Website title

Home About Articles Bulletin Contact Gifts History Links Updates


NOTE... WE NEED YOUR HELP! If you have an article that you can share be it serious, amusing or otherwise we would love to hear from you. Any memories of events, places or things be it short or long, please send it along. (If you still remember it after 60 years then it is important and should be here.)

The following articles have been kindly submitted by the individual authors as noted. They are in no particular order and represent a mixture of the serious, tragic, and humerous side of Squadron life. Keep in mind that these events took place over half a century ago, they are not intended to be detailed historical documents and there may be some errors. Remember that two different people can see the same event from an entirely different perspective. All these articles give life to the Squadron's history, without which this history becomes nothing more than a long dry list of names and pictures.
Articles :
The Valiant years - RAF Marham
By WO Shaun P Broaders MBE RAF Ret

'You are posted to RAF Marham'! The year is 1959 and fresh out of armament mechanic training the clerk advises me of my posting; 'Report to the general office on arrival'. I am posted to a Bomber unit with the first of the V Bombers, the Valiant. Leaving Melksham for Marham all kinds of visions pass across the mind, but you have to feel that a posting to a Squadron would be much better, as it would be more exiting and a chance to see the world. On arrival at Marham it took a few months to settle in and attend the Valiant Ground Servicing course at RAF Gaydon and the Martin Baker Ejection Seat course Type 3A at Higher Denham and then it happened, I was posted to No 214 Squadron, what joy.

Just to put you in the picture, concerning 214 Squadron and Marham, I think a little history would not go amiss.
On the 21st January 1956, 214 Squadron re-formed at Marham with Valiant B1's, first of the V Bombers. Powered by four Avon 204 axial-flow turbojets rated at 10,050 lb.s.t. Estimated performance: speed 610 mph at 40,000 ft cruising speed 530 mph service ceiling 50.000 ft initial rate of climb 4,500 ft/min and range was 3,000 plus miles.
The same year 207 Squadron disbanded on 27th March and re-formed at Marham on 1st April with Valiants. (During that month a party of VIP's arrived on the station and included; Mr N Kruschev, Marshall Bulganin, Mr I V Kurbachov (a nuclear scientist) and Mr A N Tupolev, the Russian aircraft designer). 148 Squadron also re-formed at Marham on the 1st July with Valiant aircraft.
During September/October 1956 all three squadrons were detached to Luqa airfield in Malta to take part in the Suez operations. In October, 214 Squadron carried out raids with 1000lb bombs on El Adem airfield, Almaza and Abu Sueir, and in November the targets were Kasfrit, Huckstep Barracks and El Agami. All of these squadrons returned to Marham in November/December 1956. Also, with the arrival of the nuclear capability in 1957, RAF Marham was awarded its own Station Crest, a Blue bull with the motto "DETER".

In-flight refuelling was not a new concept in 1956, history records that as early as 1919 some people were beginning to think that by transferring fuel from one aircraft to another in flight, thereby extending their flight times, that this idea could be used for trans-continental flights and many combinations of aircraft have been used through the years trying to prove the benefits of in-flight refuelling.
Flight Refuelling Ltd (FRL) has had a long association with the RAF but the company was about to end its long association with propeller-driven aircraft. In May 1953, following a decision to equip the RAF's 'V' bomber force for aerial refuelling, Canberra B.2, WH 734 arrived at FLR and after installation of the prototype Mk. 16 Hose Drum Unit (HDU), then being developed for the Valiant tankers, it became Britain's first jet tanker.
Further studies in flight refuelling led to trials of a modified Valiant bomber, which started flying from Boscombe Down in 1956 and after clearance for Service was obtained from the Aircraft and Armament Experimental Establishment (A & AEE) 214 Squadron began Valiant-to-Valiant flight refuelling trials in 1957, which were in the main dry contacts. Then in March 1958, whilst retaining its bombing role, 214 Squadron became the trials and development unit for flight refuelling trials number 306 and 306A, Trial No 306 was to test the capability of aircraft tanker and receiver equipment, and Trial No 306A was for developing modern rendezvous (RV) procedures and techniques. (These Trials were to last until May 1960). The initial training of both air and ground crews was carried out at Flight Refuelling Ltd. at Tarrant Rushton, while a Flight Refuelling School was developed at Marham - and all subsequent training for flight refuelling in the Royal Air Force was to be carried out at this School.
During this evaluation period Wg. Cdr. Michael Beetham commanded the Squadron; who as we all know became Chief of the Air Staff in 1977 and eventually retired from the service as Marshall of the Royal Air Force in 1982.

Conversion of the Valiant Bomber entailed the fitting of a probe to the front of the NBS scanner bay and connecting it internally to the aircraft fuel system and installing a Mark 16 HDU internally in the rear of the bomb bay and a 4500 lb fuel tank in the front of the bomb bay. The HDU control panel was positioned beside the Navigator Radar who became in addition the fuel panel operator. External floodlights were fitted to give the aircraft the capability of flight refuelling at night.
The modification to the Valiant bomber gave the aircraft in the tanker role the facility of transferring 45,000 pounds of fuel at a maximum rate of 4000 pounds per minute with a maximum drogue fuel pressure of 50 pounds per square inch. Although there were early teething problems, these were overcome, and in January 1959 two fully modified tankers began wet transfers.

The Mark 16 Hose Drum Unit and Drogue

Hose Drum Unit (HDU) installed in the bomb bay; winding in the hose.

The 214 Squadron Valiants actually gave their first public demonstration of air-to-air refuelling at the 1958 Society of British Aircraft Constructors (SBAC) Flying Display, after which, during the Trial No 306 they were involved in many long distance proving flights and various interesting unofficial records were made:

a. Marham to Aden in 7 hrs 10 mins.
b. Marham to Nairobi 4,350 miles in 7 hr 40 mins.
c. Marham to Salisbury 5,320 miles in 9 hr 42 mins.
d. UK to Johannesburg 5,845 miles in 11 hr 3 mins.
e. UK to Capetown 6,060 in 11 hr 28 mins.
f. Capetown to UK 6,060 miles in 12 hr 20 mins.
g. The longest jet flight (at that time) by an RAF aircraft, (214 Sqn Valiant) of 18hr 5 mins, which covered 7400 nautical miles, around UK.
Sqn Ldr J H Garstin flew this aircraft.

As recalled, 'IN COBHAMS' COMPANY' by Colin Cruddas ISBN 0952449907;
Sir Michael was particularly inspired to make a two-way, record-breaking flight from England to South Africa in 1959, by Sir Alan Cobham's journey to the Cape thirty -four years earlier. His Valiant was the first aircraft to fly to Capetown and back, non-stop in each direction, being refuelled by two Valiant tankers over Kano, Nigeria on both flights.

Wg. Cdr. Michael Beetham (left) and crew of the Valiant, which undertook the first non-stop flight to Cape Town, 1959.

Pat Hornbridge, Dickie Dickenson and Sir Alan Cobham, visit 214 Sqn, Marham, 10 June 1959. With Sqn Ldr Garstin, Wg Cdr Michael Beetham and Gp Capt Wilf Burnett

By now both aircrew and ground crew were enjoying flight refuelling. The main challenge to the aircrew was the exacting task of learning to fly as receivers whereas the ground crew had many opportunities for visiting exotic places such as Malta, Cyprus, Bahrain, Karachi, Mauripur, Nairobi, RAF Gan, RAAF Butterworth, Singapore and Darwin. Of course they also enjoyed flying in the Bristol Britannia, which usually stayed with the Squadron throughout the detachment. You can imagine the enthusiasm when the next detachment was mentioned; almost all the ground crew personnel were quick to volunteer for the trip, knowing all too well that some were going to be disappointed.

WZ 390 and WZ 376 During Trials

May 1960 saw the completion of the Trials No 306 and 306A, it also heralded the arrival of Wg Cdr P.G. Hill the new Squadron C O. The Squadron were kept busy with training and deployments overseas as well as taking part in trials with the Royal Navy Scimitar and Sea Vixen fighters and a compatibility exercise with a USAF Destroyer, Super Sabre and Voodoo aircraft.

214 Valiant on servicing; Inset photograph Wg Cdr P G Hill

Other long distance flights during this period include:
a. Marham to Offutt 4,336 miles in 9hr 30 mins 19 Jan 1960
b. Offutt to St Mawgan 4,400 miles in 9 hr 3 mins 25 Jan 1960
c. Marham to Changi, 8110 miles in 15 hr 35 mins 25 May 1960
d. Butterworth to Marham, 7,700 miles in 16 hr 16 mins 1 June 1960
e. Marham to Vancouver, 5007 miles in 10 hr 28 mins 5 July 1960
f. Vancouver to Marham, 5007 miles in 9 hr 35 mins 8 July 1960
Another interesting exercise in which the Squadron took part was the refuelling of a Scimitar aircraft from the Royal Naval Air Station (RNAS) at Lossiemouth to the Aircraft Carrier Ark Royal 'somewhere in the Mediterranean'. We did find it!

