COMET Chapter Seven - Old Accidents

 

I invite you to watch as the brazen brown and yellow aircraft designated “dash 80” slowly begins it's takeoff roll down runway 15/33. The four Pratt and Whitney turbojet engines, individually suspended below the 35 degree sweptback wings, roar as they produce 44,000 pounds of thrust. 

At 120 miles per hour pilot Alvin “Tex” Johnson firmly pulls back on the control column, and 200,000 pounds of aluminum alloy, wires, rubber tubing, kerosene  and ambitions float off the asphalt. It is 2:14 on Thursday afternoon of 15 July, 1954. The air above Lake Seattle is populated with puffy white clouds. And as the twin four-wheeled bogie tricycle gear of Dash 80 fold neatly into the underbelly, the grounded Comet jet transport becomes obsolete. 

When he landed, 2 ½ hours later, “Tex” said, “She flew like a bird. Only faster.”

Douglas Aviation had dominated the commercial aviation market since 1933, with their DC 3 (above) family of piston multi-engine passenger planes. 

Their competitor Boeing survived thanks to their military contracts -  beginning with 17,000 B-17's built between 1935 and 1945.

This was followed by almost 4,000 B-29 SuperFortereses built between 1942 and 1945.

Then Boeing built 2,000 swept wing B-47 Strato-jet bombers between 1948 and 1963.

 Finally, beginning in 1951, Boeing supplied the U.S.A.F with 744 swept wing B-52 StratoFortress, still flying more than 70 years later.

So, when Bill Allen, president of Boeing Aircraft Company, saw the de Havilland Comet at the 1949 Farnborough Air Show (above), he was not impressed. But what the Comet high lighted to the Boeing engineers was that jet transports promised speed and reliability to carry anything you could fit in the pressure hull.

The United States Army Air Corps had been experimenting with mid-air refueling since 1927. Developments were slow, but by 1948 the USAF had two squadrons of beefed up double body B-29's Tankers, which Boeing initially called the 367s and which the U.S. Army relabeled the B-50 (above)

The problem was the jet bombers could not comfortably fly slow enough to be serviced by these piston driven flying gas tanks.  Also, at higher altitudes where the B-50's labored, the air was “smoother”, making refueling easier. Obviously the Air Force was going to need a jet powered tanker. And that was Boeing's initial justification to nervous investors when, in 1952 Allen asked them to risk 25% of Boeing's total capital, some $16 million, to developing a jet tanker. But carrying fuel was only part of Boeing's great plan.

Boeing labeled their new aircraft Project 367-80. Eventually it became known simply as the Dash 80. It was big - 128 feet long as opposed to the 93 foot long Comet – 130 foot wingspan to 115 feet for the Comet, and a wing area of 2,400 square feet to 2,015 square feet for the Comet. All that extra wing space, devoted entirely to fuel, gave the Dash 80 a range of 3,530 miles to the Comet's 1,500.

The plane was so big a passenger version was projected to carry at least 140 seats, five abreast, compared to the Comet's 43 seats at two abreast, thus reducing the operating cost to 25 cents per seat-mile for every gallon of kerosene the four engines burned. Not to mention, the Dash 80 could cruise 100 mile per hour faster than the Comet. It looked like democracy with wings.

Boeing swept the wings of the Dash 80 back to 35 degrees, which they knew would be stable because that was the same angle as the wings on their B-26 and B-52 jet  bombers. And they avoided new engine development by using the same engines used in the bombers, and slung them beneath the wings for easy maintenance, and to free up wing space for fuel, just like the bombers. All of this would reduce the need to retool when and if the various versions of the passenger plane went into production.

Boeing also learned from the well publicized crashes on Comet take offs by designing forward and rear facing extensions (flaps, tabs, ailerons and spoilers) on the swept wings of the Dash 80 (above). These allowed the big bird to stay in the air at speeds as low as 80 miles per hour. 

To allow the passenger version to use existing airfields of 7,000 feet, bucket thrust reversers were included, to slow the jet from the landing speed of 150 miles per hour to a dead stop within 6,000 feet.

