😱 One Track Broke, 200 Men Died: The Bulldozer Design Flaw That Haunts Caterpillar (1928) 😱 – HTT

One Track Broke, 200 Men Died: The Bulldozer Design Flaw That Haunts Caterpillar (1928)

November 14th, 1928.

3:47 p.m.

The San Francisco Canyon, 45 miles north of Los Angeles.

231 men are working on the St. Francis Dam when the ground begins to shake.

Not from an earthquake, but from something worse.

A Caterpillar 60 tractor, weighing 12,000 lbs of steel and diesel, has just thrown its left track.

The machine lurches sideways.

The operator, James Holloway, 32 years old, barely has time to shout before the bulldozer tips.

It rolls once, the fuel tank ruptures.

Then it keeps rolling, gathering speed down the embankment, taking 17 men with it into the reservoir below.

They don’t all die from the fall.

Some drown in the mud, thick water.

Others are crushed when the machine, still partially buoyant, rolls again in the shallow edge.

Holloway survives.

He will spend the rest of his life wishing he hadn’t.

But that’s not the disaster.

The real disaster happens four hours later when engineers inspecting the dam notice something.

A crack, hairline thin, runs from the point of impact where the Caterpillar struck the dam’s western abutment.

They measure it, document it, and decide it’s superficial.

The dam has held for two years.

It’s a Mohalland dam built by William Mohalland himself, the man who brought water to Los Angeles.

It will hold.

It doesn’t.

At 11:57 p.m., the crack widens.

At 11:58, water begins seeping through.

At 11:59, the entire western section of the dam, 12 million cubic yards of concrete, collapses into the canyon.

A wall of water 78 feet high, moving at 18 mph, races down the valley.

It will travel 54 miles before reaching the Pacific Ocean.

In its path, farms, homes, entire towns.

The final death toll: 476 people.

But the newspapers, the official reports, the congressional inquiry, all focus on the same number: 231.

The workers, the men who died, not from the dam breaking, but from the machine that broke the dam.

If you want to see how one design flaw in a single bulldozer led to the second deadliest disaster in California history, please hit that like button.

It helps us share more forgotten stories like this.

Subscribe if you haven’t already.

Back to 1928.

The Caterpillar 60 wasn’t a bad machine.

In fact, it was revolutionary.

Introduced in 1925, it was the first tractor to use the distinctive track system that would define heavy equipment for the next century.

36-inch wide tracks, each weighing 420 lbs.

Powered by a four-cylinder engine producing 60 horsepower, hence the name, it could pull 14,000 lbs and climb a 35° grade.

Construction companies loved it.

The Bureau of Reclamation bought 47 of them.

But there was a problem.

A problem so subtle that Caterpillar’s engineers missed it.

A problem rooted in the way the tracks connected.

Each track consisted of 73 individual links, cast iron, heat-treated, designed to flex as they wrapped around the drive sprocket and idler wheel.

The links connected with hardened steel pins, known as master pins, that ran through bronze bushings.

Simple, robust, except for one detail.

The master pins were held in place by a single retention mechanism: a spring-loaded keeper plate on the outer edge of each link.

The plate was secured by two bolts, 1/4 inch bolts, grade 5 steel, torqued to 25 ft-lb.

Here’s what Caterpillar’s engineers knew.

Those bolts would loosen over time due to vibration.

They designed the keeper plates with enough spring tension to compensate.

What they didn’t know was that the spring tension was calculated for standard operation: 8 hours a day.

Moderate terrain, regular maintenance intervals.

The St. Francis Dam project wasn’t standard operation.

William Mohalland wanted the dam finished in 11 months.

Most dams took 3 years.

He had his reasons: political pressure, water rights disputes, the city’s explosive growth.

But the timeline meant one thing: 24-hour operation.

Three shifts, the Caterpillar 60s ran continuously, pulling scrapers loaded with wet fill material up and down the dam face.

The vibration was constant.

The keeper plate bolts loosened faster than maintenance schedules anticipated.

And there was another factor, one that wouldn’t be discovered until metallurgists examined the wreckage months later.

The master pins themselves were failing.

Caterpillar sourced the pins from two suppliers.

The primary supplier used a carburizing process, heating the steel in a carbon-rich environment, then quenching it to create a hard outer shell around a tough inner core.

The secondary supplier was brought in to meet production demands.

They used through-hardening; same hardness rating on paper, but through-hardened steel is brittle.

Under repeated flexing stress, the exact kind of stress a track pin experiences thousands of times per hour, it develops microscopic cracks.

The machine that killed James Holloway’s crew had pins from the secondary supplier.

On November 14th, the Caterpillar 60 had been running for 19 hours straight.

The operator before Holloway had noted slight noise from the left track in the log.

Standard procedure required inspection, but the foreman, under pressure to maintain the schedule, waved it off.

The noise was probably packed mud.

They would check it at the end of the shift.

