = F_{min} + \frac{4R_n}{Z_0}\,\frac{\vert\Gamma_S - \Gamma_{opt}\vert^2}{\left(1 - \vert\Gamma_S\vert^2\right)\vert 1 + \Gamma_{opt}\vert^2} \tag{4.10}\]Friis formula (available gains; each \(F_k\) for the source impedance stage \(k\) sees):
\[F = F_1 + \frac{F_2 - 1}{G_1} + \frac{F_3 - 1}{G_1G_2} + \dots \tag{4.11}\]Sensitivity, and the SNR in a noise bandwidth \(B\) at a bit rate \(R_b\):
\[P_{sens} = -174\ \mathrm{dBm/Hz} + 10\log_{10}B + \mathrm{NF} + \mathrm{SNR}_{min} \tag{4.12}\] \[\mathrm{SNR} = \frac{E_b}{N_0}\,\frac{R_b}{B} \tag{4.13}\]Memoryless nonlinearity, harmonics and compression for an input \(A\cos\omega t\):
\[y = \alpha_1x + \alpha_2x^2 + \alpha_3x^3 \tag{4.14}\] \[\mathrm{HD2} = \frac{\alpha_2A}{2\alpha_1}, \qquad \mathrm{HD3} = \frac{\alpha_3A^2}{4\alpha_1} \tag{4.15}\] \[A_{1dB} = \sqrt{0.145\left\vert\frac{\alpha_1}{\alpha_3}\right\vert}\quad (\alpha_1\alpha_3 < 0) \tag{4.16}\]Third-order intercept, and its distance from the compression point:
\[A_{IIP3} = \sqrt{\frac{4}{3}\left\vert\frac{\alpha_1}{\alpha_3}\right\vert}, \qquad \frac{A_{1dB}^2}{A_{IIP3}^2} = 0.109 \;(-9.6\ \mathrm{dB}) \tag{4.17–4.18}\]Intercepts from one two-tone measurement at input power \(P_{in}\) (per tone), with \(\Delta P_3\), \(\Delta P_2\) the distances from the fundamental to the IM3 and IM2 products:
\[\mathrm{IIP3} = P_{in} + \frac{\Delta P_3}{2}, \qquad \mathrm{IIP2} = P_{in} + \Delta P_2, \qquad A_{IIP2} = \left\vert\frac{\alpha_1}{\alpha_2}\right\vert \tag{4.19–4.20}\]Small-signal gain in the presence of a blocker of amplitude \(A_B\):
\[\alpha_{1,eff} = \alpha_1 + \frac{3}{2}\alpha_3A_B^2 \tag{4.21}\]Cascaded IIP3 (amplitudes with voltage gains \(\alpha_k\); powers with power gains \(G_k\)):
\[\frac{1}{A_{IIP3}^2} = \frac{1}{A_{IIP3,1}^2} + \frac{\alpha_1^2}{A_{IIP3,2}^2} + \dots, \qquad \frac{1}{\mathrm{IIP3}} = \frac{1}{\mathrm{IIP3}_1} + \frac{G_1}{\mathrm{IIP3}_2} + \frac{G_1G_2}{\mathrm{IIP3}_3} + \dots \tag{4.23–4.24}\]Spurious-free dynamic range (powers in dBm, \(P_{floor} = -174 + 10\log_{10}B + \mathrm{NF}\)):
\[\mathrm{SFDR} = \frac{2}{3}\left(\mathrm{IIP3} - P_{floor}\right) \tag{4.25}\]A.5 Modulation and Transceiver Architectures
A narrowband RF signal and its complex envelope:
\[s(t) = \mathrm{Re}\left\{[I(t) + jQ(t)]\,e^{j\omega_ct}\right\} = I(t)\cos\omega_ct - Q(t)\sin\omega_ct \tag{5.1}\]Error-vector magnitude (white noise, matched filter, \(E_s/N_0\) = SNR):
\[\mathrm{EVM}_{rms} = \sqrt{\frac{\langle\vert r_k - s_k\vert^2\rangle}{\langle\vert s_k\vert^2\rangle}} \approx \frac{1}{\sqrt{\mathrm{SNR}}} \tag{5.2}\]Bit-error rate of BPSK and Gray-coded QPSK; spectral efficiency with raised-cosine pulses of roll-off \(\alpha\):
\[\mathrm{BER} = Q\!\left(\sqrt{2E_b/N_0}\right), \qquad \eta = \frac{\log_2M}{1+\alpha}\ \ \mathrm{b/s/Hz} \tag{5.3–5.4}\]GFSK: frequency deviation, and the Gaussian filter of bandwidth \(B\) (\(BT\) = 0.5 for BLE):
\[\Delta f = \frac{h\,R_b}{2}, \qquad h_G(t) = \sqrt{\frac{2\pi}{\ln 2}}\,B\,\exp\!\left(-\frac{2\pi^2B^2t^2}{\ln 2}\right) \tag{5.5–5.6}\]Peak-to-average power ratio (quoted at a probability, e.g. \(10^{-4}\)):
\[\mathrm{PAPR} = \frac{\vert x(t)\vert^2_{peak}}{\langle\vert x(t)\vert^2\rangle} \tag{5.7}\]Mixing, and the image frequency (LO below / above the wanted signal):
\[\cos\omega_{RF}t\cdot\cos\omega_{LO}t = \tfrac{1}{2}\cos(\omega_{RF}-\omega_{LO})t + \tfrac{1}{2}\cos(\omega_{RF}+\omega_{LO})t \tag{5.8}\] \[f_{im} = f_{RF} \mp 2f_{IF} \tag{5.9}\]Flicker-noise penalty of a baseband noise \(N_0(1 + f_c/\vert f\vert)\) over a channel from \(f_1\) to \(f_2\):
\[\Delta N = 10\log_{10}\left[1 + \frac{f_c\ln(f_2/f_1)}{f_2 - f_1}\right] \tag{5.10}\]Second-order intermodulation of a modulated interferer of power \(P_B\) (\(\kappa\): fraction of its envelope-squared fluctuation in the channel; two tones: \(\kappa\) = 0.5 at their difference frequency):