The Javelin crews of No 23 Squadron from RAF Coltishall started receiver training in 1960 with practice flights to Akrotiri and back in August, then in October four Javelins were flight-refuelled from the UK to RAAF Butterworth in Malaysia via Akrotiri, Bahrain, Mauripur, and Gan.
It was on this trip that a number of the squadron ground crew suffered the usual tummy upset at Mauripur, fortunately we had the Javelin Squadron medical officer with us. To ensure we did not drink any more water, I recall him giving authorisation for the squadron to obtain soft drinks from a local soft drinks distribution firm for the duration of the detachment. This ensured that we all made it to Butterworth without further mishap. Ernie Hill, the squadron Sgt Electrician, was very ill at this time and it was doubtful whether or not he would make it to Butterworth. However, with good nursing and plenty of fizzy drinks, he recovered enough for the onward trip to Gan and after medical treatment carried on to RAAF Butterworth with the Squadron.

Valiant Tanker of 214 Sqn refuelling a Javelin of 23 Sqn

It was about December of 1960 that Vulcan crews started receiver training for the proposed non-stop flight to Sydney in Australia. Training progressed with non-stop flights from Scampton to Nairobi, and to Karachi and back then finally in June 1961 with the first non-stop flight from Scampton to Sydney, a distance of 11,500 miles in 20 hours 3 minutes at an average speed of 573 mph, the aircraft was refuelled overhead Cyprus, Karachi and Singapore. If memory serves me right we had nine Valiants on this trip and all remained serviceable for the duration of the exercise.
After this significant achievement Sir Alan Cobham presented trophies to the officer commanding of No's 214 and 617 Squadrons. It was also memorable in that the double 'Speed Bird' logo used by FRL was now incorporated in the 214 Squadrons emblem emblazoned on the tail fins of the tankers. (See picture page 9)

Other memories are of the boss, during one of the many training trips to Nairobi, coming home with a bomb bay pannier full of fruit for the squadron personnel. My wife and I recently visited Kenya and arrived at Nairobi airport only to find the aircraft parked opposite the RAF Detachment Office that we used, still proudly displaying the RAF sign. The bomb bay pannier was also handy for the Christmas run to Malta, when orders were taken for the Christmas bottle.
Valiant tanker of 214 Sqn refuelling a Vulcan

Happy Days - 214 ground crew with a Valiant tanker - staging through Malta on the way to the Far East -1961. Author; is front row, third from left.

It should be noted for the squadron enthusiast that on the 1st April 1962 (44 years after the formation of the RAF) No 214 (FMS) Squadron and No 90 Squadron officially became tanker squadrons, losing their bomber commitment and so becoming the first tanker squadrons in the RAF. HQ 3 Group retained operational control of these squadrons.

Tank Change using Simons Bomb Hoist mounted on a 4 Ton Bedford Chassis

It was during the loading of a HDU to one of our aircraft that we nearly had a serious accident, although at the time we all fell about laughing. I used to drive and operate the Simons Bomb Hoist, the corporal, (Cpl Blower) would get in the cage and I would manoeuvre him onto the spine of the aircraft and with the chief in the bomb bay we would carryout the loading procedure, during this time we would be in contact through a throat mike head set. After the load was completed it was standard operating procedure for the corporal to stow the equipment, remove his headset and throat mike and before placing them in the cage, hold the mike to his throat so as give me the instruction to withdraw the boom and remove the vehicle, (you must bear in mind that you could not see the top of the aircraft from the vehicle), he would then come down via the aircraft rear hatch. The instruction given I started to swing the boom, only to hear a strange noise in my headset; that's right, he was still attached and I was pulling him along the spine of the aircraft, I hate to think what would have happened if I had decided to raise the boom first.
It was about this time that the RAF Mk VI probe and drogue equipment was replaced with the Standard NATO Mk VIII probe and drogue. There was further training with the Vulcan's and this was to lead to yet another trip to Australia. This time with three Vulcan's of 101 Squadron who flew non-stop from Waddington to Perth in 18 hr 7 mins. Operational training was in full flow and the Sqn was to carry on training with the Vulcan's, Sea Vixen's, Victor's and the Lightning aircraft, which would be the next aircraft to be tanked to the Far East. There was never a dull moment on the Squadron. One day we had a Valiant return with the hose and drogue still extended, some electrical fault had prevented it from being wound in, but the crew being professional landed safely. There was also an incident when a Lightning lost its probe, you've guessed it, it was still in the drogue when the Valiant landed.

XD 816 Note the Squadron Crest within the Speed Logo

In December 1964 the Valiant aircraft were grounded because of severe metal fatigue in the main wing spars and were officially withdrawn from service at the end of January 1965, many of the squadron personnel were posted during December 64 and the rest during the early part of the new year. I consider myself to be one of the lucky ones as I was posted, in March 65, to RAAF Butterworth, a unit I had visited regularly during our tanking exercises. Now my wife could see for herself the terrific hardship we suffered in the Far East!

214 Squadron disbanded on 28 February 1965, a sad day for all on the squadron and those associated with it. During the tanker years the squadron had a record that anyone would be proud of and history will recall that the first tanker squadron; 214 (Federated Malay States) Squadron played a significant role for the future of air-to-air refuelling.

214 Squadron was to reform at Marham on the 21 July 1966 with Victor tankers. However, that is another story.

WO Shaun P Broaders MBE RAF Ret

I would like to offer my grateful thanks to the following, for their help and contribution.
Photographs from Cobhams Archives and quotations from In Cobhams' Company by Colin Cruddas ISBN 0 9524488 0 7
Photographs from Aerofax Vickers Valiant by Eric Morgan available from Midland Counties Publications 0145 254450

First Contacts!
By Flt/Lt Gary Weightman

Part 1:
The Big Stick and the Looped Hose.

As the air war over the Balkans demonstrates once again, the effective application of global airpower is highly dependant on air to air refuelling support provided by the fleets of NATO tanker aircraft that crowded airfields from Brize Norton to Budapest. AAR has been a vital factor in air warfare since Korea and Vietnam and the RAF tanker force has played a major role in campaigns from the Falklands to Iraq and Yugoslavia. However, despite the invention of practical AAR in Britain it was the United States and the USAF that first took up the technology and applied it to military use. It was fifty years ago that AAR evolved from a flying circus stunt into a military force multiplier. This major development in aviation took place in the skies over Southern England in 1949 and it is important to mark the 50th anniversary of this historic event. It is a story full of political intrigue, inter-service rivalry, bureaucratic obstruction and heroic British innovation.
The birth of the United States Air Force on 18th September 1947 had not been an easy one. The US Navy, who feared that they would lose control of their own aircraft, had opposed its formation at every stage. The advent of atomic weapons added a new twist to this bitter inter-service rivalry. The USAF argued that it had the sole means of delivery of the atomic bomb in its long range strategic bomber force and should receive the lion's share of defence funding to pay for its new B-36 bombers. The Navy were appalled at the possibility of losing out and argued that the USAF had misled congress on the B-36's performance and that billions of dollars would be wasted. These dollars would be better spent on new 70,000-ton supercarrier, the USS United States, that could carry a 45-ton atomic bomber and be capable of striking from the sea at an enemy anywhere in the world. When the Russians blockaded Berlin on 31st March 1948 it became clear that this enemy would likely be the Soviet Union.
Footnote to Garys comment regarding the Soviet Union. Some readers may not be aware but Churchill, in an absurd lack of foresight or blatent arrogance, put all of the Western worlds Atomic science on the table during negotiations with Stalin. A stroke of bureaucratic lunacy indefensible and beyond all comprehension. "Nothing like starting a nuclear arms race on an even playing field".What helped Churchill win some token concessions from the Russians would nearly cost us the planet in years to come.
The USAF was rattled by the Navy's case. Their existing B-29 bomber didn't have the range to reach targets in central Russia and the giant B-36 was proving to be a sluggard in early trials. The Boeing B-47 and B-52 jet bombers were years from entering service and the Navy were clearly ahead in the game. The USAF could only win the nuclear funding battle if they could prove that their Strategic Air Command bombers were capable of reaching their targets from bases in the USA. The solution to the problem appeared to lie in the hands of a small company based in Dorset, England.
Flight Refuelling Ltd had been formed by British aviation pioneer Sir Alan Cobham in 1934 to develop air to air refuelling as a commercial proposition. By 1939 the first regular flight refuelled transatlantic airmail flights, using the looped hose method, were inaugurated but the outbreak of war terminated the service. The surplus equipment was used in a series of successful wartime trials in the USA and in 1944 Sir Alan was awarded an air ministry contract for 600 sets of looped hose flight-refuelling equipment for the RAF's "Tiger Force" Lancasters. A considerable amount of work had been completed before the contract was suddenly cancelled in 1945. After the war Flight Refuelling continued the transatlantic trials leading to a successful flight refuelled passenger service to Bermuda in 1947
When the success of these trials came to the attention of the USAF they decided to adopt the procedure themselves. In April 1948 a group of senior USAF officers arrived at Flight Refuelling Ltd's Tarrant Rushton base with an immediate order for 40 sets of their "looped hose" equipment (later extended to 100). Fortunately, Sir Alan Cobham had just purchased at scrap value from the Air Ministry all the parts and components produced for the "Tiger Force" tankers and receivers and could easily meet the tight delivery schedule set by the Americans. Boeing began a rapid conversion program that would eventually produce 92 KB29M "looped hose" tankers and 131 B29 and B50 receivers. The first B-36A entered service with SAC in June 1948 and work began to convert the giant bomber into a looped hose receiver. By December 1948 the USAF was ready to try out the system.
The political battle with the US Navy was approaching a climax with a congressional inquiry into the conflicting merits of the B-36 mega-bomber and the USS United States super-carrier. The USAF needed some positive publicity and the anniversary of the Japanese attack on Pearl Harbour gave them a perfect opportunity. On the 7th December 1948 a B-50 bomber flew from Carswell AFB, Fort Worth, Texas, carried out an undetected mock attack on the US Navy's Pacific fleet in Pearl Harbour, Hawaii, dropped a dummy 10,000lb atomic bomb and returned, completing a round trip of 9,400 miles using looped hose flight-refuelling. Much to the chagrin of the US Navy, this demonstrated the vulnerability of their fleet bases to atomic attack by long-range bombers and greatly assisted the case for SAC and the B-36.
To further demonstrate its new global strike capability the USAF planned the ultimate in long-range record attempts, to fly non-stop around the world. On Saturday the 26th February 1949 a Boeing B-50 named "Lucky Lady II", of the 43rd Bomb Group, departed from Carswell AFB and flew 3,800 miles east towards the Azores. At Lages 4 KB-29M tankers waited and 2 were used for the first refuelling. Lucky Lady II then flew 5,200 miles to Dhahran for a second bracket, 4,900 miles to the Philippines for the third and 5,300 to Hawaii for a final refuelling before landing at Fort Worth on Wednesday 2nd March. The flight covered 23,452 miles and took 94 hours and 1 minute. In all, 16 KB-29M tankers of the 43rd Air Refuelling Squadron were deployed along the route, an indication of the rapid build-up of the USAF tanker force just 10 months after the contract was signed with Flight Refuelling Ltd. The around-the-world flight was final proof that the US Air Force bombers were capable of striking anywhere in the world. The USAF won its case for the B-36 in congress. Although construction had already begun on the USS United States it was cancelled in April 1949. Efforts by the Navy to blacken the name of the B-36 and even hint at corruption in high places failed to reverse the decision and the bomber went on to be the mainstay of SAC's nuclear "big stick" in the early 1950s.

Part 2:
Dustbin lids, Roller Blinds and Relief Tubes.

Despite Cobham's continuing belief that his flight-refuelling system had a future in commercial aviation it was the military that now drove developments and the USAF, rather than the RAF, its main proponent. Although Flight Refuelling's looped hose method was a proven and reliable system its complexity and the low speed required for the operation limited its use to large multi-crew piston-engine aircraft. In 1947 the Air Ministry had concluded that the system was impractical and of little value in future military operations and that any further development of flight-refuelling equipment would be discontinued. Had it not been for Cobham's persistence and the lucrative USAF contract development in Britain might have ceased altogether. In the USA Boeing began investigating the "boom" method for refuelling high speed jet bombers but the USAF were also interested in any new AAR technique that could be used for single seat jet fighters. Cobham had been told of this interest late in 1948 during a lunch with senior USAF officers at Wright Patterson AFB, in Dayton, Ohio. He rashly mentioned that his company was already working on such a system. The USAF generals were most impressed and arranged to visit Flight Refuelling Ltd's Tarrant Rushton base in the spring of 1949 for a demonstration. Unfortunately, Sir Alan had been bluffing and no such project yet existed. The "looped hose" method had taken many years to develop and now he had just over 4 months to come up with a completely new system.
Returning to England Cobham gave the new project top priority with his design team. Several options were investigated before attention began to centre on a receiver aircraft equipped with a probe mounted nozzle which would make contact with a receptor coupling in a tapered funnel or drogue on the end of a tanker's hose. Various designs of drogue were tested by towing them at high speed behind a car down the runway at Tarrant Rushton. In principle the technique looked promising but a problem remained on how to prevent the tanker hose from whipping and looping when contact was made. It was Flight Refuelling engineer Peter Macgregor who came up with the solution. He considered the method of retraction used by measuring tapes and roller blinds while lying in bed one Sunday morning. If a spring could take up the slack then the problem of hose whipping would be eliminated. Macgregor decided that a mechanical spring system would be too heavy but design work showed that an electro-hydraulic fluid drive motor attached to a hose drum unit would do the job and could be fitted in one of Flight Refuelling's Avro Lancaster tankers. All that had to be found was a suitable receiver aircraft.
Cobham was a master at getting results from an unenthusiastic Air Ministry and eventually Meteor Mk 3 EE397 was loaned by the RAF to the company for the trials. Time was running out and Flight Refuelling had just over a month to refine the coupling system and modify the Meteor with a nose-mounted probe and internal fuel lines. The system was eventually ready for the first flight trials just two days before USAF General Carroll and his team were due to see a demonstration of the new probe and drogue technique.
On Saturday the 4th April 1949 FRL's test pilot Pat Hornidge took off from Tarrant Rushton in Meteor EE397 and carried out the first series of dry contacts with the Lancaster tanker flown by Tom Marks. Throughout the weekend flight trials sufficiently refined the technique of the probe and drogue tanking to allow the planned demonstration to take place promptly at 10.00 am on Monday when the USAF officers arrived. Not only had a completely new method of AAR been developed in a matter of months, but also for the first time a jet aircraft had refuelled in flight, which heralded a new era of possibilities. The USAF delegation was most impressed with the demonstration but technical problems delayed a repeat performance for over a month. Eventually a series of seven demonstration flights were planned for VIPs from the RAF and aviation industry. Sir Alan claimed that his probe and drogue system could transfer fuel at 4,000lbs a minute while flying at 400kts at 35,000ft. However, one officer noted that the demonstrations were restricted to contacts at 130kts and 2,000ft and little fuel appeared to be transferred to the Meteor. Cobham managed to convince the sceptic that a temporary snag had curtailed that particular demonstration but in truth unless his company acquired a more capable tanker than the Lancaster from a reluctant Air Ministry it would be difficult to carry out any high altitude trials.
To dramatically demonstrate the operational advantages offered by the probe and drogue method, test pilots Hornidge and Marks organised an attempt on the world jet endurance record. On the 7th August 1949 Hornidge took off from Tarrant Rushton in the Meteor and flew a continuous circuit down the Bristol Channel and along the south coast to the Needles to rendezvous with Mark's Lancaster orbiting around the Isle of Wight. Hornidge remained airborne for 12 hours and 3 minutes flying about 3,600 miles. He made 10 refuelling contacts and received a total of 2,352 imperial gallons of kerosene. During his record flight he suffered the indignity of a "pee tube" failure, which flooded his cockpit floor, but the attempt was a great success and gained considerable international publicity for FRL's revolutional new probe and drogue AAR technique.
1949 also saw the first deployment of USAF tankers to the UK when the KB-29Ms of the 43rd Air Refuelling Sqn deployed to RAF Marham to carry out looped hose AAR exercises with B-50 bombers based in East Anglia. The old method was having mixed results in service with the USAF and by the end of the year the arrival at Tarrant Rushton of four B-29 bombers for conversion to HDU equipped tankers showed the way ahead. One of the aircraft would be converted into the world's first 3-point tanker, designated YKB-29T with Mk 11 HDUs mounted in the bomb bay and on each wingtip.
It is 50 years since the momentous events of 1949 but the year truly marked the birth of military AAR with the British invention (backed by American money) of the probe and drogue technique. The USAF used probe and drogue in TAC until the Vietnam War before standardising on the SAC fleet of boom equipped KC-135 tankers. The reintroduction of hoses on USAF tankers in recent years may indicate the error of the earlier decision. The RAF was slow to adopt AAR and it was not until November 1959 that No 214 Sqn became the RAF's first fully operational tanker unit equipped with single-hose Vickers Valiants. It was 1966 before the first 3-point Victor tanker entered service with No 57 Sqn. However, probe and drogue has proved to be the most versatile method of AAR, capable of use with aircraft as diverse as helicopters and supersonic fighters and it enables even the diminutive A-4 Skyhawk to be used as a tanker. Its invention is most worthy of celebration on its 50th anniversary year, so let's raise our glasses to Cobham, Macgregor and roller blinds everywhere! Cheers!