However, the Dash 80 was neither a tanker nor a passenger plane. It was a test bed for both. That did not matter, it seemed, because almost before the 2 ½ hour maiden flight had landed, the Army ordered 29 of the new, as yet  un-built planes to be labeled the “K” (meaning tanker) and “C” (meaning transport) -135 (above).  Another 250 KC-135's were quickly added to the order, the planes first reaching service in August of 1955.

The airlines, however, showed little interest, in part because 1954 was a recession year, but also because the disasters of the Comet were still fresh in the public mind. Few seemed eager to risk their lives on a passenger jet. So the Dash 80 flew on, amassing data to improve the design.

Meanwhile, on Tuesday, 1 February, 1955, the the British Civil Aircraft Court of Inquiry into the crashes of Comet Yoke Peter and Yoke Yoke was issued by the Royal Aircraft Establishment. The fault, they had determined, was “...metal fatigue, caused by the repeated pressurization and de-pressurization exacerbated by the thin aluminum alloy skin...” and the squared off windows which intensified pressures at the corners.

De Havilland responded with a public statement. "Now that the danger of high level fatigue in pressure cabins has been generally appreciated, de Havillands will take adequate measures...we propose to use thicker gauge materials...and to strengthen and redesign windows and cut outs and so lower the general stress to a level...(which) will not constitute a danger.” The company immediately began the work, but it would be 3 years before the redesigned Comet 4 could re-enter commercial service.

It was not until 2015, when the 50 years of silence required by the British government Secrets Act had expired that the full truth of the Comet hull failures was revealed. Said the originally unredacted report, “...metal fatigue, attributed to raised stress at the squared-off window corners, actually had another cause....the structure had been designed to be bonded – glued, in fact – by...the Redux process...” 

However, “...During production...de Havilland chief designer, R.E. Bishop (above)... decided that these areas should...be reinforced...by normal aircraft riveting...It was this 'belt and braces' riveting...that caused the failures. The cracks emanated from the rivet holes in the corner area – not from the material in the corner structure itself.”

But it was Tex Johnson (above), the Boeing test pilot,  who drove the final nail in the Comet coffin. 

The stage was the annual Gold Cup hydroplane races to be held on Saturday 6 August, 1955 - light high speed boats powered by air craft engines, racing at 80 to 90 miles an hour across the surface of the water and throwing 30 foot high rooster tails behind them. Viewed by perhaps 200,000 spectators from the bluffs above Lake Washington, in 1955 for the first time the event was broadcast on live television.

That same week, The International Air Transport Association and the Society of Aeronautical Engineers were both holding their conventions in Seattle. So Boeing's Bill Allen invited a large number of aviation industry folks to attend the races. He had also coordinated with the Dash – 80 team to do a fly by.  That was all Tex Johnson was supposed to do - fly by.

But Tex had heard that Douglas aircraft, which had started a crash program to build their own slighter smaller and slightly slower jet passenger plane, the DC 8 (above), was telling potential customers that the Boeing jet was unstable. 

Tex (above left) felt obliged to prove the critics wrong.  As the Dash 80 was in route to Lake Washington that morning, he told his co—pilot Jim Gannet (above, right) , “Hey Jim, I'm going to roll this airplane over the Gold Cup." Gannet suggested if he did, Jim Allen would fire him.

Johnson then steered the big jet down to 500 feet for the east bound fly-by.  And as he passed the spectators, Tex pulled up slightly and slipped the aircraft into a gentle roll to the right, 360 degrees - a perfect barrel roll. (Barrel Roll).  It was perfectly safe, according to “Tex”. “The airplane does not recognize attitude,” he later explained, “providing a maneuver is conducted at one G...The barrel roll is a one G maneuver and quite impressive, but the airplane never knows it’s inverted.” 

Then, on the west bound pass, he did it again.  During the second roll the flight engineer, the only other person on board, snapped a photo of the Bill Allen's $16 million gamble upside down less than a thousand feet above the city of Seattle (above).

The gamble worked. Less than 2 months later, on 13 October, 1955, Pan American World Airlines ordered 20 of the newly designated Boeing 707 jets, to replace their cancelled Comet orders. 

The only fly in the ointment landed when Douglas upgraded the engines on their DC-8's, forcing Boeing to follow suit. That delay prevented the first 707 commercial flight until October of 1958. But by 1956 even British Overseas Airway Corporation had ordered Boeing's big jet. Until the production lines shut down in 1978, Boeing built 865 various versions of the 707.

