The results of the Y1B-17A tests heightened GHQ Air Force's enthusiasm for the potential of high-altitude bombing, and so the addition of turbo-supercharging to new B-l7 production models was approved. When testing was complete, the Y1B-17A was renamed the sole example of the B-17A.
Changes were still needed however. The nose glazing with its strangely-shaped 'turret' for the single Browning machine gun was deemed impractical. Indeed, the United Kingdom's then Air Commodore Arthur Travis 'Bomber' Harris - who was visiting the USA on a purchasing mission, and who was never one not to speak his mind to all and sundry - commented in a report after he had inspected a Y1B-17 that the device would seem to be '... More appropriately located in an amusement park than a war aeroplane'. There was also a need for better high-altitude performance and internal changes to give the navigator and other crew members an ergonomically better working environment.
The B-17C started out as Boeing's Model 229H design study, specifically aimed at improving the defensive armament. The teardrop-shaped gun cupolas had been found to have restrictive fields of fire and were difficult to use and even harder to manufacture. They were eliminated, and in their place, simple removable windows allowed post-mounted machine guns more flexibility. The lower teardrop was replaced with a more extended structure, often likened to a bathtub, which allowed a kneeling gunner to operate a Browning machine against low attack from the rear. All the guns were now the heavier .50 in calibre Brownings, except the .30 nose weapon, for which two more sockets were added in the framing so that the bombardier could re-position the gun to increase his field of fire.
There were detail changes to fuel, electrical and oxygen systems and further brake improvements. The B-17C had upgraded Cyclone engines and improved turbo-superchargers, raising the high-altitude top speed during one service test to 323 mph. Supercharger failure due to inadequate lubrication had been encountered in the B-17Bs, with all units replaced early in 1940. These raised the B-17 B's performance, but there is no doubt that the B-17C's improvement in that respect also resulted from the reduced drag occasioned by the removal of gun cupolas. Deliveries were made between August and November 1940, but owing to the volume of additional changes required on the last 42 aircraft of the contract, the Air Corps bestowed a new designation, B-17D, although the Boeing model identity remained the same.
But that was still in the future. The day after Boeing and the Air Corps talked about the superbomber, Belgium surrendered. An alarmed United States heard its President ask Congress for an immediate allocation of over one billion 1940 dollars for new weapons.
Representatives from Boeing flew back to Ohio to meet with Colonel Oliver Echols to discuss drastic changes in the Flying Fortress. The first B-17C had yet to fly and the D model was way behind that. Echols was only too aware of what was happening in Europe, and what must inevitably engulf the United States. He was looking ahead and told Boeing: 'Figure out how to get more guns in those Forts. Tail guns especially. And armor plate to protect the crew'.
Boeing's engineers were almost disbelieving of the new requirements. They were being asked to carry out a massive redesign of the Flying Fortress. Tail guns? It would make mincemeat out of the centre of gravity of the aircraft. It would add weight, complexity, and problems. It would, they warned, make a mockery of the steadily improving performance of the machine.
In the meantime, the B-17C had to be developed. A program was set up in December 1940 in cooperation with General Electric Company to obtain flight test data in order to determine the critical altitude performance of the turbo supercharger using B-17C 40-2043. A standard GE type B2 turbo supercharger with factory-installed instrumentation was fitted on the No.4 engine nacelle and a series of level test runs at 15,000, 20,000, 25,000, 30,000 and 35,000 feet were planned. Boeing's Preliminary Report D-3422 dated 26 August 1941 was not only for the B-17 - as item 4, in conclusion, stated, 'The study as to the applicability of the date to the B29 is not sufficiently complete to state conclusions at this time'. As these flights progressed, the company realized it was obtaining inconsistent data and variable results. The tests were not helped the numerous delays brought about as a result of the severity of high-altitude operation.
In order to try and obtain more consistent data, a revised programme was devised and a further series of flights were conducted between 31 July and 15 August, but again the program was hit by mechanical difficulties and bad weather. Boeing discovered that the critical altitude tests revealed that the turbo-installation in the No.4 nacelle of the B-17C test aircraft developed a limiting Brake Mean Effective Pressure (BMEP - the medium pressure in a cylinder during it's combustion cycle) at much higher altitudes than was previously thought. It seems the power drop observed above the critical altitude was excessive which made high altitude cruising questionable.
The report concluded that altitude performance of the No.4 nacelle in the B-17C was considerably better that had been previously established. Unstable and surging conditions had been encountered which required further investigation and correction to realize the full potential of added altitude performance. Boeing stated that 'the critical altitude tests have resulted in opening up many important new phases of turbo operation for further consideration'. Translated into simple English, this meant that there was a whole lot more work to be done until they fully understood what was going on.
The incorrectly marked AN528 - it is painted AM528 - is seen here over the USA before crossing the Atlantic. (USAAC)
With the fall of France and the critical situation that ensued during the summer of 1940 when Britain stood alone, positive interest in the Flying Fortress was revived. The political atmosphere in the USA had changed with the Administration's open support for the British cause, and a massive expansion of America's own military forces was planned. Delivery of further Flying Fortresses, the new B-17B model, during 1939-1940, followed by the much-improved B-17C, had radically changed the position for the USAAC. To make mention of its doctrine of daylight strategic bombardment was no longer heresy.
Anxious to obtain any worthy US combat aircraft, the British showed interest in the B-l 7, and although they were critical of its small bomb load and defensive armament, they knew that if made available, it could be usefully employed. With a British order in mind, Boeing had explored the RAF's requirements under a design study labelled Model 299U. However, the Army Air Corps, soon to be the Army Air Forces, had no intention of losing production aircraft to a foreign power when it was intent on building its force of heavy bombers.
Here national politics entered the picture. Through Boeing's and the Air Corps' persistent courtship of the news media for their own purposes, the Flying Fortress had become well known abroad as the bright star in America's air arsenal. The British, eager to draw on United States aid, were appreciative that the Roosevelt administration, keen to justify the enormous investment in the nation's air arms, saw the use of the Flying Fortress by the British as a valuable means to this end. The rhetoric of Generals and other enthusiasts aside, the Air Corps remained doubtful as to how the current model B-17 would fare against modern fighters such as the British and Germans had in service.
The RAF, while still critical of the bomber's armament and other equipment, was cognisant of the mode of attack propounded by the USAAC theorists; daylight precision attack from a very high altitude where the effectiveness of enemy anti-aircraft fire and fighter interception would be minimized. The first few months of war had shown the RAF that, in general, heavy bombers were too vulnerable to enemy interception to operate effectively in daylight. The Air Corps' concept offered a new approach to day operations, and the RAF could not afford to neglect any possibility.
Approaches were made for acquiring new-model Flying Fortresses, but the Air Corps were reluctant to lose precious bombers from their own expansion scheme or risk their techniques being aborted by any precipitate action which might bring repercussions on their own efforts to establish a strategic bombing force. The British met with a more sympathetic ear in the US Government. In late September 1940 Secretary of War Henry Stimson promised, among other things, to furnish the British with five Flying Fortresses. However, with an election pending, the US administration thought it wise to defer release due to 'political difficulties' but, with the November elections over, twenty B-17Cs were released to the RAF and allotted the serial numbers AN518-537.
These aircraft, having no self-sealing fuel tanks, were returned to Boeing for Goodrich self-seal units to be installed. As the B-17D model, then in production at Seattle, had priority for these tanks, some delay resulted with a number of aircraft being placed in storage at McCord Field, Tacoma WA until modifications could be done. Other work was going on the aircraft destined for the RAF at McClellan Field, Sacramento, CA so the first four B-17Cs were not ready for movement to the UK until April 1941.
Certain pro-British members of the US Government were impatient to see the Fortresses in action and Averell Harriman, the US Ambassador, explained to the British 'that the question of using the Flying Fortress operationally had become a political question of some importance', particularly as it had been well reported that these large bombers had been sent to Britain.
This appears to have been in direct contradiction to a claim which surfaced postwar which was that it had been understood by the United States Government that the Fortress I (the B-17C) would never be sent into combat, but was to be employed for training flight and ground crews to gain experience with the new bombers. It was stated that when the RAF squadrons were fully familiar with the aircraft operation, Fortress II models (the B-17E) would be provided under the lend-lease agreements. With the new Fortress II, heavily armed and fitted out with equipment designed from the lessons of air combat in Europe, the RAF would have the nucleus of a new and powerful daylight bombing force. It seems that members of the Air Corps and engineers from Boeing went to great pains to warn the British against sending their Fortress I into combat and then directly blame them for the debacle that followed. One unnamed Air Corps officer is supposed to have said 'I was in England when the planes arrived. We explained to the British our doctrine for the use of the planes. We told them that the crews had to be well trained, that a crew should drop two hundred practice bombs before attacking a real target; that the planes were designed to fly in formation for protective purposes and that by using them as trainers, trained crews could be ready to operate the new, properly equipped Fortresses when we delivered them. For some reason, they decided to use the planes offensively'.
Inside the bomb bay of an RAF B-17C looking forward towards the cockpit. The extremely narrow walkway that connected the front of the aircraft to the rear is very noticeable, as are the bomb suspension beams. The curved ribbed panels that make up the ‘floor’ are in fact, the aircraft’s closed bomb bay doors. (USAAC)
Perhaps this was an attempt at re-writing history in an attempt to disguise just how lacking the B-17C was, or perhaps there was confusion as to how the B-17C should be operated. Whatever the case, at the time American interests did not appreciate the dilemma of priorities within the hard-pressed RAF but these proddings at a high level from both US and British sources did have the effect of accelerating supply and training.
As the Royal Air Force were to be the first to take the B-17 into combat - and therefore highlighted a whole raft of shortcomings to the design that was at least partially solved in later models - it therefore becomes vitally important to study what happened by those directly involved. For this we need to go to the Operational Record Books - the RAF Form 540s - in which anonymous squadron diarists recorded the day-to-day events as they happened.
Above: The cockpit. Below: The nose compartment
Designated 'Fortress I' the first reached squadron service at RAF Watton in Norfolk during May. The flying crews selected were to be under 24 years of age and were chosen for fitness and 'slim build', the latter requirement due to the infamous narrow 'passageway' across the bomb bay from the rear of the aircraft to the cockpit! Each flying crew member had to undergo - without any ill-effects - a four-hour simulated 35,000 feet' flight' in a pressure chamber, breathing oxygen.
Forced to part with some of their prized bombers, the USAAC then became anxious to advise in their use and assist in training RAF personnel. The RAF, while accepting all practical aid and politely acknowledging the Air Corps plans for employment, preferred to operate using their own practical experience rather than strictly adhere to Air Corps theory. The USAAF provided a number of flying personnel and civilian technicians who were engaged in training the RAF crews converting to the Fortress I.
