quantcore_circuits.analysis.epr¶
Energy-participation-ratio Hamiltonian (pyEPR calculation core, no AEDT needed).
Input: linear eigenmode frequencies (HFSS/Palace eigenmode with Lj as a lumped inductor), junction participation
ratios p_mj (M modes × J junctions), signs s_mj and junction inductances. pyEPR’s QuantumAnalysis needs a
pickle produced by DistributedAnalysis; the calculation functions below are what it calls internally.
participations() is pyEPR’s DistributedAnalysis.calc_p_junction + QuantumAnalysis renormalisation
(config.epr.renorm_pj = True) applied to the field integrals that qcc_solve.py evaluates inside AEDT
(quantcore_circuits.sim_tools.aedt_scripts), method line_voltage:
V_j = sign(∫Re E·dl) √((∫Re E·dl)² + (∫Im E·dl)²) over the junction line, I_j = V_j / (ω L_j);
U_E = ∫Re(E·εE*)/2, U_H = ∫Re(H·μH*)/2 (peak energies), U_J = ½ L_j I_j², U_Cj = ½ C_j V_j²;
p_mj = U_J / (U_E + ΣU_Cj), then rescaled so Σ_j p_mj = |(Ū − U_H)/Ū|, Ū = mean of total capacitive and inductive energy; s_mj = sign(V_j).
Sign conventions are converted to qcc’s (α < 0, χ = E11 − E10 − E01 + E00 < 0); pyEPR reports positive shifts.
Functions¶
|
Participations ( |
|
Dressed frequencies and χ matrix from EPR data. |
|
Independent QuTiP check of the EPR Hamiltonian (not pyEPR's code path). |
|
p_J = ½ Lj I² / U_total for a mode with peak junction current |
|
Modes that are a junction's lumped L_J ‖ C_J resonating on its own: p_mj ≥ |
|
Junction participations from qcc's in-AEDT EPR integrals ( |
Module Contents¶
- quantcore_circuits.analysis.epr.epr_analysis(epr: dict, freqs_GHz=None, Q=None, tan_delta: float | None = None, **kw) dict¶
Participations (
participations()) + EPR Hamiltonian (epr_hamiltonian()) in one dict.Adds
chi_MHz(M×M, diagonal = α),f_GHzdressed (numerical diagonalisation), the first-order values, and withtan_deltathe bulk-dielectric Q per mode (1 / (p_substrate tan δ));Q(per mode) is passed through.
- quantcore_circuits.analysis.epr.epr_hamiltonian(freqs_GHz, p_mj, lj_nH, s_mj=None, cos_trunc: int = 8, fock_trunc: int = 9, numerical: bool = True) dict¶
Dressed frequencies and χ matrix from EPR data.
Returns
f_GHz(dressed, numerical diagonalisation or 1st-order PT),chi_MHz(M×M; diagonal = α), and the first-order valuesf_O1_GHz/chi_O1_MHz.
- quantcore_circuits.analysis.epr.epr_qutip(freqs_GHz, p_mj, lj_nH, s_mj=None, fock: int = 8) dict¶
Independent QuTiP check of the EPR Hamiltonian (not pyEPR’s code path).
H = sum_m h f_m a_m^+ a_m - sum_j EJ_j [cos(phi_j) + phi_j^2 / 2], phi_j = sum_m s_mj phi_mj (a_m + a_m^+), phi_mj^2 = p_mj h f_m / (2 EJ_j), EJ = (hbar / 2e)^2 / Lj. The cosine is exact (
cosm), so only the Fock truncation differs from pyEPR’s Taylor series. Returns the dressed single-excitationf_GHzandchi_MHz(M x M, qcc sign: diagonal = alpha, off-diagonal chi_mn = E(1_m 1_n) - E(1_m) - E(1_n) + E(0)).
- quantcore_circuits.analysis.epr.junction_participation_from_energy(freq_GHz: float, lj_nH: float, i_peak_A: float, u_total_J: float) float¶
p_J = ½ Lj I² / U_total for a mode with peak junction current
i_peak_Aand total energyu_total_J.
- quantcore_circuits.analysis.epr.junction_self_modes(p: dict, p_min: float = 0.95, rel: float = 0.15) list[int]¶
Modes that are a junction’s lumped L_J ‖ C_J resonating on its own: p_mj ≥
p_minfor one junction and f withinrelof 1/2π√(L_J C_J) of that junction (with C_J > 0). Seen on coarse / fast meshes (tutorial 00, C_J 14.7 fF: a 15.65 GHz mode with p_J 0.9999 beside the real qubit mode, p_J 0.985); a circuit mode has the island’s capacitance in parallel and sits far below. Returned as mode indices (0-based).
- quantcore_circuits.analysis.epr.participations(epr: dict, freqs_GHz=None, renorm: bool = True) dict¶
Junction participations from qcc’s in-AEDT EPR integrals (
<sim>_project_qcc.json"epr").freqs_GHz(mode order) overrides the per-modefreq_Hzread in AEDT (e.g. the.eigexport). Returns{"junctions", "Lj_nH", "Cj_fF", "f_GHz", "p_mj", "s_mj", "p_mj_raw", "U_diff", "p_volume"}with M×J lists.