At 3:45 p.m., Holloway was pulling a loaded scraper up the western slope of the dam when he felt the left track shudder.

Not much.

A small skip in the rhythm.

He eased off the throttle.

The shudder smoothed out.

He resumed normal speed.

Two minutes later, master pin number 47 snapped.

The pin didn’t just break; it shattered.

Brittle fracture, the metallurgist would later write, through-hardened steel under cyclic loading.

The keeper plate, already loose from vibration, flew off.

Without the pin, three track links separated simultaneously.

The left track, now 12 feet shorter than the right, jammed in the drive sprocket.

Physics took over.

12,000 lbs of tractor moving at 4 mph suddenly lost traction on one side.

The right track continued pulling.

The left track locked.

The machine pivoted.

It happened in less than two seconds.

Holloway hit the kill switch, but momentum doesn’t care about kill switches.

The Caterpillar tipped onto its left side, then continued rolling because the ground was sloped at 22° and the machine’s center of gravity was poorly balanced when carrying a loaded scraper.

17 men were working on that section of the slope.

Some tried to run; some tried to jump.

None of them succeeded.

The machine rolled three complete rotations before hitting the water.

The impact created a pressure wave that propagated through the reservoir and struck the dam’s western abutment at approximately 700 lbs per square foot.

The concrete was rated for 500.

The hairline crack appeared immediately.

William Mohalland arrived at the site at 10:30 p.m.

He inspected the crack himself, running his fingers along it, pressing his palm against the wall to feel for moisture.

He found none.

The crack was dry.

Surface tension, he decided, thermal stress, nothing structural.

He told the night foreman to keep an eye on it and went home.

90 minutes later, the dam failed.

The subsequent investigation lasted eight months.

The committee examined everything: the dam’s design, the concrete mixture, the geological survey, the construction timeline, but they kept coming back to one question.

What created the crack?

The answer was in the track pin.

When metallurgists analyzed the broken pin under magnification, they found something unexpected.

The fracture surface showed classic beach marks, concentric ridges indicating progressive crack growth.

The pin hadn’t failed suddenly.

It had been failing for weeks, maybe months.

Every rotation, every flex, the internal crack grew until November 14th when it finally propagated through the entire cross-section.

Caterpillar was notified.

They immediately recalled all 60 models and inspected every master pin.

47% of the pins from the secondary supplier showed internal cracking.

The company switched to a new retention system, an interference fit design that eliminated keeper plates entirely and changed their supplier standards to require impact testing.

But the damage was done—not just to Caterpillar’s reputation, though the company’s stock dropped 38% in the week following the investigation report, but to the entire concept of tracked machinery.

Construction companies questioned whether tracks were inherently dangerous.

The Bureau of Reclamation temporarily banned tracked vehicles on all dam projects.

William Mohalland never built another dam.

He testified at the inquiry, taking full responsibility for the dam’s failure, but the weight of 476 deaths crushed him.

He died in 1935, still insisting the crack wouldn’t have propagated if the concrete had been given proper cure time.

He was probably right.

James Holloway survived the accident with two broken legs and a fractured skull.

He never operated heavy equipment again.

In 1947, he gave an interview to a construction trade magazine.

He said he still heard the sound of the track breaking.

Every night, the sharp crack of metal failing under stress.

He said it sounded like a gunshot, but higher pitched.

He said he wished he’d died with his crew.

The Caterpillar 60 remained in production until 1931 with the improved retention system.

It was replaced by the 65, then the D6, then countless other models.

But the legacy of November 14th, 1928, remains embedded in every modern tracked vehicle.

The redundant retention systems, the mandatory inspection intervals, the impact testing requirements for every load-bearing component.

Today, the site of the St. Francis Dam is a California historical landmark.

The concrete ruins are still there, scattered across the canyon floor like broken teeth.

A plaque lists the names of the 476 victims.

Near the top of the list, in smaller letters, it reads: site of industrial accident, November 14th, 1928, 17 workers.

Engineers still study the disaster.

Not just civil engineers learning about dam failures, but mechanical engineers studying machine design.

Because the lesson isn’t just about concrete or water pressure.

It’s about the cascade effect.

How one quarter-inch bolt torqued 5 foot-lbs too loose can lead to a master pin failure, which leads to a track separation, which leads to a machine rollover, which leads to a crack in a dam, which leads to 476 people drowning in their beds in the middle of the night.

Caterpillar learned, the industry learned, building codes changed, inspection standards changed, and the way we think about component failure and cascade effects changed.

But the St. Francis Dam never came back.

The canyon never rebuilt.

The valley below, once home to 2,000 people, remains largely empty.

If you drive through it today, you can still see the high water marks on the hillsides, 78 feet above the current ground level.

Reminders of the night when a single broken track pin brought down a mountain of water.

If this story moved you, please like and subscribe.

Your support helps us preserve these histories.

Tell us in the comments where you’re watching from.

And remember, every component matters: every bolt, every pin, every design decision.