\[P_{IM2} = 2P_B - \mathrm{IIP2} + 10\log_{10}(\kappa/2) \tag{5.11}\]Image rejection of an RC–CR Hartley combiner with corner \(f_0 = 1/2\pi RC\):
\[\mathrm{IRR}(f) = \left(\frac{1 + f/f_0}{1 - f/f_0}\right)^2 \tag{5.12}\]Quadrature down-conversion with gain error \(\varepsilon\) and phase error \(\theta\) (input \(z\) relative to the LO):
\[y = K_1z + K_2z^*, \qquad K_1 = \frac{1 + (1+\varepsilon)e^{-j\theta}}{2}, \qquad K_2 = \frac{1 - (1+\varepsilon)e^{j\theta}}{2} \tag{5.13}\]Image-rejection ratio:
\[\mathrm{IRR} = \frac{\vert K_1\vert^2}{\vert K_2\vert^2} = \frac{1 + 2(1+\varepsilon)\cos\theta + (1+\varepsilon)^2}{1 - 2(1+\varepsilon)\cos\theta + (1+\varepsilon)^2} \approx \frac{4}{\varepsilon^2 + \theta^2} \tag{5.14}\]Adler’s lock range of an oscillator with tank quality factor \(Q\) and injected current \(I_{inj}\):
\[\omega_L = \frac{\omega_0}{2Q}\,\frac{I_{inj}}{I_{osc}} \tag{5.15}\]IIP3 needed for two interferers at \(P_I\) whose IM3 falls on a wanted signal at \(P_w\):
\[\mathrm{IIP3} \ge \frac{3P_I - (P_w - \mathrm{SNR}_{min})}{2} \tag{5.16}\]Reciprocal mixing, and the phase noise it requires (\(C/I\) in dB, wanted over interferer):
\[P_{RM} = P_B + \mathcal{L}(\Delta f) + 10\log_{10}B \tag{5.17}\] \[\mathcal{L}(\Delta f) \le C/I - \mathrm{SNR}_{min} - 10\log_{10}B \tag{5.18}\]A.6 Transistors at Radio Frequencies
Transit frequency of a MOSFET (where the current gain h21 reaches one):
\[f_T \approx \frac{g_m}{2\pi\,(C_{gs} + C_{gd})} = \frac{g_m}{2\pi\,C_{gg}} \tag{6.1}\]Spot-frequency extrapolation (f0 in the −20 dB/decade region; 10 GHz for MOSFETs, 40 GHz for HBTs):
\[f_T = f_0\,\lvert h_{21}(f_0)\rvert, \qquad f_{max} = f_0\,\sqrt{U(f_0)} \tag{6.2}\]Mason’s unilateral gain, from the Y-parameters; \(f_{max}\) is the frequency where \(U = 1\):
\[U = \frac{\lvert y_{21} - y_{12}\rvert^2}{4\left[\mathrm{Re}(y_{11})\,\mathrm{Re}(y_{22}) - \mathrm{Re}(y_{12})\,\mathrm{Re}(y_{21})\right]} \tag{6.3}\]Maximum oscillation frequency of a MOSFET:
\[f_{max} \approx \frac{f_T}{2\sqrt{g_{ds}R_g + 2\pi f_T R_g C_{gd}}} \approx \sqrt{\frac{f_T}{8\pi R_g C_{gd}}} \tag{6.4}\]Gate resistance of \(n_g\) fingers of width \(W_f\), each contacted at both ends (contact resistance \(R_{con}\) per end):
\[R_g = \frac{1}{n_g}\left(\frac{R_{sq}W_f}{12\,L} + \frac{R_{con}}{2}\right) \qquad \left(\frac{R_{sq}W_f}{3L} + R_{con}\ \text{if contacted at one end}\right) \tag{6.5}\]Gate resistance in IHP’s SG13G2 RF model (rfmode = 1), \(m\) cells of \(n_g\) fingers (\(R_{SHG}\) = 3 Ω/sq, \(X_{GW}\) = 0.1 µm, \(R_{INT}\) = 13.0 Ω µm², \(R_{GO}\) = 40 Ω):
\[R_g = \frac{1}{n_g m}\left[\frac{R_{SHG}\,(W_f/6 + X_{GW})}{2L} + \frac{R_{INT}}{W_fL}\right] + \frac{R_{GO}}{m} \tag{6.6}\]Transit time of a bipolar transistor:
\[\frac{1}{2\pi f_T} = \tau_F + \frac{V_T}{I_C}\,(C_{je} + C_\mu) + C_\mu\,(r_e + r_c) \tag{6.7}\]Kirk current density (collector doping \(N_C\), width \(W_C\)):
\[J_K \approx q\,N_C\,v_{sat}\left[1 + \frac{2\varepsilon_{Si}\,(V_{CB} + \phi_{bi})}{q\,N_C\,W_C^2}\right] \tag{6.8}\]Maximum oscillation frequency of a bipolar transistor:
\[f_{max} \approx \sqrt{\frac{f_T}{8\pi\,r_b\,C_\mu}} \tag{6.9}\]Open-base breakdown (multiplication exponent \(n\) ≈ 3–6):
\[BV_{CEO} \approx \frac{BV_{CBO}}{\beta^{1/n}} \tag{6.10}\]Noise figure for a source admittance \(Y_s = G_s + jB_s\), and in terms of reflection coefficients:
\[F = F_{min} + \frac{R_n}{G_s}\,\lvert Y_s - Y_{opt}\rvert^2 = F_{min} + \frac{4R_n}{Z_0}\,\frac{\lvert\Gamma_s - \Gamma_{opt}\rvert^2}{(1 - \lvert\Gamma_s\rvert^2)\,\lvert 1 + \Gamma_{opt}\rvert^2} \tag{6.11}\]Minimum noise factor of a MOSFET (Fukui’s form) and of a bipolar transistor:
\[F_{min} \approx 1 + K\,\frac{f}{f_T}\sqrt{g_m\,(R_g + R_s)} \tag{6.12}\] \[F_{min} \approx 1 + \frac{1}{\beta} + \sqrt{2 g_m (r_b + r_e)\left(\frac{f^2}{f_T^2} + \frac{1}{\beta}\right) + \frac{1}{\beta}} \tag{6.13}\]Scaling at constant current density: \(R_n \propto 1/W\), \(Z_{opt} \propto 1/W\), \(NF_{min}\) independent of \(W\).
