Skip to content

Simulation: Investigate loss mismatch between simulation and bench measurements #10

Description

@luizvilla

Context

Source: session recap email (2026-02-19)

Problem

Current simulation predicts lower losses than bench measurements. Candidate causes were identified (Rac, capacitor ESR, component models), but no structured investigation exists yet.

Expected Outcome

A calibrated model investigation closes the gap between simulation and measurement, starting with DC total loss alignment before AC refinements.

Scope

In:

  • Quantify current simulation-vs-measurement mismatch.
  • Test candidate contributors: Rac, capacitor ESR, component model updates.
  • Prioritize matching Ptot DC before extending to dynamic conditions.
    Out:
  • Major control software refactor.
  • Bench hardware redesign.

Acceptance Criteria

  • Baseline mismatch (simulation vs bench) is reported with reproducible test case.
  • At least 3 candidate model adjustments are evaluated and documented.
  • A recommended updated model set is proposed with before/after error metrics.

Evidence / Notes

  • Email excerpt: "Les premiers ajustements ont montré des pertes plus faibles entre simu et manipe ... Ajouts Rac, ESR condensateurs, modèles composants ... se rapprocher au niveau Ptot DC"
  • Related files/paths: simulations/Hackathon_Si_Losses.asc, simulations/Hackathon_Si_Losses_full.asc
  • Related issue(s): none

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions