Writing

May 25, 2026

  • meta
  • site

Hello, again — re-launching kuskira.com

A short note on starting over with a simpler, faster portfolio site.

Read post

April 12, 2026

  • FEA
  • test correlation
  • fatigue
  • building block

Notes on FE-test correlation that don't fit in a slide

Practical lessons from correlating finite element models with coupon, element, and full-scale test — gauges, boundary conditions, the conservatism multiplier, and the small reference model.

Read post

February 3, 2026

  • fatigue
  • scatter
  • essay

Why fatigue is still hard, and that is fine

An essay on the parts of fatigue analysis that resist automation — scatter, history, environment, and the judgement calls no pre-processor will ever ask you about.

Read post

October 20, 2025

  • fatigue
  • spectrum
  • load interaction
  • damage tolerance

Spectrum truncation: the assumption nobody writes down

Fatigue spectra get clipped at both ends, and the clipping changes the life. Omission, truncation, load interaction, and why the conservative direction isn't obvious.

Read post

July 14, 2025

  • python
  • tooling
  • automation
  • pyNastran

Writing stress tools a team will actually use

Most internal stress scripts die on the author's laptop. What separates the few that survive — and why automation makes the assumptions louder, not quieter.

Read post

March 27, 2025

  • stress concentration
  • fatigue
  • notch
  • hand calcs

The Kt questions I keep getting asked

Stress concentration factors, gross versus net, and why the fatigue analyst and the static analyst quote different numbers for the same hole — and both are right.

Read post

November 15, 2024

  • FEA
  • meshing
  • convergence
  • practice

Mesh convergence is not a checkbox

Refining until the number stops moving is necessary, not sufficient. Singularities, stress recovery, the right quantity to converge, and knowing where to stop.

Read post

July 9, 2024

  • composites
  • failure
  • stability
  • delamination

A laminate fails three ways at once

Metals give you one failure mode to chase. A composite laminate gives you a committee — strength, stability, and the interlaminar modes that aren't in your in-plane criterion.

Read post

March 22, 2024

  • allowables
  • materials
  • knockdowns
  • certification

A-basis, B-basis, and the quiet conservatism of allowables

Where the number at the end of your margin actually comes from — the statistical basis, the knockdowns hiding inside it, and how they stack.

Read post

December 11, 2023

  • joints
  • fasteners
  • hand calcs
  • FEA

Bearing, bypass, and why I still open a hand-calc workbook

Fastener load distribution is the classic place where FE flatters you. Bearing-bypass, fastener flexibility, and why the closed-form check still rules the FE.

Read post

September 30, 2023

  • FEA
  • GFEM
  • sub-modelling
  • load path

From global model to detail: keeping the load path honest

Sub-modelling is where good global models go to die. Extracting detail loads without lying to yourself — boundaries, equilibrium, and the cases that don't commute.

Read post

June 18, 2023

  • damage tolerance
  • certification
  • crack growth
  • philosophy

Slow crack growth or fail-safe: the choice before the analysis

Damage tolerance is a design philosophy first and a crack-growth curve second. Picking the route early — and what each one actually commits you to.

Read post