The British government remained loyal to the de Havilland Comet, and in March of 1955 British Overseas Aviation Corporation ordered 19 of the new Comet 4's (above). To extend its range, the Comet 4  had a fuel pod perched in each wing, and the wings themselves were bigger, as were the engines. 

On 4 October, 1958 it was a Comet 4 that flew the first London to New York jet flight, with a westbound refueling stop at Gander, Newfoundland. But the new Comets could only squeeze 99 passengers into the larger pressure cabin, and de Havilland's plane was still 50 miles per hour slower than either the Boeing 707 or the DC-8.  De Havilland sold only some 76 Comets, before production was stopped in 1964.

In 1960 de Havilland itself was acquired by Hawker Siddeley in a government brokered sale and the name de Havilland faded from British aviation. 

The Canadian government bought the subsidiary de Havilland Canada, and for the next 20 years produced a successful line of short take off and landing civilian aircraft. But in 1984 the conservative government of Brian Mulroney privatized the company, and in 1986 de Havilland Canada was bought by Boeing. Despite promises to Canada, Boeing closed the plant and broke up the jigs and all production equipment, and that really was the end of de Havilland.

Today, (2021) Boeing is stull reeling from their own self-engineered 737 Super Max debacle - 2 crashes and 346 dead, caused by a preventable failure  - followed by the COVID 19 pandemic which gutted air travel for two years.  

Boeing in 2022 seems to be facing a similar fate to the one de Havilland faced in 1954, building airplanes nobody wants to fly aboard. 

As one of the Farnborough engineers pointed out back in 1954, ‘There are rarely new accidents, just old accidents waiting for new people to have them.”

- 30 -

COMET - Chapter Three - Spirits Of The Dead

 

I ask you to witness the sad little fleet dock in the port of Azzuro (above), on Elba's eastern shore. It is some 7 hours since Yoke Peter fell into the cold Tyrryenian Sea - 10 January, 1954. From the trawler Francesco Guiseppe they carry the pale corpse of a 10 or 11 year old boy, strapped atop a wooden plank and covered by a fresh white sheet. The somber fishermen then carry off the body of a 29 year old stewardess, Jean Evelyn Clark.

She is followed by a procession of 12 more shroud draped bodies up the dirt road and into the stone fortress of Chiesa di San Giacomo Maggiore (above). 

In the fort's small, austere, baroque chapel the burdens are laid before the altar. 

Candles are lighted and prayers are intoned. And if the spirits of the dead should have awakened, they must have been surprised to find themselves in such a dark sanctuary, such a long way from the cold thin air so close to heaven.

 Because de Havilland had built Britain's second jet fighter, the single engine Vampire (which went into service in April of 1945), they believed they were uniquely qualified to build the Trans-Atlantic passenger jets envisioned by the Brabazon Committee. However these would require a range of over 3,000 miles. But the second generation Ghost engines burned 3 – 4 times more fuel at 10,000 feet than conventional internal combustion engines, putting New York well beyond their range. So the cautious de Havilland company decided to approach trans-Atlantic flights in stages.

The Comet 1 would establish profitable routes tying the the British empire together, with flights between London and Singapore or South Africa. 

These largely overland routes would allow refueling stops every thousand miles or so, over well established flight paths, using already established ground facilities. Once the Comet was proven and the three times more powerful Rolls-Royce Avon engines had completed their testing, de Havilland would build the larger transcontinental Comet 2, and then the even larger Comet 3. So from the beginning the Comet 1 and it's Ghost engines were envisioned as test beds.

Throughout 1947 and 1948 de Havilland subjected the Comet fuselage and its over-sized square windows (above)  to twice anticipated pressures, through 16,000 cycles from sea level to 36,000 feet and back, simulating over 40,000 service hours. 

Since each rivet holding the Comet's aluminum skin required a hole be punched, the decision was to use the adhesive “Redux”, proven in combat on the plywood fighter/bomber, the Mosquito. It would be used on the Comet primarily “...to affix stiffening...to wing and fuselage panels, the resulting panel being both stronger and lighter than a riveted structure.” After all this was tested and proven, the design was finalized and built.

