Automotive

Ford cruise control fires: one unfused switch, 14 million vehicles, ten years

Case file #101·November 22, 2026·6 min read·analysis by Peter Stasko

Case file

  • What happened: A brake-pressure-activated cruise control deactivation switch supplied by Texas Instruments could leak brake fluid internally and overheat, igniting fires — including in parked vehicles with the ignition off, because the switch stayed electrically live from the battery.
  • Scale: More than 14 million vehicles recalled across roughly a decade of recurring campaigns, 1999 through 2009, spanning multiple Ford platforms. NHTSA linked the switch to hundreds of vehicle fires. One of the largest fire-related recall campaigns in U.S. automotive history.
  • Root cause: A low-cost single-point component carrying a safety function, wired unfused and unswitched — no redundant protective barrier between the battery and a part sitting in a hot, vibration-rich engine bay for its whole life.
  • The bill: Remedies evolved from fused wiring harnesses to full switch replacement, campaign after campaign. Direct recall costs ran to enormous sums; the reputational and regulatory cost of ten years of recurring recalls was larger still.

Here is an uncomfortable observation from two decades of fixing quality systems: the most expensive component failure I have ever studied cost almost nothing to design out at the drawing stage. One fuse. A few cents and one line on a wiring schematic stood between Ford and a decade of returning recalls. When a €2 part carries a safety function and has nothing standing behind it, you have not saved money on the part. You have financed ten years of CoPQ on an instalment plan.

14M+vehicles recalled, 1999–2009
10 yrsof recurring recall campaigns
1 fusethe missing protective barrier

The situation

The cruise control deactivation switch is an innocent-looking part. It sits on the brake master cylinder, senses brake pressure, and tells the cruise control to disengage when you touch the pedal. Not glamorous. Nobody wins engineering awards for it. And because it was electrically simple and cheap, it was treated — implicitly — as low-consequence.

Two design decisions turned that assumption lethal. The switch was wired hot: power straight from the battery, live at all times, ignition on or off. And the circuit had no fuse. So the switch held brake fluid under pressure on one side of its diaphragm and constant battery power on the other, sealed by a design expected to survive heat-cycled engine-bay conditions for fifteen years or more.

When the seal degraded and fluid migrated into the electrical side, the brake fluid — hygroscopic and conductive once contaminated — created a resistive path. Resistance means heat. Heat, fed by an unfused 12-volt supply that never sleeps, means fire. In a parked vehicle. In a garage. Attached to a house.

How it unfolded

What followed is the part every quality professional should study, because it was not one failure — it was the same failure, rediscovered repeatedly. Ford's early recalls addressed the symptom: dealers installed fused wiring harnesses inline with the switch. The harness added the missing barrier, and the campaign closed.

Then fires continued. Investigations continued. More platforms came into scope, and later campaigns replaced the switch itself rather than merely fusing it. From 1999 through 2009 the recall expanded and expanded again — different model years, different nameplates, the same root cause still sitting in the fleet. NHTSA linked the component to hundreds of fires over the years.

Ten years. That is what it looks like when containment is fielded as corrective action and the failure mechanism is never fully removed from the vehicle population.

Root-cause anatomy

Technically, this is a textbook single-point failure on a safety function. In PFMEA terms: occurrence may have scored low in early years, detection moderate, but severity — fire, property destruction, potential loss of life — is a 10. A severity 10 with no redundancy behind it is the one combination a design review must never wave through. The switch's location. Its duty cycle, live 24/7 for the vehicle's entire life. The known aggressiveness of brake fluid toward electrical materials. The missing fuse. Each is an individually defensible engineering choice; combined, they form an unprotected energy path waiting for a seal to age.

The organisational failure is subtler and, frankly, more instructive. The part was cheap, commodity-class, supplied by a major tier-one. Cheap components attract shallow scrutiny — nobody assigns their best failure-mode analyst to a cruise-control cut-out switch. The supplier-quality boundary was presumably clean on paper: the part met its specification. But the specification itself embedded the fatal assumptions — always-on power, no fuse, service life equal to vehicle life. And when field failures emerged, the remedy was additive (a harness) rather than subtractive (removing the failure mode). That is the signature of an organisation managing a recall rather than solving an engineering problem.

