LAB NOTE 001
← iSight Laboratories 𝑖 SIP · i5.1.4 The weekly lab note Hangar
Lab note 001 · two ways to keep a fleet in the air · iSIP i5.1.4 — Durability & Damage Tolerance Control

ASSUME FLAWLESS,
OR ASSUME FLAWED.

The U.S. Navy and the U.S. Air Force keep aircraft airworthy with two different philosophies at the core of their programs. Neither service runs its philosophy pure — the real programs are hybrids, and this note says where — but the two postures underneath are real, they are opposites, and each has a bill. Read them as engineering first. Then notice that you have been running one of them on yourself, without ever choosing it.

Safe-life: retire before the crack

Safe-life starts from a clean assumption: the airframe is flawless when it leaves the factory. Engineers test a full-scale copy until it cracks, divide the result by a safety multiplier, and that number becomes the airplane's life. When the hours run out, the aircraft retires — before any crack is expected to start. The slogan is safety by retirement. Nobody has to find the crack, because the airplane is gone before it exists.

This is the posture at the backbone of the Navy's carrier-aircraft program, and for reasons that make sense on a flight deck: catapult launches, arrested landings and salt air produce loads and corrosion that argue for a fixed, conservative life. The Navy does not guess at the number. Every tail carries a running tally called Fatigue Life Expended — the fraction of the allotted life that airframe has actually used, updated from what it flew — so "retire at the number" is an accounting, kept per aircraft. And the Navy is not purely safe-life: it runs damage tolerance on some fleets. The label is the backbone, not the whole skeleton.

SAFE-LIFE · RETIRE BEFORE THE CRACK
hours flown →crack size ↑nobody looksretire hereassumed flawless up to this linethe airplane is gone beforethe crack is expected to start

The crack that would have grown is drawn dashed, because in this philosophy nobody ever looks for it. The red line is the retirement point, fixed on day one.

It is simple to communicate, easy to track fleet-wide, cheap to run, and it makes deployment decisions fast. The bill comes due elsewhere: the number is fixed, the airplane retires on schedule whether or not it was flown gently, and the whole scheme leans on the assumptions made on day one. If the missions change, or the sensors go quiet, the only safe move is to be more conservative — and sometimes to ground the fleet.

Damage tolerance: inspect and fly on

Damage tolerance starts from the opposite assumption: the airframe is already flawed. Every one. Somewhere in the structure there is a crack too small to see, and the question is not whether it exists but how fast it grows. So you calculate the growth, you set an inspection interval short enough to catch the crack before it turns critical, and you fly on. The slogan is safety by inspection. The aircraft's life is not a fixed number; it is whatever the record and the inspections will support.

This is the posture at the backbone of the Air Force's program, and it was learned the hard way. In 1969 an F-111 lost a wing in flight because of a flaw the safe-life assumptions said could not be there. The service rebuilt its structural-integrity standard on the opposite assumption — that a flaw is there — and has run it that way for fifty years, with fleet-wide load recording and per-tail usage tracking to keep the inspection intervals honest. Civil transports, broadly, took the same road: damage tolerance for the main structure, with safe-life kept for the parts that cannot practically be inspected, landing gear being the usual example.

DAMAGE TOLERANCE · INSPECT AND FLY ON
hours flown →crack size ↑flawed from day onecritical sizethe blue marks are inspections,timed to catch the crack beforeit reaches the red line

The crack is assumed present from the first hour. The blue marks are inspections, spaced so the crack is always found before it reaches the red line. The airplane keeps flying with a known flaw under a known limit.

It is accurate, flexible, and empirical — each tail number is tracked against what it actually flew, so a gently-used airframe can serve past its design life while a hard-used one gets watched more closely. The bill: complexity, cost, onboard instrumentation, and a total dependence on the inspections being done and the analysis being right. Miss an inspection and the philosophy has nothing underneath it.

Who flies which

THE BACKBONE, AND THE HYBRID
FleetBackboneWhere it is a hybrid
U.S. NavySafe-life, with Fatigue Life Expended tracked per tailDamage tolerance on some fleets
U.S. Air ForceDamage tolerance, fleet-wide, since the 1970sDurability (economic-life) analysis alongside; safe-life logic for certain components
Civil transportsDamage tolerance for the main structureSafe-life for parts that cannot practically be inspected, such as landing gear

Two postures, three fleets, no purist. That is worth holding onto for the second half of this note: the mature programs are the ones that know which assumption they are running where, and why.

Side by side

ENGINEERING FIRST · THE HUMAN LINE UNDER EACH ROW
Safe-life (the Navy backbone)Damage tolerance (the Air Force backbone)
Starting assumptionFlawless until proven otherwise
"I'm fine. I'll deal with it when it's a problem."
Flawed from day one
"There's a crack somewhere. Let's find out how fast it's growing."
How safety is keptRetire before the crack starts
Quit the job, the marriage, the town — before anything is looked at.
Inspect on an interval, fly on
Look on a schedule. Keep flying with a known flaw under a known limit.
What sets the lifeA fixed number from a test article and a safety multiplier
Somebody else's timeline — parents', church's, employer's.
Your own record, updated
What you actually flew, measured, not what the brochure said you could take.
StrengthsSimple · visible · cheap · fast decisions
Everybody understands "I'm done."
Accurate · flexible · proactive · fewer surprises
You find out early, and you get to choose.
WeaknessesInflexible · premature retirement · relies on day-one assumptions · goes conservative when data is missing
Leaves early, or grounds itself, because it never looked.
Complex · costly · needs instrumentation · everything depends on the inspection actually happening
It only works if you log the number and keep the appointment with yourself.
Safety by retirement, or safety by inspection.
You are already running one of them.

Which one have you been flying?

Most of us were issued safe-life without being asked. It is the default of every institution that raised us: assume the person is sound, give them a fixed allotment, and when they run out, retire them — from the team, the role, the family's good graces. It is simple and it is visible and it never, ever looks inside. The failure mode is not dramatic. It is a person who leaves early, again and again, because the only tool they were given for a crack was the exit.

There is a second failure mode, and it is the one turned on other people. Safe-life applied to a human being says: fixed allotment, then disposal. Aircraft do not even get treated that way. Outside Tucson, at Davis-Monthan Air Force Base, sits the boneyard — thousands of retired airframes parked in the desert — and it is not a graveyard. Aircraft come out of it regularly — repaired, refitted, sometimes re-winged — and go back into service, because the engineers know that "retired" was a number on a page and the structure is still there. A fleet that keeps that hangar open for aluminum and closes it for people has its philosophy backwards. Nobody is a write-off at their allotment. They are an airframe in the desert waiting on a work order.

Damage tolerance is the posture this whole lab is built on, and the Insider Threat page says why in one line: you are the load and the crack. Assume the flaw is already there and already yours. Then the question stops being "am I okay?" — which safe-life can only answer with a date — and becomes "how fast is it growing, and when do I look?" That is a question with a schedule attached, and a schedule is something you can keep.

One honesty the engineers would insist on: damage tolerance is the more expensive philosophy. It asks for an instrument, a record, and an inspection you do not skip. The lab's answer to the cost is the 80/20 rule — one number a night, thirty seconds, and the analysis builds itself. Small burden, four-fold return, or the gauge does not ship.

This is a gauge, not a diagnosis. Nothing on this page generates a mental-health label or a prediction, and none of it is medical advice. It cannot see a crisis. If you are in one, the right instrument is a person: in the US, call or text 988, any time.

Fly it

The posture is a decision, and the program has a place to write it down.