Not until after three years of testing did the Comet fly. The test pilot John Cunningham described this marvel of modern technology as “Very promising. Very quick.” British Overseas Aircraft Company bought 8 Comet 1's. 

Air France  (above) bought a pair,  South African Airways bought another two, as did Canadian Pacific Airways.  And the American giant, Pan American World Airways, signed up to buy three of the Comet 3s, with an option to buy 7 more the following year. With each Comet sold abroad representing over half a million pounds in badly needed foreign currency, and with “eight BOAC Comets leaving London each week: three to Johannesburg, two to Tokyo, two to Singapore and one to Colombo”, by the summer of 1953 it seemed the Comet was indeed a revolution in passenger travel.

However there were some annoying problems. The vacuum tubes in the navigation system tended to overheat. The cockpit's windows would occasionally fog over. And the seals on the hydraulic control system, which powered the flaps and tail rudder, were faulty on the first four Comets. Crews solved this problem by carry extra cans of fluid, and would top off the system whenever they spotted fluid on the tarmac.

And then there those under powered Ghost engines. In late October 1952 a BOAC Comet taking off from Rome, Yoke Zebra, failed to reach V-2, the speed at which pilots would lift the nose to become airborne. Unable to lift up or stop, Yoke Zebra overran the runway and broke the landing gear (above).  Pilot Peter Duffrey, who had flown the Comet into Rome,  found Yoke Zebra, “... just sitting there hissing – it (had) ripped the belly tank out and there was fuel everywhere.” Luckily there was no fire and no loss of life.  

But Duffrey diagnosed the problem immediately. The pilot had raised the nose too high, and stalled the aircraft. But the pilot was not helped by the hydraulic system. Duffrey noted, “...the lack of feedback on the controls...allowed you to pull-back as much as you wanted.” That made it easy to stall the aircraft.

That mistake was repeated on 3 March 1953 during the delivery flight of a Canadian Pacific Comet, this time just after lifting off from Calcutta. This time all six crew members and the six passengers on board were killed. A week later a Comet test flight out of Karachi, Pakistan, stalled after lift off and slammed into a stone bridge, killing 11 technicians and crew. This crash cost de Havilland the second Comet the Canadians had ordered but now refused to accept. 

Then, in June, it happened again, this time in Dakar, Senegal. Said Duffrey, “As pilots we were becoming very concerned.” In response, new take-off instructions were issued to pilots, and in all future Comets modifications would be made to the wings, and all future Comets would be fitted with weather radars.

Then had come the Yoke Victor tragedy over Jagalgori, India (above). It had been blamed on the weather, but pilot error had not been ruled out. Nor had sabotage. There were a lot of people in Asia and Africa who had an grudge with the British Empire. And a lot of older pilots were suspicious about those new hydraulic controls. And now there was this new disaster over the Tyrryenian Sea. The negative publicity was certain to kill desperately needed sales. 

With 24 hours of prodding from Arnold Alexander Hall (above), Director of the Royal Aircraft Establishment, both de Havilland and BOAC agreed to temporarily ground all Comets. Hall then opened a comprehensive investigation, which began by looking for nefarious causes and actors.

Admiral Louis Francis Mountbatten (above), 1^st Earl of Burma and Commander of the British Mediterranean Fleet, was ordered to recover as much of Yoke Peter as possible. 

However the Comet tragedy coincided with a series of devastating 7.0 and above earthquakes in Greece, which followed the ruinous Greek Civil War. Mountbatten was also the NATO commander in the Med and in that role, the fragile state of Greek politics and economics made relief efforts the priority. 

Although he had to continuously shift his resources, Mountbatten still managed to maintain at least one frigate with SONAR and at least one British minesweeper on station in a grid pattern search for debris of Yoke Peter, directing 5 local fishing trawlers dragging the bottom.

It was not easy. One engineer, viewing the search through a primitive video system, described the ocean bottom as looking “as though someone had upended a waste-paper basket”. Still, by 15 March – 65 days after the crash – the Royal Navy had recovered a large section of the front of the fuselage and part of the wings. Four days later they had recovered all 4 Ghost engines. If anyone had noticed the similarity in the way Yoke Peter had broken apart in January and the the pattern Yoke Victor had fallen from the sky 7 months earlier, they were not talking. Not yet.

- 30 -