By early 1941 both the RAF and Luftwaffe had resorted to night operations for their medium and heavy bombers, generally leaving daylight operations to fast light bombers that stood a better chance of evading intercepting fighters, or operating in bad weather when extensive cloud gave protective cover. With Fortress I in British service, RAF Bomber Command planners proposed to try a new approach in evading enemy opposition by the sheer altitude of the attack.
At this time sub-stratospheric flight was rare and almost unknown in military operations. The turbo-superchargers on the Fortress's Cyclone engines enabled maximum performance to be obtained at 25,000 feet, some 10,000 feet more than the normal operational altitude of British bombers at that time. However, to afford maximum protection, the RAF proposed to operate Fortresses as near to their combat-loaded ceiling as possible, an optimum of 32,000 feet eventually being considered as desirable. It was assumed that enemy fighters would have great difficulty reaching this altitude and locating the bomber, and the aircraft would be fairly safe from anti-aircraft fire. If the American high-altitude bomb sights proved their worth, then targets could be hit with some measure of accuracy and such operations could be of inestimable value. This was basically the Army Air Corps strategic concept, with the exception that the RAF intended to operate at some 7-10,000 feet higher than the Americans.
The temporary US Army olive drab paint applied to the upper surfaces and neutral gray to the lower faces was the washable distemper type, and much disappeared during the trans-Atlantic trip. Both British and US crews were used for delivering Fortresses via Newfoundland to Prestwick. (USAAC)
Despite neutrality, the supply of war equipment to Britain under the Lend-Lease agreement was public governmental policy and much was made of this by both British and American news agencies. The Americans were anxious to impress upon the enemy how deeply they were involved. Although the Air Corps would have preferred not to have its concept of high altitude bombing with the Fortress put to the test by the RAF, many factions in the US administration were anxious to bring the US aircraft industry to the forefront. No sooner had the first Fortress arrived than pressure was placed on Bomber Command to commit the aircraft to battle.
Aware of the political implications of getting the much-vaunted Flying Fortress into battle, RAF Bomber Command decided to accept the bombers with a minimum amount of modification. Other aircraft types purchased from America were normally processed through Burtonwood, Lancashire, where standard British radio and signals equipment and identification lights were installed. It had also been intended to adapt the bomb racks to hold British bombs, but this was evidently abandoned with the availability of American 1,000 pound high explosive types in the UK.
AN521 went to Burtonwood for some modifications on 7 May, and AN534 arrived, giving 90 Sqn their first aircraft, thus allowing the squadron to start conversion training, with Capt Connally USAAC converting RAF pilots. AN529 arrived on 11 May, thus doubling squadron strength.
With so few aircraft, plus the need to replace or install equipment to RAF standards it is not surprising that there was never more than around twelve machines in service at any one time. There were oxygen system deficiencies and the electrically heated clothing was neither efficient or reliable when flying at 30,000 feet - and the guns regularly froze up.
Brigadier General Millard Harmon, a US observer of the air war in Europe, was acquainted with the RAF's Bomber Command opinions of the Fortress at this time and is quoted as saying: '...we are more than satisfied with performance, layout and crew comfort. The only point on which we are seriously doubtful is the efficiency of the defensive armament... experience will show.'
On 6 June 1941 British Prime Minister Winston Churchill visited RAF West Raynham, Norfolk, to review the latest RAF aircraft and equipment. USAAC officers Major James Walsh and Captain James Connally demonstrated what is believed to be Fortress I AN522 for him.(USAAC)
Training of the air crews was difficult to hurry due to the specialized nature of their mission. The twenty crews had to be carefully selected to stand the physical demands of operating at what were then incredibly high altitudes. It was also important that none was over twenty-four years old and that all had operational experience on four-engined aircraft. Each man was required to undergo a test in the decompression chamber at Farnborough, spending four hours at a simulated 32-36,000 feet. This was to eliminate those susceptible to 'bends' - pains caused by bubbles in the blood and other body fluids and tissues resulting from the decreased barometric pressure at high altitude. These varied from dull aches in leg or arm joints to crippling agony and the intensity varied with individuals. Chest constrictions or itching and pricking of the skin were other symptoms. Many men were eliminated through their susceptibility to bends and 60 per cent of the prospective crews were rejected on medical grounds.
.jpg)
One completely unforeseen local problem was the danger of flints on grass airfields piercing tyres of these heavy machines with disastrous consequences if this should occur on take-off. Because of this hazard at Watton an attempt had been made to operate from its nearby satellite at Bodney but one alarming landing was enough to convince Major Walsh and Wg Cdr MacDougall that the surface there was far too undulating for operating Fortresses safely. A move was made to nearby Great Massingham where the surface seemed better, although regular inspections were carried out to locate stones. This was but a temporary move until Polebrook in Northamptonshire, the most advanced of the many new airfields in the area with hard surfaced runways, was available.
Two days later, on 6 June, Winston Churchill came to see things for himself, with Sir Richard Pierse, Sholto Douglas, the AOC 2 Group and General Royce of the USAAC. Walsh and Connally gave a flying demonstration, but the Americans urged caution when the question of flying the aircraft operationally was raised. The British, with the experience of a number of years of wartime flying behind them, considered themselves better able to judge this aspect, and felt that operations were now feasible.
When the weather allowed, further test flying occurred, with Fortresses climbing to around 32,000 feet, giving crews a taste of things to come and showing up faults in equipment. Oil leaks were already a nuisance and were to continue to be. Finally, on 21 June practice bombing began on Watton's range at Stanford Warren.
The tumbling tail section fell a further 12,000 feet before Steward was able to free himself and fall clear. He was the only man on the Fortress able to parachute to safety, landing not far from the wreckage at Catterick Bridge in Yorkshire. Lt Bradley was the first airman of the US Army Air Force - which had changed from USAAC to USAAF in June 1941 - to be killed on active service in World War Two.
AN519 and '526 went to Burtonwood for the fitment of long-range tanks and the two machines returned to Massingham before the entire squadron moved to Polebrook between 27 and 29 June.
Smaller bombing targets were now in use and at Ashton Range, a pyramid target twenty yards square and thirty feet high was set up for practice with the Sperry sight. Bombing errors were averaging 150 yards from 10,000 feet. But even at this relatively low altitude there were few days when practice could be undertaken due to clouds. Release of bombs had only been made from up to 20,000 feet, considerably lower than the RAF-proposed attack altitude.
The Sperry vs Norden saga
As a result of political in-fighting between the US Navy and the AAC, coupled with a high degree of sales' hype' from Norden's meant that the O-1 Sperry sight and the A-2 Sperry auto-pilot were installed before delivery of the Fortresses to the RAF.
Before the Norden and Sperry bombsights, accurate high altitude bombing was considered impossible. In the 1930s, comparatively simple mechanisms guaranteed fair accuracy in hitting targets from altitudes below 5000 feet. But at heights above the effective range of ant-aircraft guns, aircraft moved too fast for normal calculations of firing data. The problem of calculating in real time the proper point for releasing a bomb was formidable for the equipment then in use. A bomber traveled rapidly in three dimensions and rotated about three axes, and was often buffeted by air turbulence. The path of the dropped bomb was a function of the acceleration of gravity and the speed of the machine, modified by altitude, wind direction, and the ballistics of the specific bomb. The bombardier's problem was not simply an airborne version of the artillery- gunner's challenge of hitting a moving target; it involved aiming a moving gun with the equivalent of a variable powder charge aboard a platform that was evading gunfire from enemy fighters.
The Norden sight was designed for use on US Navy aircraft by Carl L Norden, a Dutch engineer educated in Switzerland who emigrated to the US in 1904. He had been a Navy consultant on different projects since 1915. For the four years before that, he was an engineer working on ship gyrostabilizers with the newly formed Sperry Gyroscope Co., and continued as a consultant to Sperry through World War One. Norden began studying bombing problems in 1921 as a consultant to the U S Navy. In 1923, Norden went into partnership with another Navy consultant, a former Army colonel named Theodore H Barth, who provided valuable know-how in sales. Over the next five years Norden designed bombsights, and Barth built and tested prototypes from Norden's top secret drawings. In 1928, Norden and Barth received their first order from the Navy for forty bombsights. At that point the two incorporated as Carl L. Norden Inc.
The Norden sight was initially built at the Norden plant in New York City and then at several other companies with a wide variety of different versions being built, all with minor differences. The sight was used in conjunction with the aircraft's automatic pilot, the bomb sight being the 'governing agent' of the automatic pilot.
To make the bombsight's telescope independent of the buffeting of the aircraft, it was hung from gimbals. Inside the sight were two DC-powered gyroscopes one for vertical orientation and one for azimuth reference. Both spun at 7800 rpm. Through an electromechanical servo mechanism similar to those that operated ship stabilizers, the azimuth gyro steadied the bombsight optics in the horizontal plane so the crosshairs could be synchronized with the aircraft's approach.
The Norden design had at least four operating problems. Firstly, the carbon DC brushes wore out and had to be replaced frequently; moreover, carbon dust from the wearing brushes would settle into the sensitive gimbal bearings, increasing friction, and necessitating the repeated cleaning and oiling. Secondly, accurate leveling of the vertical gyro was a tricky procedure, especially in rough air, as it required manual setting of two liquid levels like the bubble in a carpenter's level. The process could take several minutes, a significant amount of time during the bombing run. The third problem was that both the azimuth and range operating knobs were on the right hand side of the bombsight, making simultaneous two-hand sighting on a target almost impossible. Fourthly, the angular freedom of the vertical gyro was such that in rough air the gyro would hit the limit stops and tumble off its axis of rotation, losing the bombing run.
Meanwhile, Sperry Gyroscope Co., which had been founded by Elmer Sperry in 1909, had begun designing and building bombsights as a natural outgrowth of its development of gyroscopes for commercial and military aircraft and ships. As early as 1914, when Carl Norden had been on the payroll for three years, Sperry's company had built and granted a patent for a vertically stabilized bombsight that relied on a vertical gyro assembly driven with DC power. The company went on to develop improved models of this first synchronized sight, culminating in a model called the Sperry O-1. But like the Norden sight, the Sperry gyros had problems. Moreover, there was no market for the Sperry bombsights until the Army began having procurement problems with the Norden company in 1936.
In the 1930s, the US Army was building up its own airborne fighting arm, known as the General Headquarters (GHQ) Air Force, which had been established in 1922. The Army was structured so that the GHQ Air Force had to arrange training and procure supplies through another arm, the Army Air Corps.
The GHQ Air Force, as impressed with the Norden bombsight as the Navy, made it standard equipment on its own bombers by 1934. But because the Norden company was a dedicated source to the Navy, the only way the Army Air Corps could get Norden bombsights was by ordering them through the Navy, a pass-along arrangement that complicated design, development and delivery.
Since the Norden bombsight had been developed primarily for the medium altitudes and slow speeds of small Navy flying boats, it had to be modified for the higher speeds and extremely high and low altitudes of the heavy, long-range Army GHQ Air Force bombers. For Air Force purposes, the Norden's optical field of the telescope was too limited, giving insufficient forward and thwartship vision. The Norden bombsight also did not allow bombs to be accurately targeted if the aircraft were descending in a glide, a manoeuvre preferred to level flight during bombing runs because changing altitude made the bomber a more elusive target for anti-aircraft guns and its trail settings were too limited to accommodate the wind resistance encountered by the faster Air Force machines.
In January 1936, the Navy suspended deliveries of the Norden sight to the Army Air Corps until the Navy's own requirements were satisfied. The commander of the GHQ Air Force, Major General Frank M. Andrews, expressed his concern in a memo to the Chief of the Air Corps and to the Navy. He then openly encouraged the Sperry Gyroscope Co. to develop the O-1 bombsight to meet Air Force specifications.
n's President, Theodore H. Barth, 'spoke somewhat disparagingly of the Sperry Company' and 'stated he was very much grieved that the Air Corps was purchasing... ...an inferior sight' from Sperry, and offered to set up a separate factory exclusively for the Air Corp's a suggestion the Navy did not accept.
When the Air Corps asked Norden to cooperate with Sperry to make a Sperry autopilot standard equipment even on Norden equipped machines, the company balked. To get around the stalemate, in January 1942 the Air Corps contracted for autopilots with the Honeywell Regulator Co., Minneapolis, Minn. The Honeywell autopilot, called the C-1, was based on the Norden SBAE gyros, but incorporated the electronic rate circuits and servos from the Sperry A-5. At the request of the Air Corps, Honeywell engineers went to Sperry for information and a demonstration of the Sperry A-5, and the Air Corps acquired a manufacturing license from Sperry so that Honeywell would have a free hand in incorporating certain features.
Meanwhile, Norden's Barth was working hard to ensure Norden's primacy in military procurement. Barth was a personable and flamboyant salesman for the company, with extensive contacts in both the Navy and the Army, all of whom he enthusiastically wined and dined. The Navy in particular had reservations about Sperry's status as a multi-national commercial company, which, before the war, had licensees in London, Germany and Japan. Barth made great efforts to point out that as a dedicated source Norden could not only 'devote its entire attention to the interests of the Government' but also 'maintain a high degree of secrecy' not possible with an 'international organization' engaged in 'world trade'.
Even though by World War Two the Norden bombsight's classification had been reduced from Top Secret to merely Confidential, Barth and others within Norden skillfully cultivated a 'Top Secret' mystique about the Norden bombsight that remains to this day. Literally thousands of photographs taken by AAF photographers of the bombardiers position all have the sight covered up if it is fitted, for usually the top portion of the sight, dubbed the 'football', was removed from the bottom stabilizer when the aircraft was on the ground, and was escorted by armed guards to the Norden Lockup on each base. Bombardiers, it seems, had to swear an oath during their training '...to protect the secrecy of the American bombsight, if need be with my life itself'. Norden bombardiers would often say that they could drop a bomb into a pickle barrel from 20,000 ft, and legend was, they complained that they were not told which pickle to hit.
Despite all the hype regarding the 'Top Secret' nature of the sight, Herman W. Lang, an employee of the Carl L. Norden Company turned out to be a German spy, and during a visit to Germany in 1938, Lang conferred with German military authorities and reconstructed plans of the confidential materials from memory. Remarkably, the German's decided against its use, having already developed dive-bombing techniques. In 1941, Lang, along with the thirty-two other German agents of the Duquesne Spy Ring, was arrested by the FBI and convicted in the largest espionage prosecution in US history up to that time.
Combat Commences
Maintenance problems with the Fortress I brought a high unserviceability rate, often reducing the bombers available to half. On the evening of 3 July, after an engine check on Fortress AN528 WP:B, the groundcrew commenced a test running. After priming No 2 engine, an attempt to start it brought a backfire and flames. Although the fire extinguisher button was pressed, the flames were not doused and in a matter of seconds the fire had spread to the wing. The station fire force was soon on the scene- but efforts to contain the fire were in vain and AN528 burnt out. The Court of Inquiry into this incident found that the groundcrew had used an incorrect starting procedure, partly as a result of ambiguous instructions in the manual - which was later changed.
In better weather more height tests took place. On one bombing practice the error was cut to 95 yards. A height of 34,200 feet had been reached after 1.5 hours and all the guns fired satisfactorily at 32,000 feet, where gunners reported easy handling. On 6 July the bombing error fell to 78 yards in a drop from 20,000 feet. All was set for operations to begin.
Throughout the 7 July three aircraft, AN519, AN526 and AN529 were prepared and crews fitted out with their special clothing. At 13.00 hours on 8 July briefing began. The chosen target was the centre of the submarine building dock at Wilhelmshaven. Three aircraft in open formation were to bomb from 27,000 feet then climb to 31,000 to 32,000 feet for their getaway.
AN529 and AN526 reached Wilhelmshaven shortly before 16.50 hours and in the next ten minutes delivered their attacks from between 28,000 and 30,000 feet altitude. Four bombs fell from AN526 and hits were recorded 250 yards west of the Bauhafen at Wilhelmshaven and others at 150 yard intervals to the south-east.
Aboard AN529 Plt Officer Mathieson had two bombs hang up, a problem later identified as moisture freezing in a solenoid. Sqn Ldr Maclaren aboard AN519 had engine trouble at 20,000 feet whilst climbing and oil leaks from all four engines. At 27,000 feet oil had frozen one inch thick on the tailplane. As oil pressure was falling Maclaren decided to bomb the Frisians instead. He ran up on Norderney at 16.45, but his bombs fell on the sands 500 yards from the town.
The astrodome on board AN529 froze up at 18,000 feet making fire control, impossible, but the guns and mountings were frozen too. Only two Messerschmitt Bf 109s were seen after the attack, some 2,000 feet below the bombers and west of Terschelling. Both fighters closed on the starboard beam of AN529 to about 800 yards when one seemed to enter an involuntary spin. The second followed it down and no fire was opened. Maclaren then encountered lateral flutter due to oil freezing on the tailplane where it had built up to seven inches and caused considerable vibration. All propellers were feathered in turn to no effect. The situation only improved after they descended to 5,000 feet for the journey home.
This initiation of the Fortress into action highlighted many problems. Freezing equipment and oil throwing had not been met in the low humidity environment in which B-I7s generally flew in the USA. Over half the engines of the Squadron's Fortresses were now throwing excessive oil out of the crankcase breather at high altitudes, the phenomenon generally beginning in the neighbourhood of 27,000 feet. Prolonged loss could lower crankcase oil to a dangerous level and oil freezing on the tail surfaces caused control problems. The behaviour of a number of engines made further high altitude sorties a dubious prospect and some aircraft were confined to training duties at lower levels.
Another difficulty was that of contrails which formed behind the aircraft at certain unpredictable levels in the upper atmosphere. As these immediately gave away the position and direction of the bomber to any hostile interceptors, the Fortress had either to climb higher or descend to get out of the contrail level. Frequently neither course was possible and the mission had to be abandoned.
The squadron accepted a service ceiling of 36,000 feet, but physiological problems meant that this height could rarely be contemplated for operations. There were range limitations too, since much fuel was consumed on the battle climb although climb performance which took the bomber to 30,000 feet in about 45 minutes was good. Some attacks on Berlin from 35,000 feet were planned but never executed. To achieve such range additional fuel tanks were to replace two bombs cutting the offensive load to 2,200 lb.
It was soon discovered that physiological problems were not the only ones at such heights. Taking the bombers up two miles beyond the 20,000 feet altitude which the Air Corps considered the optimum for bombing brought mechanical and equipment difficulties. Extreme temperature change caused fractures in the welding of the engine exhaust flanges. The turbo-super-charger controls became unduly sensitive and abrupt control movement could cause the turbos to surge and cause disintegration of a turbine blower. Another problem at 30,000 feet was the build-up of frost on the inner sides of windows, restricting visibility and requiring dispersal by heated panels. Radio signals were weak at this altitude and further changes had to be made to the equipment.
Modification of the American oxygen system to take British regulators and masks had also to be effected. Oxygen masks were found to be uncomfortable and, if allowed to get damp on the ground, tended to freeze up at high altitude. In an effort to eliminate moisture build-up, both oxygen masks and the electrically heated clothing were later kept in a specially heated building.
Of all the technical troubles, however, the most persistent was the tendency of the Wright Cyclones to throw oil from the crankcase breathers. Mysteriously, the problem lay with individual engines for while one performed impeccably to 35,000 feet, another might throw oil at 20,00 feet and the flow would grow alarmingly as altitude increased.
On 26 July AN530 alone bombed Emden from 32,000 feet despite severe icing and storm clouds topping 30,000 feet, but had to land, when his engines gave trouble, at Horsham St Faith in Norfolk.
Two days later the second flying accident occurred shortly after 17.00 hours on 28 July. AN534 departed Polebrook on a high-altitude test and encountered severe turbulence. It crashed at Wilbarston, Northamptonshire, after losing the starboard wing when the pilot attempted recovery.
Two aircraft were to operate against Kiel on 2 August flown by Sqn Ldr Mathieson and Plt Off. Sturmey. Due to a misunderstanding Sturmey formated with a training aircraft and, after finding his error, landed. He was later despatched to bomb Emden. Mathieson successfully bombed his primary without interference but as Sturmey approached Emden cloud was encountered so the target was switched to Borkum, which was bombed through a cloud gap. Twenty miles northwest of Texel two Bf 109Fs attacked the bomber and although it was hit in several places damage was not serious and the crew were uninjured. Defensive fire was seen to enter the engine of a fighter. Four days later Mathieson and Sturmey targetted the German battleships Scharnhorst and Gneisenau in Brest. Mathieson's bombsight froze over and bombs were jettisoned in the dock area. Sturmey attacked from 32,000 feet, placing his load in the Rade Abri.
Sqn Ldr Mathieson took Mr Vose of the Sperry Company as his bomb aimer on 2 September when Bremen was the target. There was heavy flak and bombs fell wide despite the skill of the expert.
On 5 September four Fortresses were bombed up and proceeded to Kinloss. Sqn Ldr Maclaren, flying the reserve aircraft, took along ground personnel. Next day four crews set out to bomb the Admiral Scheer in Oslo harbour. Because of a troublesome supercharger, one aborted, but the rest arrived over Oslo but could not locate the ship. Instead they bombed docks and oil tanks from 30,000 feet. Because they had missed the primary target they were told to stand by for further operations, bombs for which had to be brought from Polebrook.
Sqn Ldr Mathieson with AN533, Fg Off Romans in AN525, Plt Off Sturmey aboard AN532 and Sgt Wood in AN535 set off for Oslo soon after 09.00 hours on 8 September. This time enemy fighters were alerted in time and the most disastrous of the Fortress operations evolved.
In Norway, the German fighter base at Sola airfield near Stavanger had just been equipped with the Messerschmitt 109T, which had a greater wingspan and allowed it to match the B-17's altitude performance. At 11.27 two Bf.109Ts from 13/JG 77 flown by Lt Alfred Jakobi and Lt Wolfgang Steinicke had no difficulties in locating the B-17s in the area of the Setesdal Vally, about 40 miles north of Kristiansand. They were observed by the other two B-17s as the German fighters intercepted and attacked Fg Off Romans' machine at 25,000 feet. His aircraft was soon ablaze and crashed in the Norwegian mountains but not until they had damaged both German fighters. Sgt Wood some four miles behind, seeing interception was inevitable, abandoned his sortie. He jettisoned his load, then climbed to 34,000 feet, and encountered serious oxygen troubles. In addition the intercom system was not working well. In the rare atmosphere voice production was bad, the pilot's vocal chords failing to vibrate effectively. A gunner misunderstood an instruction, disconnected himself from the main oxygen system and began to use his portable bottle. Its supply was soon used and he collapsed. The other gunner, who went to his aid, failed to plug correctly into his portable supply and soon passed out. Sgt Wood was unaware of what had happened and flew on for fifteen minutes before, failing to contact his gunners, he asked the radio operator to see if they were all right. When he found out what had happened Wood dived, but at 29,000 feet a Bf.109 came into the attack, firing a long burst from fifty yards astern. Both gunners were hit, Sgt Wilkins soon died of his wounds, and Flt Sgt Tates had arm injuries. A piece of shrapnel fractured the radio operator's oxygen lead and soon he was unconscious. Damage to the glycol system in the cabin caused it to fill with dense white smoke which blinded the crew. They jettisoned the perspex astrodome and fumes poured out, probably misleading the fighter pilot into thinking the bomber was ablaze.
As it was the first B-17 to be shot down, the Germans expressed much interest in the remains of AN525. From their base at Evjemoen, elements of the Wehrmacht Heer located the crash site near the village of Bygland and removed most of the aircraft, which was freighted to Germany for detailed inspection.
Thereafter came a spell of concentrated fighter affiliation with a Defiant and Spitfire from Castle Kennedy to improve tactics. Bad weather and persistent contrails were now making operations difficult and unlikely, but one special mission was billed. Information was received that a fair of household goods and textiles—particularly from Belgium and Holland—was to be held in Cologne. It was believed the fair would rival that of Leipzig. A large exhibition was placed on the Deutscher Unfer on the Rhine embankment north of the Hohenzollern Bridge. It was further thought that at the opening on 14 September important Nazis would be present. This would be a fine time to attack the exhibition hall making a grand opening—and finale—to the event. All depended on the weather, and because this was a deep penetration the attack must not take place below 32,000 feet. If contrails formed the raid would have to be abandoned and attempted later.
Unfortunately the weather on the 14th was too bad for operations but on 15th AN536 set out. At 32,000 feet trails formed and the sortie had to be halted. Next day Sturmey tried for Cologne, but the operation was ill fated. An engine lost power then a gunner contracted stomach trouble. Operation Ingo was abandoned.
Two more sorties were flown by Sturmey, one on 20th when bombs were accurately placed on the target in Emden. On 25th trails formed at 27,000 feet and this final raid was abandoned fifty miles from target.
Training continued for the next five months, but operations were impossible in winter. Detachments were made to Kinloss for possible operations over Norway, but nothing came of these. On 6 December it was finally agreed that the Fortress 1 was useless for day raids over Europe. It was suggested that flame dampers be fitted for night operations. These never materialized.
Something of what the crews of 90 Sqn had undergone in that summer was related on the BBC on 26 September by Flt Sgt Mulligan. He had flown with Flg Off Sturmey, who had received a well-earned DFC.
'I have flown in the sub-stratosphere in a Fortress over Holland, France, Norway and Germany. If the people on the ground in those countries have seen us at all, we have appeared no more than the tiniest dot in the sky.'
'On your first ascent you are very much aware of flying in unexplored space, relying completely on oxygen. After a few trips you become accustomed to new colours in the sky. I was in the Fortress which was attacked by seven fighters when we were returning from Brest. Three minutes after our bombs had gone the fire controller called out that there were enemy fighters coming up to us from the starboard quarter, 1,000 feet below. They closed in and there was almost no part of the Fortress which was not hit. Some of my friends in the crew were killed, others wounded. A petrol tank was punctured, bomb doors were thrown open, flaps were put out of action, tail tab shot away, tail wheel stuck half down, brakes not working, only one aileron any good and the rudder almost out of control. The centre of the fuselage had become a tangle of wires and broken cables, square feet of the wings had been shot away, and still the pilot managed to land the Fortress on a strange aerodrome. This is a testimony to the makers in America.
'Fortunately these thrills are rare. Our attack on Emden last week was almost without incident, except, of course, for the dropping of the bombs by the Sperry sight with beautiful accuracy on the target. We lost sight of our aerodrome at 2,000 feet and never saw the ground again until we were off the Dutch islands. Foamy white cloud, like the froth on a huge tankard of beer, stretched all over England and for about thirty miles out to sea. The horizon turned—quite suddenly—from purple to green and from green to yellow. It was hazy but I could see Emden fifty miles away.
'I called out to the pilot "Stand by for bombing, bombsight in detent, George in. O.K. I've got her." The drill is that I push a lever on my left for the bomb doors to open, and on a dial in my cabin two arms move out like the hands of a clock to show me the position of the bomb doors. I keep my eye down the sighting tube which, incidentally, contains 26 prisms, and with my wrist I work the release. As the cross hairs centred over a shining pinpoint in Emden on which the sun was glinting, the bombs went down. We were still two miles away from Emden when we turned away. Almost a minute later one of the gunners told us through the intercom 'There you are, bursts in the centre of the something target," and back we came through those extraordinary tints of the sky.
'During the whole sortie I only had one thrilling moment. I saw a Messerschmitt coming towards us. He seemed an improved type, and I looked again. It was a mosquito which had got stuck on the perspex in the take-off and had frozen stiff. Otherwise it proved an uneventful typical trip in a Fortress, with the temperature at minus 30 degrees below zero Centigrade.'
In November 1941 the use of the Fortress Mk.I was reviewed. Clearly it was not a stratosphere bomber, and its fighter defences were poor. Before it could fight any engagement waist guns needed to be fitted into position after huge side panels had been removed. This was an almost impossible task for gunners at high altitudes, and the gaping holes made the aircraft bitterly cold inside. Contrailing made lone operations hazardous, and for half the year possible on about only one day per month. With the Wellington VI pressure cabin bomber soon becoming available consideration was given to switching Fortress Is to night operations. The squadron was therefore passed to 8 Group on 2 January 1942 which was forming to use Liberators for night raids. Night bombing of Germany was not the task of 2 Group, and it was now left to others to make the Fortress work in daylight.
Gradually the survivors were split between the RAF's Coastal Command for Search and Rescue duties and a small bombing contingent in the Middle East. The latter group of aircraft moved to India and eventually one survivor was returned to the USAAF. It is clear from surviving records that the Royal Air Force was not impressed with 'Boeing's Best'.
Lessons learned from the Royal Air Force
Over the years, it has been suggested that there were 'differences of opinion' as to how the Fortress Is should have been operated and that the Royal Air Force was negligent in using them as they did. Many historians - mainly American - have claimed that despite warnings from the US Army Air Corps that the B-17 should only be operated at 20,000 feet and as part of a large formation so that each aircraft could provide overlapping zones of defensive fire to the other machines, the RAF still used their B-17Cs in small formations of two or three aircraft flying at up to 32,000 feet as defence against fighters and flak. Whilst that statement is correct, it is also a far too simplistic viewpoint, for it does not take into account that they really had no choice in using them as they did. What is not generally realized is that as early as August 1941 the RAF's Bomber Command were calling for more B-17s to be made available than just the ten aircraft that were on squadron strength so as to be able to operate in just such a manner as theorized by the Americans. A strong strategic and operational case was made for the USAAF - the Air Corps became the Air Force on 20 June 1941 - to divert more B-17s to the RAF, but it was turned down flat.
The RAF also discovered that the guns froze up, and the aircraft desperately needed some power-operated turrets. The Royal Air Force certainly did not like the lack of defensive guns at the extreme rear. There was a huge 'blind spot' where enemy fighters could approach from and attack with impunity. The RAF realized rearward facing guns - preferably in the form of a power-operated turret behind the tail - were essential for bomber defence. Whether the RAF operation of the Fortress I was a foretaste or a fiasco is debatable, but clearly lessons were learned that were soon put into effect.
1. The Fortress cannot rely on the evasion of enemy aircraft below 32,000 feet.
2. Once intercepted, the Fortress has little chance against modern fighters.
3. Despite the reputed accuracy of the Sperry sight, precise bombing with the Fortress above 32,000 feet is extremely difficult due to the physical and mental strain imposed in using equipment at that great height.
4. Dependence on suitable weather to carry out attack limits the frequency of operations.
5. Condensation trails can cause a raid to be abandoned. Gives away the position of the B-17.
6. The average bomb load is uneconomical in relation to the manning and maintenance effort.
The RAF made every possible effort to help the Eighth Air Force to get in to the war as soon as was humanly possible but many remained very sceptical about its ability to bomb accurately in daylight. They based their views on their own bitter experiences of 1939-40 and their own disastrous daylight bombing campaign and also on the successful survival of the 'Blitzkrieg' attacks by the German Luftwaffe on the airfields of southern England and later London. Nevertheless, American faith in daylight bombing was unshakable - American proponents of air power claimed that it could destroy German industry, for, in theory, 90% of all bombs dropped on a clear day would explode within 1,250 feet of the Mean Point of Impact. This belief originated in the 1930s, when it was thought that unescorted long-range bombers would be able to successfully reach a target if they were heavily armed and organized into massed formations.