Part 3:
Bust and Boom.

The triumphant British invention of the world's first practical in-flight refuelling system in 1949 can, with hindsight, be seen as a major milestone in the development of airpower. Indeed it can be argued that the development of probe and drogue rivals that of the jet engine in the UK's contribution to modern aviation. Yet far from bringing glory and profit for Flight Refuelling Ltd the technology continued to be regarded as an unnecessary circus stunt by the mandarins of the Air Ministry and by the end of the year the firm faced bankruptcy. Of even more concern was that the small Dorset based company now found itself in direct competition with the most powerful aviation industrial complex on the planet, Boeing!
The unenthusiastic attitude of the Air Ministry was very much a symptom of the time. Money for research and development in Post war Britain was almost non-existent. Even Whittle's company, Power Jets, was wound up, as the government appeared more willing to give away jet technology to the Americans and Russians rather than develop it at home. What funds existed were poured by beauracrats into dead-end prestige projects such as the Bristol Brabazon long-range airliner. The successful trans-Atlantic in-flight refuelling trials of 1947 appeared irrelevant if an aircraft could be built to cross the ocean on internal fuel alone. The Air Ministry had already concluded that "flight refuelling on future types of aircraft is not a paying is not proposed to continue any further development.but to rely on the aircraft carrying internal fuel for the ranges required." The difference in opinion between the British and Americans over the advantages of AAR could be explained by the very different wartime experiences of their air forces. The RAF had fought mainly in the European theatre over what would be considered today as tactical ranges. Lancasters, Lincolns and even Mosquitoes could easily reach their targets in Eastern Europe from bases in England and aircraft under development promised even greater ranges. However, the USAAF 20th Air Force had fought their campaign in the Pacific over ranges unimagined a few years earlier. It is 635 miles from Brize Norton to Berlin but the USAAF B-29 Superforts had been flying 3128-mile round trips from Guam to Tokyo. This vast difference in experience may partly explain why the British were reluctant and the Americans ready to embrace AAR.
However, profits from the successful looped-hose deal with the USAF barely covered losses incurred by Flight Refuelling when its Berlin airlift contracts were suddenly cancelled earlier than expected in August 1949. To save the company Cobham was forced to sell a majority shareholding to a merchant bank who promptly began an asset stripping operation. It is in these harsh financial conditions that we reach the next stage of our story. The arrival of 4 Boeing B-29 bombers and 2 Republic F-84 Thunderjet fighters at Tarrent Rushton for probe and drogue conversion.
Codenamed "Operation Layette" (later changed to "Operation Outing") this project represented a huge leap forward from the Lancaster/Meteor trials of the spring and summer of 1949. The first B-29 was fitted with a single Mk 8 HDU in the rear fuselage and two others were fitted with a probe extending forward above the cockpit for the first bomber trials with the new system. The single point tanker was redesignated the KB-29T and the receivers became B-29MRs. It had been intended that the B-29MRs would have a pneumatically powered "ejector probe" which, when selected by the pilot, would shoot out to make contact with the drogue. This exciting development was abandoned at the time as overcomplicated but has been born again in recent years for helicopter AAR.
The most significant conversion was made to the second B-29. This was fitted with 3 electrically driven Mk 11 HDUs, two of which were fitted in wing-tip pods, becoming the world's first three-point tanker, designated YKB-29T. Both F-84s were fitted with wing mounted probes, becoming EF-84Es in the USAF inventory. Unfortunately the Superfortress and Thunderjet were considerably more advanced and complex than the Lancaster and Meteor and the conversions proved to be far more difficult than expected. A fixed price contract had been agreed before Cobham's engineers had been able to see what work would be required and although the company was able to meet its commitment the costs overran significantly. With no sign of British government support and in order to cover his losses, Cobham was forced to sell the manufacturing rights for the probe and drogue system to the Americans, less than a year after its invention in the UK.
To add insult to injury Flight Refuelling Ltd also lost its effective monopoly in practical refuelling techniques. Boeing, who had benefited from FRL's work with huge USAF contracts to convert B-29s into KB-29M tankers, had been working in secret since November 1947 on a new in-flight refuelling method for its future swept-wing B-47 jet bomber. On the 19th October 1949 Boeing publicly revealed its "hitherto classified device", the flying boom refuelling system.
Boeing claimed several advantages of their new boom over probe and drogue. The rigid boom could allow a higher fuel flow than a hose. The boom could operate at greater altitudes and speeds than had been demonstrated with probe and drogue. Also, as a boom operator in the tanker aircraft made the coupling with the receiver, receiver pilots would not need to be trained in the formation skills required for probe and drogue contacts. The final advantage of the boom over probe and drogue in the eyes of America was that it was an invention of good old Yankee know-how and not some Limey lash-up.
Although many of the so-called advantages of the boom system were illusory the Boeing challenge threatened FRL's AAR lead with probe and drogue. As 1949 drew to a close the future adoption of the probe and drogue technique on a large scale by the USAF was still in doubt and interest from the RAF was non-existent. 1950 would be a make or break year for the system and the future of Flight Refuelling Ltd.

Part 4:
The "Tripple Nipple" goes to War.

At the beginning of 1950 in-flight refuelling was a well-established technique in the USAF. The first Air Refuelling Squadrons, the 43rd at Davis-Montham AFB in Arizona and the 509th at Roswell AFB in New Mexico, had been in business since June 1948 using the looped hose system and Boeing was making great progress with its flying boom tanker. FRL's probe and drogue had been proven as a system that could be used successfully by jet fighters and work was under way at Tarrent Rushton to convert two B-29 bombers to HDU tankers and fit more bombers and F-84 Thunderjet fighters with receiver probes. However, without British government support FRL's attempts to compete on equal terms with Boeing were doomed to failure. The conversion work on the American aircraft was much more complex than anticipated and it brought FRL to the edge of bankruptcy to meet the contract deadline. The commander of SAC, General Curtis LeMay, was only interested in a system that could refuel bombers and Boeing's flying boom was his method of choice. FRL needed another demonstration of the advantages that their probe and drogue could give to jet fighters and the opportunity was not long in coming.
Back in July 1948 the USAF had deployed 16 F-80 Shooting Star jet fighters across the Atlantic from Selfridge AFB in Michigan to RAF Odiham. This was Operation "Fox Able One" (Fighters Atlantic One) and the movement took 10 days staging via Bangor, Goose Bay, Greenland, Iceland and Stornaway. In May 1949 "Fox Able Two" took 16 days over a similar route. USAF Col David Shilling had led "Fox Able One" and was very interested in any way that the deployment of jet fighters could be improved upon. The outbreak of war in Korea on the 25th June gave added urgency to finding a solution. At the 1950 SBAC Display at Farnborough he witnessed the first public demonstration of probe and drogue refuelling between FRL's Lincoln tanker and Meteor Mk 4 flown by Marks and Hornidge. In a novel experiment the R/T was relayed over the public address system. However, Hornidge's expletives when fuel washed over the Meteor's canopy on disconnect were heard by all! Col Schilling realised that probe and drogue solved the fighter deployment problem and was aware that two F-84s had been converted by FRL as receivers with wing mounted probes. He was granted authority to make an attempt at a non-stop flight refuelled crossing of the Atlantic by the F-84s.
The plan was to fly the pair of fighters from Manston to Mitchell Field in New York, refuelling from Lincoln and KB-29 tankers over Prestwick, Keflavik and Goose Bay. The first attempt "Fox Able Three" took place on the 19th September 1950 but was aborted after Schilling made the first contact and found that the Lincoln's HDU was malfunctioning. It was just as well since the Keflevik based Lincoln tanker had suffered a fire after take off and had to have an engine replaced. A second attempt, "Fox Able Four", was made on the 22nd September. The first RV went well but poor weather over Iceland led to problems joining up with the second Lincoln tanker. Practical RV techniques had yet to evolve and too much reliance was put on a special radar in the FRL tanker. Eventually it was agreed that the Lincoln and F-84s would orbit in the cone of silence from Keflevik radio until they were visual. Schilling made a successful contact but Col Bill Ritchie in the second F-84 had a problem. The refuelling probes fitted to the F-84s had a manually operated valve that had to be opened by the pilot after contact and closed before disconnect. Ritchie disconnected before the valve closed and damaged his probe. Despite this they decided to press on for the third tanker, the KB-29, near Goose Bay. Schilling and Ritchie encountered strong headwinds on the next stage and became tanker dependent before reaching Goose. Ritchie found he was unable to take on fuel with his damaged probe so he decided to climb for height and glide for Goose Bay before he flamed out. He broke out of cloud at 12,000 ft, over Lake Melville, 40 miles from the airfield. He was still 30 miles short when his engine died and after passing a message to his wife via Schilling he ejected at 3,000 ft. Schilling continued to make a landing in the US at Limestone (Loring) AFB in Maine after a flight time of 10 hours and 8 mins! Ritchie was rescued by helicopter after landing in a tree.
Despite the partial failure of the flight, "Fox Able Four" was a transatlantic first for jets and proved that the flight-refuelled deployment was practical. The lessons learned led to the development of an automatic probe valve. The Farnborough demonstration and the publicity surrounding "Fox Able Four" finally provoked the British Air Ministry into taking a closer look at flight refuelling. They decided to convert the 16 Meteor Mk 8 fighters of 245 Squadron, based at RAF Horsham St Faith in Norfolk, to probe receivers for refuelling trials with Lincoln tankers. These trials, code named "Operation Pinnacle", eventually began on the 8th May 1951 and continued until the end of October. Although completely successful, "Pinnacle" failed to convince the Air Staff who considered that the provision of tankers for the RAF would simply mean fewer fighters. This incredibly flawed logic could not be further from the opinion of the USAF who by this time had already used flight refuelling in combat!
By spring 1951 FRL had completed the conversion of the B-29s into a single point KB-29T and a 3-point YKB-29T. The YKB-29T was immediately nicknamed the "Triple Nipple" by the Americans and a demonstration of its capability was arranged. Two probe-equipped Meteor Mk 8s were borrowed from 245 Sqn and with FRL's Meteor Mk 4 made the first simultaneous 3 aircraft refuelling from the "Triple Nipple". The Americans were very impressed. They realised that numbers of fighters was not a factor when their targets in Korea were out of jet range from their bases in Japan. Flight refuelling would solve the problem and in June 1951 the single point KB-29T was flown to Yokota AB, Japan and attached to the 91st Strategic Reconnaissance Wing. There wasn't time to convert enough jet fighters to single point probe refuelling but USAF engineers at Wright Patterson AFB in Ohio came up with a solution. By fitting probes to their drop tanks the fighters did not need the additional internal plumbing which FRL had found so difficult with the F-84s. Lockheed, Republic and North American were tasked with converting the tanks on the F-80 Shooting Star, F-84 Thunderjet and F-86 Sabre. The swept wing and underwing tanks of the F-86 proved to be unsuitable but the straight wing and tip tanks on the F-80 and 84 gave no problems and conversions were soon underway. The refuelling technique required three contacts to avoid fuel asymmetry. The left tank would be half filled, then the right filled to full before the receiver topping up the left again. It was far from ideal but in the time available it was an acceptable system.
The program was code named "Project Collins" and culminated in the first flight refuelled combat mission in history on the 6th July 1951. Three RF-80As took off from Taegu in South Korea and rendevoused with the KB-29T tanker off Wonsan, North Korea. It was 210 nm from Taegu to Wonsan and the RF-80A had a range of 330 nm. After refuelling the 3 RF-80As were able to use their full range to photograph targets in the far north of Korea. At 0510 hours on the 28th September 1951 Major Harry Dorris took off from Japan in his F-80A, armed with a full war load of bombs, rockets and guns. He refuelled 8 times from the KB-29T tanker, between which he carried out a series of low level strikes on targets in North Korea before landing back in Japan at 1925 hours. He had established a new endurance record for jet aircraft of 14 hours and 25 minutes and done it under combat conditions!
The USAF now began a full-scale flight refuelling operation over Korea code named project "High Tide". Ten KB-29M looped hose tankers were released from SAC for conversion to a "quickie fit" probe and drogue tanker by FRL. Even the "Triple Nipple" got in on the act and was deployed to Japan. "High Tide" initially involved three F-84 squadrons of the 136th FBW, which were converted with tip tank probes for an operational test of large tactical units using flight refuelling. The program had three phases; training the 3 squadrons, establishing combat air patrols over Japan and finally carrying out strikes on North Korea. A number of units took part in "High Tide" including the Air National Guard squadrons of the 116th FBW. "High Tide" culminated on the 29th may 1952 when 12 F-84Es of the 159th FBW flew from Japan to carry out successful full scale ground attack strikes on North Korea, refuelling en-route from the KB-29T tankers. The USAF concluded from "High Tide" that although the wing tip refuelling had been a success it was only an ad hoc emergency substitute for a single point system. However, the value of flight refuelling to tactical aircraft operations was undisputed. Probe and drogue was now combat proven and the system of choice for tactical aircraft. However, the battle with the boom was far from over, indeed it had hardly begun.

Part: 5
The Battle of the Boom.

It had taken just three years for the probe and drogue AAR system to progress from Peter MacGregor's drawing board to use in combat during the Korean War. Although the RAF continued to disregard the advantages of AAR, the Project "High Tide" trials convinced the USAF that in-flight refuelling not only had a vital role with SAC's nuclear bomber force but also was of significant value in tactical and theatre jet operations. However, the question remained as to which AAR system the USAF should adopt, either the now combat proven (but British) FRL probe and drogue, or the All-American Boeing flying boom. Boeing had been working on their boom system since November 1947 and it must have been galling for their engineers to watch the rival probe and drogue system see action first. However, the Boeing boys had powerful support for the boom from SAC commander-in-chief, General Curtis Le May, who saw the boom tanker as his own pet project. In March 1950 the first of 116 KB-29P boom tankers entered service with SAC and by December that year Boeing had converted the first C-97 Stratocruiser with an improved boom. The first KC-97E boom tanker entered service with the 306th Air Refuelling Squadron at McDill AFB, Florida, on the 14th July 1951 and eventually 814 KC-97 boom tankers would be built for SAC by Boeing.
Although the USAF had been able to rush probe and drogue tankers into combat over Korea, the job of converting the relatively complex American jet fighters like the F-84E as single point probe receivers had proved to be more difficult than expected. The initial contract had almost driven FRL to bankruptcy and forced Cobham to sell the probe and drogue manufacturing rights to the Americans. The USAF engineers' interim solution of twin tip-tank probes worked after a fashion but required pilots to make 3 separate contacts per onload to ensure fuel asymmetry was kept within limits. In 1951 Republic Aviation, who built the F-84 Thunderjet, began production of the F-84G model with a redesigned fuel system for single point refuelling. The F-84G was the first single seat fighter-bomber with the ability to carry nuclear weapons. It was equipped as an AAR receiver to enable it to carry out long range missions as part of the SAC nuclear strike force. However, Republic had not opted for a probe but had fitted a boom receptacle in the leading edge of the left wing. SAC's rapidly growing fleet of boom tankers and pressure from Le May undoubtedly influenced this decision.
On July 4 1952, sixty F-84Gs of SAC's 31st Fighter Wing took off from Turner AFB, Georgia, and 1,800 NM non-stop to Travis AFB, near San Francisco, California. They were refuelled over Texas by 24 KB-29P boom tankers. This operation was a dress rehearsal for Fox Peter One (Fighters Pacific One), the transpacific movement with AAR support of all three squadrons of the 31st FW to Japan.
The Fox Peter One deployment was led by Co, David Schilling, the pioneer of the first transatlantic AAR trail in 1950, Fox Able Four, which used the probe and drogue technique. The plan was for the F-84s to be refuelled on the long 1.860 NM leg from California to Hawaii and then island hop via Midway, Wake, Eniwetok, Guam and Iwo Jima to their destination, Misawa and Chitose Air bases in Japan. KB-29P tankers would orbit at 18,000 ft in position over Weather Station Alpha with 6 backup tankers over Weahter Station Uncle a few hundered miles east of Hawaii. The first Squadron crossed on 6th July 1952 followed by the other two waves on subsequent days. All the F-84Gs arrived in Japan on the 16th July, just 12 days after leaving Georgia. The crossing was substantially faster than the time taken to disassemble the aircraft and ship them across the Pacific as deck cargo, which had been the only option up till then. On the 29th July the first non-stop jet crossing of the Pacific was made by an RB-45C Tornado flying from Elmendorf AFB, Alaska to Yokota AB in Japan. The Tornado jet bomber was refuelled twice by KB-29P boom tankers. Both this crossing and the Fox Peter One deployment were intended to give the Soviet Union notice of the new USAF rapid reinforcement capability bought about by AAR. In October "Fox Peter Two" saw the F-84Gs of the 27th Fighter Escort Wing make the crossing and transpacific "Fox Peter" movements became routine in 1953.
It is interesting to note that despite the growing force of Boeing boom tankers, there were still numbers of KB-29M tankers using FRL's looped hose system in service with SAC. Indeed in December 1952 60 aircraft of the 301st Bomber Wing deployed to RAF Brize Norton on 90 days TDY, including 15 KB-29M looped hose tankers. These were the first AAR tankers to operate from RAF Brize Norton, which must be one of the few air bases in the world to have seen all three AAR systems in operational service!
As the more capable KC-97 boom tanker began to enter service in numbers the USAF's ability to mount routine transoceanic deployments began to be exploited more fully. On the 20th August 1953 SAC carried out Operation Longstride, a double transatlantic deployment of 28 nuclear capable F-84G Thunderjets from Turner AFB to bases in the UK and North Africa. Once again Col Schilling led the first wave of 8 fighters from the 31st Fighter Escort Wing, flying 3,800 NM via KC-97 refuelling tracks over Bermuda and the Azores, to land at Nouasseur AB in French Morocco 10 hours and 20 minutes later. The second wave of 20 F-84Gs from the 508th Fighter Escort Wing, led by Col Thayer S Olds, took off a few minutes after Schilling and refuelled 3 times from KC-97s over Boston, Goose Bay and Keflavik. Three fighters diverted into Keflavik but 17 landed at RAF Lakenheath, 11 hours and 20 minutes after leaving Georgia. Operation Longstride showed just how far AAR deployment techniques had progressed since Col Schilling made his first crossing in 1950.
As capable as the KC-97s were, they were hard pressed to match the performance of the Boeing B-47 Stratojet bombers now entering service. To refuel from the KC-97 the B-47 had to descend from 35,000 ft to 18,000 ft where it could begin to tank. As the B-47 filled and got heavier the KC-97 would have to enter a shallow "toboggan" decent to keep the bomber in contact and could go as low as 12,000 ft. The B-47 would then have to use half of the fuel it had just received just to climb back to its cruising altitude. The SAC bomber crews considered the "toboggan" manoeuvre with the KC-97's flying boom to be difficult at the best of times and potentially dangerous in poor weather conditions. Clearly the solution was to have a jet tanker with similar performance to the Stratojet bomber.
The quickest and easiest solution was to fit a B-47 with a HDU in its bomb bay for probe and drogue tanking. Early in 1953 the USAF engineers at Wright-Patterson AFB converted two B-47s for probe and drogue trials, the tanker becoming a KB-47G and the receiver a YB-47F. Considering it was a cheap "off the shelf" solution the KB-47G was a good tanker. Fitted with bomb bay tanks it could carry 59 tonnes of fuel, which gave it an AAR capability similar to the much later RAF Victor K2 tanker. To many in the USAF at the time, probe and drogue had a number of significant advantages over the Boeing flying boom. It was simpler, lighter and cheaper than the boom and imposed much less aerodynamic drag for the tanker making it more fuel-efficient. Its simple operation did not require the specialist skilled boomer of the Boeing system and there was potential to rapidly modify a fleet of KB-47 jet tankers to support SAC's new B-52 Stratofortresses about to enter service. However, although the KB-47G tanker and YB-47F receiver made the first jet to jet AAR contact on 1st September 1953 and the trials proved that the system could be used successfully by the B-47, it did nothing to change General Le May's prejudices against probe and drogue.
Le May believed that his pilots would have great difficulties controlling the sluggish 200 tonne mass of a B-52 while making contact with a small and unstable drogue and also considered that rubber hoses would be unreliable in the low temperatures above 30,000ft. Although these problems proved to be generally unfounded the boom did enjoy one undeniable advantage over a hose system. The wider pipe diameter of the boom would always transfer fuel faster than the smaller hose if the receiver were capable of taking it at the higher rate. The optimum transfer rate of the probe and drogue system at the time was 250 gallons a minute but the boom could pass 700 gallons per minute. However, the boom-equipped tankers could only refuel one aircraft at a time whereas the "Triple Nipple" probe and drogue tanker had refuelled 3 fighters at once. Also the tactical fighters could not take on fuel at the full rate of the boom system.
In February 1953 the USAF carried out a fly-off trial between the flying boom and probe and drogue tankers at Offutt AFB, Nebraska. A single KC-97 was fitted with a HDU for the trial and a variety of aircraft equipped as probe receivers. The outcome of the trial between the two systems was entirely predictable. Fighter pilots preferred probe and drogue while the bombers opted for the boom. The trouble was that all the tankers belonged to SAC, which was General Le May's train set. Unsurprisingly, the Boeing test pilots that carried out the trials considered that the probe and drogue would be unsuitable for use by the B-52. SAC also preferred the KC-97 to the KB-47 because the converted transport could also be used to deploy personnel and equipment to its overseas bases. The Boeing lobby prevailed and SAC went with the flying boom. As far as Le May and SAC were concerned the battle of the boom was over. The probe and drogue had lost and the system would have no place in SAC's growing tanker force. However, outside SAC General Le May's opinions held less sway. Already plans were being made for rival probe and drogue tanker fleets in USAF Tactical Air Command and the US Navy, and across the Atlantic the RAF began to realise that they could no longer ignore AAR in their future plans. Probe and drogue was just too good to fade away.

Part: 6
Tankers Aweigh!

When SAC, the progenitors of military AAR, committed its growing fleet of tankers to the Boeing flying boom system there appeared to be little future for the probe and drogue technique within the USAF. SAC's maverick commander, General Curtis Le May, had ordered over 800 of the piston engined Boeing KC-97 boom tankers and its planned jet replacement would also be equipped solely with the Boeing system. With the Cold War nuclear bomber force dominating USAF doctrine for the foreseeable future, the tactical airpower lessons of the Korean War were soon forgotten or ignored by the air force.
However, it was the USAF's rival in global airpower, the US Navy, that would next embrace probe and drogue AAR and go on to become its greatest champion. US Naval Aviation had suffered a severe blow to its strategic plan when the giant 100,000 ton aircraft carrier USS United States was cancelled in April 1949. The large Neptune atomic bombers intended for the carrier were forced to languish ashore and the US Navy's nuclear aspirations were in disarray. When the Korean War broke out the US Navy entered the fray with 30,000 ton straight decked WWII carriers, which soon proved impractical for intensive jet operations. The early Navy jets were underpowered and thirsty. The weights that allowed safe carrier take-offs and landings left them with only a small margine for weapons and fuel, which limited their mission flexibility and left them vulnerable to attack by Communist Mig15 fighters. After initial carrier deployments of Grumman F9F-2 Panthers jets in July 1950, it was piston engined types such as the AD-4 Skyraiders and F4U Corsairs that were to dominate US navy flight decks throughout the conflict.
By 1950 it was clear that jets would get much bigger and faster and to operate them successfully at sea would require nothing short of a revolution in aircraft carrier design. Ironically it was the British, the poor relations to American naval power, that would provide the technical answers and lay the foundations of the fleet of US Navy supercarriers that exist today. Despite the severe economic restrictions after WWII the Royal Navy had embarked on an intensive R&D programme aimed at solving the problems of carrier jet operations. Initially, much money and effort was spent on developing a flexible rubber flight deck that would allow jets to land without undercarriage but this concept was a technological dead end. A more practical development was the powerful steam catapult first installed on HMS Perseus in 1951. This allowed faster launch speeds at much heavier weights than before. The second great breakthrough was the angled flight deck, first suggested on the 7th August 1951 by Captain Denis Campbell at a Ministry of Aviation conference on methods of using the flexible flight deck. This innovation immediately eliminated the danger of a catastrophic crash into parked aircraft if a jet missed the arrester cable. The third development was Lt Cdr Nick Goodhart's gyro-stabilised mirror landing system. At higher jet landing speeds there was too much lag in the WWII system of hand signals from a "DLCO" on the flight deck. The mirror sight allowed pilots to monitor their own approach and react promptly to deviations in the flight path. The combination of steam catapult, angled deck and mirror landing sight made carrier operations by high performance jets possible. However, it was another British development that would help make carrier jet operations a practical military proposition.
The US Navy was aware of the USAF's tactical AAR trials over Korea in 1951 and saw this as a solution to the weight limitations, which would always be a factor of carrier operations. It was clear that probe and drogue was the only system the US Navy could use, as a boom tanker would be far too large for any carrier. In June 1952 a North American AJ-1 Savage bomber was fitted with a FR Inc A-12 HDU in its bomb bay to become the first carrier borne tanker. An F9F-2 Panther jet was fitted with a probe for trials which demonstrated the advantages of AAR to naval aviation. The Savage tankers first deployed to sea with VC-9 on the USS Midway in 1953. Now more take off weight could be dedicated to weapon loads with aircraft topping up from the tankers orbiting over the carrier before proceeding on their missions. The tankers could also help aircraft recovering to the carrier short of fuel to prevent them from diverting to airfields ashore. AAR also vastly increased the range and endurance of carrier based jets to a much larger degree than land based aircraft. The US Navy began a program to equip a number of jet fighter types with refuelling probes and carried out demonstrations of the advantages it had gained. On the 1st April 1954 the first USA transcontinental flight in less than 4 hours was made by three pilots of VF-21 in F9F Cougars in a 2,438-mile flight from San Diego to New York, refuelling from AJ-1 Savage tankers over Kansas. These successful trials and demonstrations convinced to US Navy that probe and drogue AAR was a vital element in the future of naval airpower. On the 1st September 1955 the US Navy announced that all naval fighters in production would be fitted with probes for in-flight refuelling, establishing the technique as a standard operational procedure.
One of the most impressive tanker aircraft ever produced was the US Navy's huge Convair R3Y-2 Tradewind 4-engined turboprop flying boat transport. It was equipped with four wing mounted HDUs and became the first (and only) tanker to simultaneously refuel 4 aircraft, US Navy Grumman Cougar jet fighters, in September 1956. However, this enormous tanker was of little use to carrier air wings operating far from shore and even the KAJ-1 Savage was too large to operate from most US Navy carriers in the early 1950s. To provide an on board AAR capability until the new supercarriers of the Forrestal class entered service the Douglas aircraft company developed the D-704 refuelling pod which was initially carried by the AD-6 Skyraider but could be fitted to any carrier aircraft turning it into a "buddy-buddy" tanker. The D-704 pod represented the biggest advance in AAR since the invention of probe and drogue. When the tiny nuclear capable A4D-2 Skyhawk entered service in 1957 "buddy-buddy" tanking gave it a range of almost 2,000 miles and gave the US Navy a nuclear delivery system easily able to operate from any of its carriers. The US Navy regained a strategic strike role it had thought lost since the cancellation of the USS United States.
The US Navy had the resources and the enthusiasm to rapidly adopt all the British innovations in naval aviation and by 1957 it had converted most of its existing carriers to the new design as well as commissioning a fleet of new large jet supercarriers. The Royal Navy, on the other hand, had been constrained by British economic difficulties and could only convert four relatively small WWII designed carriers to modern jet operations. With their much smaller aircraft complement the RN carrier wings would clearly benefit from significant force multiplication if they adopted "buddy-buddy" AAR. In 1958 FRL developed the Mk 20 refuelling pod for use with the Supermarine Scimitars and DeHavilland Sea Vixens about to enter RN service. The new jets quickly obtained their receiver clearance after tanking from a Canberra equipped with the new pod. In 1961 Mk 20 pods were delivered to 899 Sqn at FAA Yeovilton for use with their Sea Vixens which became the first RN tankers. 899 Sqn were able to demonstrate "buddy-buddy" tanking at the 1961 Farnborough air show, just a week after the delivery of their first Mk 20 pods. Following the US navy lead the Blackburn Buccaneer incorporated an AAR probe at an early stage of its design. The initial retractable probe was replaced by a fixed probe after trials revealed that the original design disrupted the airflow to the starboard engine. When the Buccaneer S1 entered service with 800 Sqn in 1964 the unit had a dedicated flight (800B Flt) of four Scimitar "buddy-buddy" tankers. Early plans to convert some Buccaneers to tankers called for a Mk 21 HDU pack to be fitted in its bomb bay but this was abandoned in favour of a Mk 20 pod under the starboard wing. The superb Buccaneer retained its secondary "buddy-buddy" tanker role throughout its long RN and RAF service.
The commissioning in the US Navy of the 70,000 ton Forrestal class carriers now enabled much larger and faster jet aircraft to operate at sea. In 1956 the US Navy finally got a long-range jet bomber in the Douglas A-3D Skywarrior. This large and adaptable aircraft would serve in many roles in the US Navy until 1991 and even proved able to operate from the smaller modernised WWII Essex class carriers. Supersonic fighter operations also began with the entry into service of the Vought F8U-1 Crusader. In June and July 1957 US Navy Crusaders made high speed in-flight refuelled transcontinental flights culminating in Project Bullet on the 16th July when Marine Major John Glenn flew his Crusader from Los Angeles to New York at an average speed of Mach 1.1. Glenn and his wingman, Lt Cdr Charles Demmler, refuelled en-route from a piston engined Savage tanker, but the high performance Crusader had problems at the Savage's slow refuelling speed and Demmler damaged his probe forcing him to abandon his attempt. By the early 1960s the faster jet Skywarrior had replaced the Savage in the AAR tanker role and was capable of lifting a 10 tonne fuel offload. The Skywarrior tankers would prove to be a vital asset for the US Navy. The importance of AAR was about to be conclusively demonstrated during air operations over a little known Asian country called Vietnam and for the second time probe and drogue would be used in combat.

Part: 7
The TAC Tankers.

The US Navy's rapid adoption of probe and drogue AAR kept the system alive and secured its development in the face of SAC's overwhelming championing of the flying boom. However Tactical Air Command, the USAF's second combat arm, had been the instigators of the Project "High Tide" combat trails over Korea and recognised the many advantages that AAR offered tactical airpower. The SAC tanker force was growing rapidly with over 800 KC-97 entering service and over 700 of the revolutionary Boeing KC-135 jet tankers on order from 1954. Unfortunately these were dedicated almost exclusively to supporting SAC's expanding force of B-47 and B-52 nuclear bombers. After Korea the concept of all out nuclear warfare dominated USAF doctrine and its senior commanders critically undervalued the role of tactical forces, a factor that would cost dearly in the Vietnam conflict. If TAC were to fully exploit the advantages of AAR it would have to get its own dedicated fleet of tankers.
The new "Century Series" of tactical fighters, the F-100, F-101, F-104 and F-105, were all equipped with refuelling probes. The F-105 Thunderchief actually had both a retractable probe and a boom receptacle due to its original development as an escort fighter for SAC bombers. To service these aircraft TAC initially looked to the KB-47 Statojet tanker originally converted for the 1953 probe and drogue trials. The KB-47 would have been an excellent probe and drogue tanker, able to fly at fast jet deployment heights and speeds and carrying a respectable fuel load in bomb bay tanks. On July 23, 1956, the USAF authorised the development of a KB-47 two-drogue prototype tanker. However, the cost of the KB-47 conversion promised to be excessively high and SAC was reluctant to release sufficient airframes from its bomber force. The USAF cancelled the KB-47 program on July 11, 1957.
TAC was left with converting the B-50D piston engined bombers that had been replaced by the B-47 in SAC's strike force. Plenty of B-50D and TB-50H airframes were available but they were too slow for modern fighters and had performance problems at the high take-off weights required for world wide tanker operations. To equip the KB-50s as tankers it was decided to use the 3-hose system of the original "Tripple Nipple" KB-29 converted by FRL and used successfully in Korea. The KB-50 was fitted with a Mk 12 HDU in the tail and two further Mk 12 HDUs in wingtip pods. Each HDU had a 75-ft hose and could transfer up to 1,000 kgs per minute. The KB-50's piston engines were fed with AVGAS from its five wing tanks and it carried 13,000 kg of JP-4 jet fuel in its bomb bay tanks. An additional further 3,500 kgs of JP-4 could be uplifted in its centre wing tank. At that time the total internal and external fuel capacity of an F-100 was just 5,000 kg so a single KB-50 was easily capable of giving a 50% top up to a flight of 6 fully armed F-100s. The widely spaced HDU fit also enabled 3 fighters to refuel simultaneously, albeit at a reduced flow rate. As the photographs show, it was quite routine for the TAC KB-50s to refuel fighters and tactical bombers three at a time.
In 1956 the Hayes Aircraft Corporation of Birmingham, Alabama, began converting 136 TB-50D, TB-50 and TB-50H into KB-50 tankers, the first of which entered service with TAC in 1957. To solve the speed and performance problems the KB-50s were fitted with two auxiliary 5,200 LB thrust GE J47 turbojets, which allowed an increase in cruising speed for tanking from 230kts to 300kts at 25,000 ft and greatly assisted take-off performance. The jet assisted KB-50J first flew in December 1957 and entered service with TAC early the following year. With a fleet of KB-50Js gave TAC the ability to deploy its aircraft between theatres. KB-50J tankers based at Bermuda and Lajes enabled TAC fighters to rapidly deploy to NATO bases in France and Germany via the safer southern transatlantic route rather than risky staging operations via Goose Bay, Greenland and Iceland. Transpacific deployments to Japan and Okinawa took place via Hawaii and Wake Island. The tactical role of the KB-50J saw numbers employed with both the USAFE in Europe and with PACAF in the Far East where the KB-50J's ability to operate from the shorter runways at TAC airfields gave it an advantage over the heavier KC-135.
The KB-50J was never more than a short-term solution to TAC's AAR requirements. Although the last B-50 had come off the production line in 1953, the 1940s wartime design was worn out and suffering from fatigue and corrosion by the early 1960s. TAC cast envious eyes on SAC's growing fleet of new KC-135 tankers. By 1961 SAC possessed a fleet of 1, 095 tankers to support its bombers and the USAF decided it would have to share its ample assets with TAC. In exchange SAC would gain control of all future USAF tankers including the planned fleet of 620 KC-135s, 32 squadrons in all. On the 17th November 1961 the deal was done and TAC was promised that at least one wing of SAC KC-135s, about 70 aircraft, would be available for its use if required. TAC's probe and drogue jets would have to get used to refuelling from the Boom Drogue Adapter (BDA) used by the KC-135s.
The last few TAC dedicated KB-50Js served in the PACAF with the 421st ARS, based at Yokota, Japan. However, the KB-50J has one claim to glory, it was the first tanker to see use in the Vietnam War. In May 1964 the United States began flying low level reconnaissance missions with USAF RF-101 and US Navy RF-8 jets over the Plain of Jars in Laos in order to trace the North Vietnamese infiltration routes into South Vietnam. This was the period before formal US involvement in the fighting and PACAF had very few assets available to it. On the 6th June a US Navy F-8 Crusader was shot down over Laos and the US government decided to retaliate with air strikes against communist Pathet Lao anti-aircraft sites. The Plain of Jars strike was flown on the 9th June using 8 F-100D Supersabres and an RF-101 pathfinder flown from Da Nang and these were refuelled by four KB-50J tankers from the 421st ARS deployed in the Philippines. At the time this was considered an isolated event but just two months later the Gulf of Tonkin incident would lead to full involvement of the US military in the growing Vietnam conflict. SAC KC-135s began to arrive in theatre in August 1964 and an initial deployment of 6 tankers would grow to a peak of 172 KC-135s in 1972.
The last KB-50J, the last USAF pure probe and drogue tanker, was retired from the 421st ARS in January 1965. Deemed too corroded to fly back to the USA for scrapping, most KB-50Js were allowed to rot on the fire dumps in at their home bases. Their jet engines were removed and fitted to ex-SAC KC-97 boom tankers that were now relegated to service with the Air National Guard. With the passing of the KB-50s the three-point AAR tanker, which had been proven in combat over Korea and now Vietnam, would not be seen in USAF service again for almost 30 years. It would be the RAF, belatedly adopting AAR in 1958, who would next put a "Triple Nipple" tanker into service.