A.7 On-Chip Passives
Quality factor of a reactive component \(Z = R + jX\), and of a series RC or RL:
\[Q = \frac{\lvert X\rvert}{R} = 2\pi\,\frac{\text{peak energy stored}}{\text{energy lost per cycle}}, \qquad Q_C = \frac{1}{\omega R C}, \quad Q_L = \frac{\omega L}{R} \tag{7.1, 7.2}\]Self-resonance frequency of an inductor with parasitic capacitance \(C_p\) (or a capacitor with parasitic inductance \(L_p\)):
\[f_{SRF} = \frac{1}{2\pi\sqrt{L C_p}} \quad\text{or}\quad \frac{1}{2\pi\sqrt{L_p C}} \tag{7.3}\]MIM capacitance (SG13G2, typical: \(C_A\) = 1.5 fF/µm², \(C_P\) = 40 aF/µm):
\[C = C_A\,W L + C_P\cdot 2(W + L) \tag{7.4}\]Modified Wheeler formula for a planar spiral (octagon: \(K_1\) = 2.25, \(K_2\) = 3.55), with \(d_{avg} = (d_{out} + d_{in})/2\), \(\rho = (d_{out} - d_{in})/(d_{out} + d_{in})\) and \(d_{in} = d_{out} - 2Nw - 2(N-1)s\):
\[L \approx K_1\,\mu_0\,\frac{N^2 d_{avg}}{1 + K_2\,\rho} \tag{7.5}\]Current-sheet expression (octagon: \(c_1\) = 1.07, \(c_2\) = 2.29, \(c_3\) = 0, \(c_4\) = 0.19):
\[L \approx \frac{\mu_0 N^2 d_{avg}\,c_1}{2}\left[\ln\frac{c_2}{\rho} + c_3\,\rho + c_4\,\rho^2\right] \tag{7.6}\]Skin depth, and the resistance of a strip of thickness \(t\) carrying current on both faces (1-D):
\[\delta = \frac{1}{\sqrt{\pi f \mu_0 \sigma}}, \qquad \frac{R(f)}{R_{DC}} = \frac{t}{2\delta}\cdot\frac{\sinh(t/\delta) + \sin(t/\delta)}{\cosh(t/\delta) - \cos(t/\delta)} \;\to\; \frac{t}{2\delta} \quad (\delta \ll t) \tag{7.7}\]Inductance and Q of a two-port spiral, single-ended (port 2 grounded) or differential:
\[L = \frac{\mathrm{Im}\,Z}{\omega}, \qquad Q = \frac{\mathrm{Im}\,Z}{\mathrm{Re}\,Z}, \qquad Z = \frac{1}{Y_{11}}\ \text{or}\ Z_d = Z_{11} + Z_{22} - Z_{12} - Z_{21} \tag{7.8}\]Q of the π model (Yue and Wong; \(R_p\), \(C_s\): parallel equivalent of the shunt branch):
\[Q = \frac{\omega L_s}{R_s} \cdot \frac{R_p}{R_p + \left[(\omega L_s/R_s)^2 + 1\right]R_s} \cdot \left[1 - \frac{R_s^2\,(C_s + C_p)}{L_s} - \omega^2 L_s\,(C_s + C_p)\right] \tag{7.9}\]Coupling coefficient of a transformer:
\[k = \frac{M}{\sqrt{L_1 L_2}} \tag{7.10}\]Switched capacitor with switch resistance \(R_{on}\) and off capacitance \(C_{sw}\):
\[Q_{on} = \frac{1}{\omega R_{on} C_{on}}, \qquad Q_{on}\cdot\frac{C_{on}}{C_{off}} \approx \frac{1}{\omega R_{on} C_{sw}} \tag{7.11}\]Propagation constant of a line from two lines of different length (ABCD matrices \(\mathbf{T}\)):
\[\text{eig}\left(\mathbf{T}_{long}\,\mathbf{T}_{short}^{-1}\right) = e^{\pm\gamma\,\Delta l}, \qquad \gamma = \alpha + j\beta, \quad \alpha_{dB} = 8.686\,\alpha, \quad \varepsilon_{eff} = \left(\frac{c\beta}{\omega}\right)^2 \tag{7.12}\]A.8 Impedance Matching
Conjugate match and the optimum load of a power amplifier:
\[P_L = P_{avs}\left(1 - \vert\Gamma\vert^2\right), \quad \Gamma = \frac{Z_L - Z_S^*}{Z_L + Z_S} \tag{8.1}\] \[R_{opt} = \frac{(V_{CC} - V_{knee})^2}{2P_{out}} \tag{8.2}\]L-section (series element on the low side) and its bandwidth:
\[Q = \sqrt{\frac{R_{high}}{R_{low}} - 1}, \quad X_s = Q R_{low}, \quad B_p = \frac{Q}{R_{high}} \tag{8.3}\] \[B_{-10\,\mathrm{dB}} \approx \frac{2}{3}\,\frac{f_0}{Q} \quad (Q \gtrsim 2) \tag{8.4}\]π network (choose Q of the high side) and T network (choose Q of the low side):
\[R_v = \frac{R_{high}}{1 + Q^2}, \ Q_1 = \sqrt{\frac{R_{low}}{R_v} - 1}, \ C_{high} = \frac{Q}{\omega_0 R_{high}}, \ C_{low} = \frac{Q_1}{\omega_0 R_{low}}, \ L = \frac{(Q + Q_1)R_v}{\omega_0} \tag{8.5}\] \[R_v = R_{low}(1 + Q^2), \ Q_2 = \sqrt{\frac{R_v}{R_{high}} - 1}, \ L_{low} = \frac{Q R_{low}}{\omega_0}, \ L_{high} = \frac{Q_2 R_{high}}{\omega_0}, \ C = \frac{Q + Q_2}{\omega_0 R_v} \tag{8.6}\]Series trap resonant at 2f0 that looks like a capacitor C at f0:
\[L_t = \frac{1}{3\omega_0^2 C}, \qquad C_t = \frac{3C}{4} \tag{8.7}\]Insertion loss of an L-section with component quality factors QL, QC (Q of the network as built):
\[IL \approx 10\log_{10}\left(1 + \frac{Q}{Q_L} + \frac{Q}{Q_C}\right) \tag{8.8}\]Transformer: ratio, and maximum efficiency with optimum loading (\(Q_s = Q_2/\sqrt{1+x}\)):
\[a = k\sqrt{L_1/L_2}, \quad Z_{in} = a^2 Z_L \tag{8.9}\] \[\eta_{max} = \frac{x}{\left(1 + \sqrt{1 + x}\right)^2}, \quad x = k^2 Q_1 Q_2 \tag{8.10}\]Bode–Fano limit for a load R in series with C (\(Q = 1/\omega_0 RC\), \(\vert\Gamma\vert \le \Gamma_m\) over \(\Delta\omega\)):
\[\int_0^\infty \frac{1}{\omega^2}\ln\frac{1}{\vert\Gamma\vert}\,d\omega \le \pi R C \;\Rightarrow\; \frac{\Delta\omega}{\omega_0} \le \frac{\pi}{Q\ln(1/\Gamma_m)} \tag{8.11}\]A.9 Low-Noise Amplifiers
Resistively terminated common-source stage (terminating resistor \(R_P = R_S\), load \(R_D\), \(R_n = R_g + \gamma/g_m\)):
\[F = 2 + \frac{4R_n}{R_S} + \frac{4}{g_m^2 R_D R_S} \;\geq\; 2 \tag{9.1}\]Common-gate stage: input impedance and noise factor (channel noise \(4kT\gamma g_m\), \(g_i = g_m + g_{mb}\), \(\chi = g_{mb}/g_m\)):
\[Z_{in} \approx \frac{1}{g_m + g_{mb}} \tag{9.2}\] \[F = 1 + \frac{\gamma g_m}{g_i^2 R_S} + \frac{R_g g_m^2}{g_i^2 R_S} + \frac{(1 + g_iR_S)^2}{g_i^2 R_S R_D} \tag{9.3}\] \[F\big\vert_{g_iR_S = 1} = 1 + \frac{\gamma}{1 + \chi} + \frac{R_g}{R_S(1+\chi)^2} + \frac{4R_S}{R_D} \tag{9.4}\]Inductively degenerated common-source stage: input impedance, its real part and the resonance:
\[Z_{in} = j\omega(L_g + L_s) + \frac{1}{j\omega C_{gs}} + R_g + \frac{g_m}{C_{gs}}L_s \tag{9.5}\] \[\mathrm{Re}\,Z_{in} = R_g + \omega_T L_s, \quad \omega_T = \frac{g_m}{C_{gs}}, \qquad \omega_0 = \frac{1}{\sqrt{(L_g + L_s)\,C_{gs}}} \tag{9.6}\]With \(C_{gd}\) and a cascode node of gain about one (\(\omega_T = g_m/C_{gg}\)), and with a pad capacitance \(C_{pad}\) in front of the gate:
\[\mathrm{Re}\,Z_{in} \approx R_g + \omega_T L_s\left(\frac{C_{gg}}{C_{gg} + C_{gd}}\right)^2 \tag{9.7}\] \[\mathrm{Re}\,Z_{in,pad} \approx \mathrm{Re}\,Z_{in}\left(\frac{C_{gg} + C_{gd}}{C_{gg} + C_{gd} + C_{pad}}\right)^2 \tag{9.8}\]Noise factor at resonance and match (channel noise only; losses of \(L_g\), \(L_s\) and \(R_g\) in series with the source), and the input-network quality factor:
\[F = 1 + \frac{R_{Lg} + R_g + R_{Ls}}{R_S} + \gamma\,g_m R_S\left(\frac{\omega_0}{\omega_T}\right)^2 \tag{9.9}\] \[Q_L = \frac{\omega_0(L_g + L_s)}{R_S} = \frac{1}{\omega_0 R_S C_{gs}}, \qquad \gamma\,g_m R_S\left(\frac{\omega_0}{\omega_T}\right)^2 = \frac{\gamma}{Q_L}\,\frac{\omega_0}{\omega_T} \tag{9.10}\]Power-constrained optimum width with induced gate noise (long-channel, Shaeffer and Lee: \(Q_L \approx 4.5\)):
\[W_{opt} \approx \frac{1}{3\,\omega_0 L C_{ox} R_S} \tag{9.11}\]Effective transconductance of the matched input stage:
\[G_m = \frac{i_{out}}{v_s} = g_m\left\vert\frac{v_{gs}}{v_s}\right\vert \approx \frac{\omega_T}{2\omega_0 R_S} = \frac{g_m Q_L}{2} \tag{9.12}\]Resistive shunt feedback around an amplifier of transconductance \(g_m\) and load \(R_L\), and the noise-cancelling condition (auxiliary gain \(-A\) from the input node, added to the output):
\[R_{in} \approx \frac{R_f + R_L}{1 + g_m R_L} \tag{9.13}\] \[A = 1 + \frac{R_f}{R_S}, \qquad F \rightarrow 1 + \frac{R_S}{R_f} + F_{aux} - 1 \tag{9.14}\]A.10 Mixers
Fourier amplitude of the n-th harmonic of a 0/1 LO pulse train of duty cycle D, and the current-mode conversion gain of one differential output:
\[a_n = \frac{2}{\pi n}\,\lvert\sin \pi n D\rvert, \qquad \frac{i_{IF}}{i_{RF}} = a_1 = \frac{2}{\pi}\sin\pi D \tag{10.2}\]Voltage conversion gain of the 25 % current-mode mixer with a TIA, against the available source voltage:
\[G = \frac{2\sqrt{2}}{\pi}\,\frac{R_F}{R_S + R_{sw}} \tag{10.3}\]SSB noise factor of the 25 % current-mode mixer (ideal switches, source at \(T_0\), circuit at \(T\)):
\[F_{SSB} = \frac{\pi^2}{4}\left(1 + \frac{T}{T_0}\frac{R_{sw}}{R_S}\right) + \frac{T}{T_0}\,\frac{\pi^2 (R_S+R_{sw})^2}{R_SR_F} + \frac{\overline{e_n^2}\,\left(1 + \frac{R_F}{4(R_S+R_{sw})}\right)^2}{4kT_0R_S\,(2/\pi^2)\,R_F^2/(R_S+R_{sw})^2} \tag{10.5}\]Noise factor after ideal I/Q image rejection: \(F_{IR} = F_{SSB}/2\) (10.6)
Pass-band half-width and stop-band floor of an N-path filter (\(N\) paths, capacitance \(C\) each):
\[\Delta f \approx \pm\frac{1}{2\pi N(R_S + R_{sw})C}, \qquad \left.\frac{v_{rf}}{v_s}\right\vert_{\lvert f - f_{LO}\rvert \gg \Delta f} = \frac{R_{sw}}{R_S + R_{sw}} \tag{10.7}\]Gilbert cell (square-wave switching), gain and the white-noise SSB noise factor:
\[G = \frac{4}{\pi}\,g_mR_L, \qquad F_{SSB} \approx \frac{\pi^2}{4}\left(1 + \frac{2\gamma}{g_mR_S}\right) + \frac{2\pi^2}{4g_m^2R_LR_S} + \frac{2\pi\gamma I_D}{A_{LO}\,g_m^2R_S} \tag{10.8}\]A.11 LC Oscillators and Phase Noise
Barkhausen criterion for a steady oscillation at \(\omega_0\) (start-up needs \(\lvert A\beta\rvert > 1\)):
\[A(j\omega_0)\,\beta(j\omega_0) = 1 \tag{11.1}\]Parallel LC tank: resonance, quality factor and parallel loss resistance:
