Skip to content

pdnkit: CavityModel uses HJ roughness#52

Open
UnsignedChad wants to merge 1 commit into
mainfrom
pdnkit/cavity-hj
Open

pdnkit: CavityModel uses HJ roughness#52
UnsignedChad wants to merge 1 commit into
mainfrom
pdnkit/cavity-hj

Conversation

@UnsignedChad

@UnsignedChad UnsignedChad commented May 25, 2026

Copy link
Copy Markdown
Owner

CavityConfig gains conductor_roughness_rq_m. When > 0, the cavity routine augments the dielectric loss tangent at each frequency with K_HJ * (delta_s / d) -- the parallel-plate surface-loss term scaled by Hammerstad-Jensen roughness.

Default 0 keeps the existing behavior bit-for-bit. Typical Rq:

  • 0.4 µm -- smooth ED foil
  • 1.0 µm -- standard rolled copper
  • 2–5 µm -- black-oxide-treated

Effect concentrates near resonances (where Q matters) and at high frequency (where δ_s shrinks toward Rq).

3 tests in [cavity-hj][validation]: zero roughness identical to old; 5 µm Rq damps the analytic TM10 peak; off-resonance change is small (< 5%).

CavityConfig gains conductor_roughness_rq_m. When > 0, the cavity
impedance routine augments the dielectric loss tangent at each
frequency with K_HJ * (delta_s / d), the standard parallel-plate
surface-loss term scaled by Hammerstad-Jensen roughness.

Default 0 = smooth, preserves the bit-for-bit old behavior. Typical
values: 0.4 um (smooth ED foil), 1 um (standard rolled), 2-5 um
(black-oxide-treated). Effect is biggest at high frequency and near
resonances where Q matters.

3 tests in [cavity-hj][validation]:
  * rq = 0 matches the old result bit-for-bit
  * rq = 5 um damps the cavity peak at the analytic TM10 freq
  * off-resonance change is small (< 5 percent)
@UnsignedChad UnsignedChad enabled auto-merge (squash) May 25, 2026 00:19
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant