Blog — Aerospace Stress, Fatigue & FEA Notes — Berkay Kuşkıra

Writing

Long-form notes on aerospace structural analysis. Looking for a quick definition instead? See the FAQ →

June 6, 2026

buckling
stability
stress
FEA

When stiffness, not strength, kills the panel

Thin panels go unstable long before they run out of strength, and the linear eigenvalue load FE hands you is an optimistic upper bound, not an answer. Notes on eigenvalue versus post-buckling, stiffener pitch, boundary conditions, and imperfection knockdowns.

May 25, 2026

meta
site

Hello, again — re-launching kuskira.com

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

May 16, 2026

thermal
stress
loads
composites

Thermal stress, the load case people forget

Constrained thermal expansion makes real stress with no external load on the part. Why a free part sees nothing, what alpha-delta-T-E buys you as a sanity check, CTE mismatch in joints, and why you can't always just add it to the mechanical case.

April 22, 2026

fractography
fatigue
failure
damage tolerance

What a broken part tells you

A short note on reading a fracture surface. Beach marks and striations, fatigue versus overload, finding the origin, and why the broken part is the one piece of evidence the analysis can't argue with.

April 12, 2026

FEA
test correlation
fatigue
building block

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

Practical lessons from correlating models with coupon, element and full-scale test. Gauges, boundary conditions, the conservatism multiplier on fatigue, and the little reference model.

February 3, 2026

fatigue
scatter
essay

Fatigue is still hard, and I've made peace with that

A short note on the parts of fatigue analysis that resist automation, mostly scatter and history and the judgement calls no pre-processor asks you about.

October 20, 2025

fatigue
spectrum
load interaction
damage tolerance

Where you clip the spectrum changes the life

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

July 14, 2025

python
tooling
automation
pyNastran

Why most of our stress scripts die on someone's laptop

What separates the few internal tools that survive from the brilliant ones nobody runs again, and why automating the arithmetic makes the assumptions louder rather than quieter.

March 27, 2025

stress concentration
fatigue
notch
hand calcs

The Kt question a junior asks me about once a month

Why the static analyst and the fatigue analyst quote different Kt for the same hole, gross versus net, and how both of them are right at the same time.

November 15, 2024

FEA
meshing
convergence
practice

Mesh convergence is not a checkbox

Refining until the number stops moving is necessary but nowhere near sufficient. Singularities, stress recovery, picking the right quantity to converge, and knowing when to stop.

July 9, 2024

composites
failure
stability
delamination

A laminate doesn't fail the way a metal does

Metals give you one failure mode to chase. A laminate gives you several at once, and the one that loses the part is usually out of the plane your in-plane criterion can see.

March 22, 2024

allowables
materials
knockdowns
certification

A-basis, B-basis, and the conservatism you don't see

Where the number at the end of your margin actually comes from. The statistical basis, the knockdowns hiding inside it, and how they stack up on a composite part.

December 11, 2023

joints
fasteners
hand calcs
FEA

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

Fastener load distribution is the classic place FE flatters you. Some thoughts on bearing-bypass, fastener flexibility, and why the spreadsheet still gets the last word.

September 30, 2023

FEA
GFEM
sub-modelling
load path

Don't trust GFEM stress, trust GFEM load

A short note on sub-modelling: the global model is a load bookkeeper, and the one check that catches most of the mistakes.

June 18, 2023

damage tolerance
certification
crack growth
philosophy

Slow crack growth or fail-safe: decide before you mesh

Damage tolerance is a design decision first and a crack-growth curve second. Some notes on picking the route early and what each one signs you up for.