\[\omega_0 = \frac{1}{\sqrt{LC}}, \qquad Q = \frac{R_p}{\omega_0 L} = \omega_0 R_p C, \qquad R_p = Q\,\omega_0 L \tag{11.2}\]Cross-coupled pair: differential input resistance and start-up condition (complementary pair: \(g_m \to g_{mn} + g_{mp}\)):
\[R_{in} = -\frac{2}{g_m}, \qquad \frac{g_m}{2} \ge \frac{1}{R_p} \tag{11.3}\]Tank Q from the Q of its elements (capacitances \(C_i\) out of a total \(C\)):
\[\frac{1}{Q_{tank}} = \frac{1}{Q_L} + \sum_i \frac{C_i}{C}\,\frac{1}{Q_i} \tag{11.4}\]Differential amplitude in the current-limited regime (tail current \(I\)):
\[V_A \approx \frac{4}{\pi}\,I R_p \ \text{(complementary)}, \qquad V_A \approx \frac{2}{\pi}\,I R_p \ \text{(NMOS-only)} \tag{11.5}\]Colpitts oscillator: negative resistance seen at the transistor:
\[R_{in} = -\frac{g_m}{\omega^2 C_1 C_2} \tag{11.6}\]Tuning gain of an LC VCO tuned by a capacitance \(C_{var}(V)\) out of a total \(C\):
\[K_{VCO} = \frac{\partial f}{\partial V_{tune}} = -\frac{f_0}{2C}\,\frac{\partial C_{var}}{\partial V_{tune}} \tag{11.7}\]Leeson’s model (noise factor \(F\), signal power \(P_{sig}\), loaded \(Q\), flicker corner \(\Delta f_{1/f^3}\)):
\[\mathcal{L}(\Delta f) = 10\log_{10}\!\left[\frac{2FkT}{P_{sig}}\left(1 + \left(\frac{f_0}{2Q\,\Delta f}\right)^2\right)\left(1 + \frac{\Delta f_{1/f^3}}{\Delta f}\right)\right] \tag{11.8}\]The same for a differential LC tank with \(P_{sig} = V_A^2/2R_p\), in the \(1/f^2\) region:
\[\mathcal{L}(\Delta f) = 10\log_{10}\!\left[\frac{F\,kT R_p}{V_A^2}\left(\frac{f_0}{Q\,\Delta f}\right)^2\right] \tag{11.9}\]Impulse sensitivity function: phase step for a charge \(\Delta q\) injected at phase \(\omega_0\tau\) (\(q_{max} = C V_A\)):
\[\Delta\phi = \Gamma(\omega_0\tau)\,\frac{\Delta q}{q_{max}} = h(\omega_0\tau)\,\Delta q \tag{11.10}\]Phase noise from a white current-noise source of one-sided density \(\overline{i_n^2}/\Delta f\) (Hajimiri and Lee):
\[\mathcal{L}(\Delta f) = 10\log_{10}\!\left[\frac{\Gamma_{rms}^2}{q_{max}^2}\,\frac{\overline{i_n^2}/\Delta f}{2\,\Delta\omega^2}\right] \tag{11.11}\]Cyclostationary sources (noise modulating function \(\alpha(x)\)), and the \(1/f^3\) corner from the DC part of the effective ISF:
\[\Gamma_{eff}(x) = \Gamma(x)\,\alpha(x), \qquad \Delta f_{1/f^3} = \Delta f_{1/f}\left(\frac{\Gamma_{eff,dc}}{\Gamma_{eff,rms}}\right)^2 \tag{11.12}\]Phase noise from all sources \(k\) (ISF \(h_k(x)\) in rad/C, white density \(S_{w,k}(x)\) and flicker density \(S_{f,k}(x,\Delta f)\) at the instant \(x\) of the cycle; \(\langle\cdot\rangle\): average over one period):
\[\mathcal{L}(\Delta f) = \frac{1}{2\Delta\omega^2}\sum_k\left[\Big\langle h_k^2(x)\,S_{w,k}(x)\Big\rangle + \Big\langle h_k(x)\sqrt{S_{f,k}(x,\Delta f)}\Big\rangle^2\right] \tag{11.13}\]Phase noise from the jitter-diffusion constant \(c\) (variance of the timing error growing as \(c\,t\)), \(1/f^2\) region:
\[\mathcal{L}(\Delta f) = \frac{f_0^2\,c}{\Delta f^2} \tag{11.14}\]Oscillator figure of merit (power \(P\)):
\[\mathrm{FoM} = -\mathcal{L}(\Delta f) + 20\log_{10}\frac{f_0}{\Delta f} - 10\log_{10}\frac{P}{1\ \mathrm{mW}} \tag{11.15}\]Band calibration: frequency resolution of a counter that counts the divide-by-two output for \(n_{ref}\) reference periods, and the tuning-voltage window the loop settles in (band spacing \(\Delta f_{band}\)):
\[\Delta f_{res} = \frac{2 f_{ref}}{n_{ref}} \tag{11.16}\] \[\Delta V_{tune} \approx \frac{\Delta f_{band} + \Delta f_{res}}{K_{VCO}} \tag{11.17}\]Quadrature VCO (parallel coupling, coupling-to-switching transconductance ratio \(m\)): frequency offset from the tank’s resonance
\[\frac{\Delta\omega}{\omega_0} \approx \frac{m}{2Q} \tag{11.18}\]N identical coupled oscillators:
\[\mathcal{L}_N(\Delta f) = \mathcal{L}_1(\Delta f) - 10\log_{10}N, \qquad P_N = N P_1 \tag{11.19}\]A.12 PLL Dynamics and Phase Noise
Loop gain of a charge-pump PLL (\(K_{VCO}\) in Hz/V, \(Z(s)\) the transimpedance of the loop filter from the charge pump to the VCO):
\[G(s) = \frac{I_{CP}K_{VCO}}{N}\,\frac{Z(s)}{s} \tag{12.1}\]Third-order passive filter (\(R_1\)–\(C_1\) branch, shunt \(C_2\), extra section \(R_3\)–\(C_3\)): one zero, two poles:
\[Z(s) = \frac{1 + sT_2}{s\,C_{tot}\,(1 + sT_1)(1 + sT_3)}, \qquad T_2 = R_1C_1, \quad C_{tot} = C_1 + C_2 + C_3 \tag{12.2}\]Phase margin, maximum at \(\omega_u\) when the zero and the poles straddle it:
\[\mathrm{PM} = \arctan\omega_uT_2 - \arctan\omega_uT_1 - \arctan\omega_uT_3 \tag{12.3}\] \[T_2 \approx \frac{b}{\omega_u}, \qquad T_1 + T_3 \approx \frac{1}{b\,\omega_u}, \qquad \mathrm{PM} \approx \arctan b - \arctan\frac{1}{b} \tag{12.4}\]Total capacitance for \(\lvert G(j\omega_u)\rvert = 1\):
\[C_{tot} = \frac{I_{CP}K_{VCO}}{N\,\omega_u^2}\,\frac{\sqrt{1 + \omega_u^2T_2^2}}{\sqrt{(1 + \omega_u^2T_1^2)(1 + \omega_u^2T_3^2)}} \tag{12.5}\]Noise transfer functions to the output phase, with \(H = G/(1+G)\):
\[\phi_{out} = N H\,(\phi_{n,ref} - \phi_{n,div}) \tag{12.6}\] \[\phi_{out} = \frac{2\pi N}{I_{CP}}\,H\,i_{n,CP} \tag{12.7}\] \[\phi_{out} = \frac{2\pi K_{VCO}}{s}\,\frac{Z_{n}(s)}{1 + G}\,i_{n,R} \qquad (i_{n,R}^2 = 4kT/R\ \text{per hertz}) \tag{12.8}\] \[\phi_{out} = \frac{\phi_{n,VCO}}{1 + G} \tag{12.9}\]Effective charge-pump noise when both sources conduct for \(t_{on}\) per reference period (\(d = t_{on}f_{ref}\)):
\[S_{i,eff}(f) \approx d\,S_{i,white} + d^2\,S_{i,1/f}(f) \tag{12.10}\]Integrated phase error and jitter over an offset band:
\[\phi_{rms} = \sqrt{2\int_{f_1}^{f_2}\mathcal{L}(f)\,df}, \qquad \sigma_t = \frac{\phi_{rms}}{2\pi f_0} \tag{12.11}\]Reference spur from a ripple of peak amplitude \(V_1\) at \(f_{ref}\) on the control voltage (narrow-band FM):
\[\frac{P_{spur}}{P_c} = 20\log_{10}\frac{K_{VCO}V_1}{2f_{ref}}, \qquad V_1 = I_1\,\lvert Z(j2\pi f_{ref})\rvert \tag{12.12}\]Fundamental of the charge-pump current in lock: from a leakage \(I_{leak}\), and from an UP/DN mismatch \(\Delta I\) during a reset pulse \(t_{rst}\):
\[I_1 \approx 2I_{leak} \tag{12.13}\] \[I_1 \approx 2\pi f_{ref}^2\,\Delta I\,t_{rst}^2 \tag{12.14}\]Sampled (z-domain) loop: once per period \(T\) the charge pump adds a charge \(I_{CP}T\phi_e/2\pi\); with state transition \(\Phi = e^{AT}\) the loop is stable if every eigenvalue of the closed-loop matrix lies inside the unit circle:
\[\mathbf{x}[k+1] = \Phi\left(\mathbf{I} - \mathbf{b}\,\mathbf{c}^T\right)\mathbf{x}[k] + \Phi\,\mathbf{b}\,\phi_{ref}, \qquad \max\lvert z_i\rvert < 1 \tag{12.15}\]A.13 Frequency Dividers
Phase and phase noise through an ideal divide-by-\(N\):
\[\phi_{out}(t) = \frac{\phi_{in}(t)}{N}, \qquad \mathcal{L}_{out}(\Delta f) = \mathcal{L}_{in}(\Delta f) - 20\log_{10}N \tag{13.1}\]CML latch: output swing, and loop gain and regeneration time constant at the balanced point:
\[\Delta V = I_T R_L \tag{13.2}\] \[A_0 = g_mR_L = \frac{g_m}{I_D}\,\frac{\Delta V}{2} > 1, \qquad \tau_{reg} = \frac{C_L}{g_m - 1/R_L} \tag{13.3}\]Quadrature error of a master–slave divide-by-two driven by a sine of amplitude \(A\) (each side) with a differential offset \(V_{os}\):
\[\Delta\phi_{IQ} = \arcsin\frac{V_{os}}{2A} \approx \frac{V_{os}}{2A} \tag{13.4}\]Self-oscillation frequency of a CML divide-by-two (latch delay \(t_d\)); the input sensitivity is best near \(2f_{so}\):
\[f_{so} = \frac{1}{4\,t_d} \tag{13.5}\]Dynamic power of a TSPC (or any CMOS) divider:
\[P = C_{eff}V_{DD}^2f \tag{13.6}\]Pulse-swallow divider with a \(P/(P+1)\) prescaler, program counter \(N_P\) and swallow counter \(S\) (contiguous from \(P(P-1)\)):
\[N = P\,N_P + S, \qquad 0 \le S \le P - 1,\quad N_P \ge S \tag{13.7}\]Total lock range of a tail-injected (hard-switching) divide-by-two with tank quality factor \(Q\):
\[\left(\frac{\Delta f_{in}}{f_{in}}\right)_{total} \approx \frac{2}{3}\,\frac{1}{Q}\,\frac{I_{inj}}{I_0} \tag{13.8}\]Phase-noise floor of white edge jitter \(\sigma_t\) at the output frequency, and its in-band contribution at the PLL output:
\[\mathcal{L}_{floor} = 10\log_{10}\left(4\pi^2 f_{out}\,\sigma_t^2\right), \qquad \mathcal{L}_{out} = \frac{4\pi^2 f_{out,PLL}^2\,\sigma_t^2}{f_{out}} \tag{13.9}\]Edge jitter from the noise voltage and the slew rate at the threshold crossing:
\[\sigma_t \approx \frac{\sigma_v}{SR} \tag{13.10}\]A.14 Fractional-N Synthesizers
Average division ratio of a fractional-N divider (\(N_{int}\) integer, \(\alpha\) the fraction, \(K\) a \(B\)-bit word):
\[N = \frac{f_{LO}}{f_{ref}} = N_{int} + \alpha, \qquad \alpha = \frac{K}{2^B} \tag{14.1}\]First-order (accumulator) dithering: fractional spurs at multiples of
\[f_{spur} = \min(\alpha, 1 - \alpha)\,f_{ref} \tag{14.2}\]Output of an order-\(L\) MASH modulator (\(E\) the last stage’s quantization error, uniform, variance 1/12):
\[Y(z) = \alpha + (1 - z^{-1})^L\,E(z) \tag{14.3}\]Phase noise at the divided output’s frequency (one step = one period of the divider input), before the loop:
\[\mathcal{L}_{q}(f) = \frac{(2\pi)^2}{12\,f_{ref}}\left[2\sin\frac{\pi f}{f_{ref}}\right]^{2(L-1)} \tag{14.4}\]Through the loop (\(H = G/(1+G)\), \(H(0) = 1\)), and with cancellation of the predicted error to a residue \(\delta\):