A recall is a confession with a part number; containment that satisfies the regulator is still not a corrective action.

Where the quality system failed

Name the discipline and this case breaks it.

  • PFMEA: a severity-10 failure mode (vehicle fire, ignition-off) with no second barrier. Either the PFMEA never scored it honestly, or it scored it and the recommended actions never carried the weight a 10 demands.
  • APQP / design review: an always-live, unfused circuit through a fluid-wetted component should have died in design verification. A single-point-failure review — standard practice in aerospace under AS9100 thinking — would have flagged it in an afternoon.
  • Change control: the fleet aged, conditions shifted, supplier process changes accumulated over a decade, and the recurring campaigns suggest the field-failure loop never forced the hard question: is the component design itself still valid?
  • CAPA gate: the deepest wound. A fused harness is D3 interim containment in 8D language — protect the customer now. Selling D3 as D5/D6 (permanent corrective action, implemented) left the root cause in the fleet, and the 8D loop reopened itself year after year, at a cost of billions in aggregate remedies and irrecoverable trust.

What would have caught it

None of it exotic.

  • A fuse. The cheapest answer in the whole file. Any overcurrent protection between battery and switch converts a fire into a blown fuse and a dead cruise control. Severity 10 becomes an inconvenience.
  • Switched power. Powering the circuit through the ignition removes the parked-car scenario entirely. Fires with the ignition off are what turned this from a quality problem into a national news story.
  • A single-point-failure review on every battery-direct circuit. One page per circuit: where does the energy go if this component fails resistive, short, or open? Any answer involving "heat, unfused" fails the review.
  • A severity-10 escalation rule. Any PFMEA line scoring 10 on a safety function triggers mandatory redundancy — a second barrier, a design change, or a documented executive waiver. No commodity exemptions.
  • An honest CAPA gate. Interim containment gets a fixed expiry date and a named owner for the permanent fix. If D3 ships, D5 must still close — or the problem formally reopens.

My take

I have lived the small version of this. Running QRQC and A3 problem-solving at Witte Automotive, and later building a greenfield quality organisation for 900-plus people at SNOP, I learned one rule the hard way: the moment a customer accepts your containment, your own urgency dies. The part that scared me was never the defect we were fighting. It was the containment everyone had stopped questioning. We drove a 70% reduction in defect costs at SNOP largely by refusing to let a harness-equivalent fix close an 8D. Containment protects the customer this week; only corrective action protects them in year ten.

Severity discipline is the other parallel. In my aerospace years and in IATF 16949 environments alike, the teams that wrote "10" in the severity column and then behaved as though it said "6" were the ones generating recurring escalations. Ford's file is that behaviour at national scale. A severity 10 is not a scoring convention. It is a design mandate for a second barrier.

What this means on your floor

  • A cheap part with a safety function deserves more scrutiny, not less. Component cost is not a proxy for consequence.
  • Every battery-direct, unfused circuit through a fluid-adjacent component is an open audit finding until proven otherwise.
  • Containment that satisfies the regulator is not corrective action. Give every D3 fix an expiry date and an owner for the permanent one.
  • Severity 10 means redundancy is mandatory — second barrier, redesign, or an executive signature accepting the risk in writing.

Fourteen million vehicles. Ten years. Hundreds of fires. All downstream of one engineering decision and one quality system that failed to challenge it. The fuse that was never designed in cost Ford more than any fuse in history. So when you are tempted to ship a barrier-free single point because the part is cheap and the failure seems implausible, remember this file. You are not saving the cost of the second barrier — you are deferring it, with interest, for a decade.

This case file analyses publicly documented events and reports. I had no involvement in the engagements described; company statements and official findings are matters of public record. The lessons and opinions are my own.

Peter Stasko

Peter Stasko

Corporate operator across automotive and aerospace — Airbus, SNOP and Witte Automotive. Building production AI hands-on since 2016.

Seeing the same failure mode on your floor?

Every case in this library ended as a headline. A one-hour conversation is cheaper. No slides, no pitch — just an experienced pair of eyes on your situation.

Book a conversation