\[\mathcal{L}_{\Delta\Sigma}(f) = \delta^2\,\lvert H(f)\rvert^2\,\mathcal{L}_{q}(f) \tag{14.5}\]Charge per reference period of a charge pump with UP/DN mismatch \(\varepsilon\) (timing error \(\Delta t\), reset pulse \(t_{rst}\)):
\[\frac{Q}{I_{CP}} = \Delta t\left(1 \pm \frac{\varepsilon}{2}\right) + \varepsilon\,t_{rst} \qquad (+\ \text{for}\ \Delta t > 0) \tag{14.6}\]Two-point modulation with a VCO-path gain error \(\gamma\): response to the frequency command
\[\frac{\Delta f_{out}}{\Delta f_{cmd}} = H + (1 + \gamma)(1 - H) = 1 + \gamma\,(1 - H) \tag{14.7}\]Behavioural DTC: delay of edge \(n\) (step \(\Delta\), gain \(g\), digital gain \(\hat g\), jitter \(\sigma_j\)) for the MASH 1-1-1 timing error \(x[n]\) in LO periods:
\[\tau[n] = g\,\Delta\,c[n] + \mathrm{INL}(c[n]) + \sigma_j\,w[n], \qquad c[n] = \mathrm{round}\!\left[\hat g\,(2 - x[n])\,\frac{T_{LO}}{\Delta}\right] \tag{14.8}\]White timing noise \(\sigma_t\) at the phase detector, at the LO through the loop:
\[\mathcal{L}(f) = \lvert H(f)\rvert^2\,\frac{(2\pi f_{LO}\,\sigma_t)^2}{f_{ref}}, \qquad \sigma_t^2 = \sigma_j^2 + \frac{(g\Delta)^2}{12} \tag{14.9}\]A.15 Digital PLLs and Clock Recovery
Phase-domain ADPLL: frequency control word and phase error (in UI of the DCO):
\[\text{FCW} = \frac{f_{DCO}}{f_{ref}}, \qquad \phi_E[k] = k\cdot\text{FCW} - \big(R_V[k] + \varepsilon[k]\big) \tag{15.1}\]Flash TDC: stages needed to cover one DCO period:
\[N_{stages} \ge \frac{T_V}{\Delta t_{res}} = \frac{1}{f_{DCO}\,\Delta t_{res}} \tag{15.2}\]In-band phase-noise floor of TDC quantization, referred to the DCO:
\[\mathcal{L}_{TDC} = \frac{(2\pi)^2}{12}\left(\frac{\Delta t_{res}}{T_V}\right)^2\frac{1}{f_{ref}} = \frac{(2\pi\,\Delta t_{res}\,f_{DCO})^2}{12\,f_{ref}} \tag{15.3}\]DCO frequency step of a capacitance step:
\[\Delta f = \frac{f}{2}\,\frac{\Delta C}{C} \tag{15.4}\]DCO quantization noise, word updated at \(f_{ref}\), and dithered by an \(n\)-th-order delta-sigma modulator at \(f_d\):
\[\mathcal{L}_{DCO,q}(\Delta f) = \frac{1}{12}\left(\frac{\Delta f_{res}}{\Delta f}\right)^2\frac{1}{f_{ref}}\,\mathrm{sinc}^2\!\left(\frac{\Delta f}{f_{ref}}\right) \tag{15.5}\] \[\mathcal{L}_{DCO,q}(\Delta f) = \frac{1}{12}\left(\frac{\Delta f_{res}}{\Delta f}\right)^2\frac{1}{f_d}\left[2\sin\frac{\pi\Delta f}{f_d}\right]^{2n}\mathrm{sinc}^2\!\left(\frac{\Delta f}{f_d}\right) \tag{15.6}\]Type-II ADPLL, open-loop gain (one cycle of latency) and its continuous-time approximation:
\[L(z) = \left(\alpha + \frac{\rho}{1-z^{-1}}\right)\frac{z^{-1}}{1-z^{-1}} \tag{15.7}\] \[L(s) \approx \frac{\alpha f_{ref}}{s} + \frac{\rho f_{ref}^2}{s^2}, \qquad \omega_n = f_{ref}\sqrt{\rho}, \quad \zeta = \frac{\alpha}{2\sqrt{\rho}}, \quad f_c \approx \frac{\alpha f_{ref}}{2\pi} \tag{15.8}\]Output phase noise:
\[\mathcal{L}_{out} = \mathcal{L}_{TDC}\,\lvert H\rvert^2 + \left(\mathcal{L}_{DCO} + \mathcal{L}_{DCO,q}\right)\lvert 1-H\rvert^2, \qquad H = \frac{L}{1+L} \tag{15.9}\]Bang-bang phase detector with Gaussian jitter of rms \(\sigma\) (UI): gain and equivalent in-band floor:
\[K_{BB} = \sqrt{\frac{2}{\pi}}\,\frac{1}{\sigma} \tag{15.10}\] \[\mathcal{L}_{BB} = (2\pi)^2\left(\frac{\pi}{2}-1\right)\frac{\sigma^2}{f_{ref}} \tag{15.11}\]Slew-rate limit of the jitter tolerance of a bang-bang CDR (step \(K_p\) UI per decision, transition density \(D_T\)):
\[A_{tol} \approx \frac{D_T\,K_p\,f_{bit}}{2\pi f_{SJ}} \tag{15.12}\]A.16 Power Amplifiers
Collector (drain) efficiency and power-added efficiency, with power gain \(G = P_{out}/P_{in}\):
\[\eta = \frac{P_{out}}{P_{DC}}, \qquad \mathrm{PAE} = \frac{P_{out} - P_{in}}{P_{DC}} = \eta\left(1 - \frac{1}{G}\right) \tag{16.1, 16.2}\]DC and fundamental components of a collector current pulse of peak \(I_{max}\) and conduction angle \(\alpha\):
\[I_{DC} = \frac{I_{max}}{2\pi}\,\frac{2\sin(\alpha/2) - \alpha\cos(\alpha/2)}{1 - \cos(\alpha/2)}, \qquad I_1 = \frac{I_{max}}{2\pi}\,\frac{\alpha - \sin\alpha}{1 - \cos(\alpha/2)} \tag{16.3}\]Collector efficiency against conduction angle, for a fundamental voltage swing \(V_1\) (class A 50 %, class B 78.5 %):
\[\eta = \frac{1}{2}\,\frac{V_1}{V_{CC}}\,\frac{I_1}{I_{DC}} = \frac{V_1}{V_{CC}}\,\frac{\alpha - \sin\alpha}{2\left[2\sin(\alpha/2) - \alpha\cos(\alpha/2)\right]} \tag{16.4}\]Optimum (load-line) resistance for an output power \(P_{out}\):
\[R_{opt} = \frac{(V_{CC} - V_{knee})^2}{2P_{out}} \tag{16.5}\]Nominal class E (50 % duty cycle, high loaded Q):
\[R = \frac{8}{\pi^2 + 4}\,\frac{V_{DD}^2}{P_{out}} = 0.577\,\frac{V_{DD}^2}{P_{out}}, \quad C_{sh} = \frac{0.184}{\omega R}, \quad X_x = 1.15\,R, \quad v_{sw,max} = 3.56\,V_{DD} \tag{16.6}\]Efficiency in output back-off \(P_{out}/P_{max}\) (ideal class A and class B):
\[\eta_A = 0.5\,\frac{P_{out}}{P_{max}}, \qquad \eta_B = \frac{\pi}{4}\sqrt{\frac{P_{out}}{P_{max}}} \tag{16.7}\]Efficiency of a matching network whose inductors have quality factor \(Q_L\) (L-section of network Q \(Q_n\); shunt feed inductor of reactance \(X_L\) across \(R_{opt}\)):
\[\eta_{net} \approx 1 - \frac{Q_n}{Q_L}, \qquad \eta_{feed} \approx 1 - \frac{R_{opt}}{Q_L X_L} \tag{16.8}\]A series L–C harmonic trap resonant at \(n f_0\) (inductor quality factor \(Q_L\)), seen at the fundamental: its reactance, and its series resistance relative to that reactance:
\[X_{trap}(f_0) = -(n^2 - 1)\,\omega_0 L, \qquad \frac{R_{trap}}{|X_{trap}(f_0)|} = \frac{1}{(n^2 - 1)\,Q_L} \tag{16.9}\]A.17 Millimetre-Wave Design: a 50 GHz Power Amplifier
Series R‖C stabilizing network in the base, and its corner frequency:
\[Z_{st} = \frac{R_{st}}{1 + j\omega R_{st}C_{st}}, \qquad f_{st} = \frac{1}{2\pi R_{st}C_{st}} \tag{17.1}\]Design conditions of the matching networks: a matched input port and the load-line load at the collector, with the output susceptance tuned out:
\[Z_{in}(\text{port 1}) = 50\ \Omega, \qquad Y_L(\text{collector}) = \frac{1}{R_L} - j\,\mathrm{Im}\,Y_{out} \tag{17.2}\]Efficiency of a matching network of loaded Q \(Q_n\) built from elements of quality \(Q_L\); quality factor of a line:
\[\eta_{net} \approx 1 - \frac{Q_n}{Q_L}, \qquad Q_{L,\,line} = \frac{\beta}{2\alpha} \tag{17.3}\]A.18 RF Layout and Electromagnetic Verification
Inductance and capacitance per unit length of a line of characteristic impedance \(Z_0\) and effective permittivity \(\varepsilon_{eff}\) (\(c\): speed of light):
\[L' = \frac{Z_0\sqrt{\varepsilon_{eff}}}{c}, \qquad C' = \frac{\sqrt{\varepsilon_{eff}}}{Z_0\,c} \tag{18.1}\]Pad over a ground shield (area \(A\), height \(h\)), parallel-plate part of its capacitance:
\[C_{pad} \approx \frac{\varepsilon_0\varepsilon_r A}{h} + C_{fringe} \tag{18.2}\]Smallest parallel resistance a series element can see behind a pad loaded by \(R_0\) = 50 Ω:
\[R_{min} = \mathrm{Re}\left(R_0 \parallel \frac{1}{j\omega C_{pad}}\right) = \frac{R_0}{1 + (\omega C_{pad} R_0)^2} \tag{18.3}\]Rational (vector-fitted) model of an EM-simulated n-port, with common stable poles \(p_k\):
\[\mathbf{S}(s) \approx \mathbf{D} + \sum_{k} \frac{\mathbf{R}_k}{s - p_k} \tag{18.4}\]Port of the circuit realization (reference \(Z_0\); \(I_i\) into the port): incident and reflected waves in volts,
\[V_i = Z_0 I_i + 2b_i, \qquad a_i = V_i - b_i, \qquad \mathbf{b} = \mathbf{S}\,\mathbf{a} \tag{18.5}\]A.19 The Complete Transceiver
Share of block \(k\) in the noise factor \(F\) of a cascade (available gains \(G\) as ratios); the same construction with \(G_1\cdots G_{k-1}/\mathrm{IIP3}_k\) gives the shares of \(1/\mathrm{IIP3}\):
\[s_1 = \frac{F_1}{F}, \qquad s_k = \frac{F_k - 1}{F\,G_1G_2\cdots G_{k-1}}\ (k > 1), \qquad \sum_k s_k = 1 \tag{19.1}\]Second-order intercept of a mixer referred to the antenna (gains in dB):
\[\mathrm{IIP2}_{ant} = \mathrm{IIP2}_{mixer} - \left(G_{filter} + G_{LNA}\right) \tag{19.2}\]I/Q phase error from the threshold mismatch of a divider’s clock pair (Pelgrom), with the divider’s sensitivity \(d\theta/dV_{os}\):
\[\sigma_{\theta} = \frac{d\theta}{dV_{os}}\,\frac{A_{VT}}{\sqrt{WL}} \tag{19.3}\]Adler lock range for an injected power \(P_{inj}\) reaching a tank that oscillates with power \(P_{osc} = V_{osc}^2/2R_p\):
\[f_L = \frac{f_0}{2Q}\sqrt{\frac{P_{inj}}{P_{osc}}} \tag{19.4}\]Ripple at the reference frequency on the tuning line that produces a spur \(S\) (dBc at the VCO):
\[V_r = \frac{2f_{ref}}{K_{VCO}}\,10^{S/20} \tag{19.5}\]Largest blocker for 3 dB of desensitization by reciprocal mixing (\(\overline{\mathcal{L}}\): the LO’s phase noise averaged over the offsets the blocker occupies), and the rejection \(R\) a filter must add for a transmitter of power \(P_{TX}\) behind an antenna isolation \(I_{ant}\):
\[P_{B,max} = \left(-174 + \mathrm{NF} + 10\log_{10}B\right) - \overline{\mathcal{L}} - 10\log_{10}B, \qquad R \ge P_{TX} - I_{ant} - P_{B,max} \tag{19.6}\]