Part 8.
V is for Tanker.

The first decade of probe and drogue AAR saw its successful use in combat during the Korean war and growing fleets of airborne tankers entering service with the armed forces of the United States. 136 Boeing B-50 bombers were being converted into three point KB-50 tankers for the USAF Tactical Air Command and Douglas KA-3B twin jet tankers were serving aboard the new class of 70,000 ton US Navy aircraft carriers. The British invention was paying enormous dividends in the rapid development of American military and naval aviation in the 1950s.
The British Air Ministry's 1947 assertion that in-flight refuelling was "not a paying proposition" was proving to be a ludicrously unimaginative policy and by the mid 1950s even their "Airships" in Whitehall had to take notice of the emerging role of AAR in modern airpower. Whether it was the rapid growth of Strategic Air Command's tanker forces or the steady bombardment of lobbying letters from Sir Alan Cobham that finally changed the minds of the AAR "Luddites" in the RAF is not known. However, by January 1954 the Air Staff had decided that most of the Valiant and all the Vulcan and Victor bombers on order for the RAF's nuclear "V" Force should be capable of flight refuelling.
Bomber Command's tactical doctrine in the early 1950's differed little from that used against Germany in 1945. Indeed the Valiant and Canberra air raids on Egypt during the Suez crisis in 1956 still relied on pathfinder marking and bombing techniques used by Lancaster crews a decade earlier. The Air Staff realised that the nuclear "V" Force would have to use completely different tactics and procedures which would inevitably be modelled on those of the much larger and more experienced Strategic Air Command. SAC's heavy reliance on AAR to carry out its mission must have influenced the RAF to change its opinion on flight refuelling.
The RAF's flight refuelling requirement differed from the Americans'. SAC's B-52 bombers could not reach their targets from the USA without AAR whereas the RAF's new "V" bombers had the range to reach their targets directly from the UK at high altitude. RAF and USAF aircraft had been able to fly unmolested at high level over the Soviet Union for some years on secret reconnaissance missions and many hoped that the high performance of the "V" bombers would be enough to get them through. However, as the Soviet air defences grew in sophistication and capability it became clear that performance alone would not be enough. The "V" bombers would now need to carry heavy ECM jammers and be able to out-flank the Russian fighter and missile bases protecting their target areas. A flight refuelling capability offered solutions to the increased AUW and range problems that the "V" Force was bound to encounter as it developed.
Despite the Air Staff's new commitment to AAR no mention was made of a dedicated tanker force. In April 1955 it was vaguely hoped that all Vulcans and Victors would be equipped for rapid conversion from bombers to tankers as required. However, Bomber Command was reluctant to release any of its aircraft from its Main Force for use as tankers and ordering additional aircraft for the role was unacceptable. The question of a tanker force for the RAF remained unanswered until early 1957 when a plan for the future composition of the "V" Force envisaged a fleet of 184 bombers made up of 120 Mk 2 Vulcans/Victors, 40 Mk 1 Vulcans/Victors and 24 Valiants. No aircraft were set aside specifically as tankers but it was intended that as the Mk 1 Vulcans and Valiants were withdrawn from the Main Force they could be converted to tankers. 32 sets of bomb-bay tanks had been ordered for the Valiants for delivery in March 1958 and Vulcans were to be equipped with tanker fittings from the 26th aircraft on the production line. At the time there was no intention for the Victor to be used as a tanker, surprising in the light of future developments.
During 1957 the difficulty for those responsible for planning the future size and composition of the "V" Force was the necessity to maintain a large strike force to guarantee that a significant percentage would penetrate Soviet defences. Most of the Mk 1 "V" bombers were allocated to fringe targets in support of the deep strike Mk 2 aircraft and few would be available for use as tankers. Many had underestimated the problems involved in developing an effective flight refuelling capability within existing resources. The Treasury doubted that the RAF fully understood what the future size, composition and operating methods of the "V" Force would be and were strongly against additional funds for tankers. The future of AAR in the RAF hung in the balance.
The situation was clarified in a note from the VCAS, Air Marshal E C Hudleston, to the Air Council on the 6th December 1957. In it he stated that flight refuelling was essential if the "V" Force was to be given the maximum tactical freedom in routing to maintain its viability as a deterrent. In the note the advantages of AAR for the RAF were listed. Bombers could route to avoid heavily defended areas and reduce penetration distances, vital targets beyond normal operating ranges could be attacked and a measure of tactical surprise achieved. The note also envisaged that flight refuelling made possible rapid deployment to overseas bases and avoidance of unfriendly territories en-route. It also enabled the "V" Force to operate from shorter runways and be based beyond the radius of action of an enemy. HQ Bomber Command estimated that a tanker force of four squadrons (32 aircraft) would be required but financial constraints reduced this recommendation to two squadrons of Valiants plus a shadow unit formed by the Valiant OCU.
The Air Council accepted VCAS' recommendations and approached the Treasury for authority to resume work on development of the Valiant tanker. By January 1958 the Treasury had only authorised minimal short-term development work on the Valiant tanker and they would not commit themselves to funding any additional aircraft for a tanker force. In November 1958 they insisted that any aircraft used as tankers should be found within the approved "V" Force establishment of 144 bombers and they would only sanction further development of the Valiant if the Air Ministry agreed not to develop the Vulcan or Victor in the tanker role. The Air Ministry could not accept any reduction in the 144 Vulcan and Victor bombers they considered the minimum to provide a viable deterrent force and they pressed the Treasury for approval to retain additional Valiants as tankers. To promote their case they stated in January 1959 that the tankers would not only be used just to support "V" Force operations but would also be essential in overseas deployments and reinforcements by Fighter Command and tactical transport aircraft.
Again the Treasury prevaricated and in April 1959 letters were exchanged between ministers. Once again the RAF's case for tankers was put, this time adding the advantages that AAR would bring to the new Lightning and TSR2 aircraft in development and requesting funding for two squadrons of Valiants. However, the potential size of the RAF's AAR commitment now alarmed the Treasury who, with incredibly convoluted logic, considered that it would be too much for just 16 Valiants and that this was just the thin end of a wedge of further expenditure for a much larger tanker force. If this were required then the Air Ministry would have to find further cuts to compensate. The MoD quickly replied that, while it could not guarantee that more tankers would not be needed in the future, there were no plans at present to go beyond the 16 Valiants requested. Finally, on the 27th April 1959 the Treasury approved the establishment of a 16 Valiant tanker force as an addition to the 144 "V" bombers in the Main Force. They stated that this would add £2.8 million a year to Bomber Command's running costs and asked the RAF to consider reducing the number of its aircraft equipped as receivers so "they might bear a closer relation to the size of the tanker force."
Thus it was that, just over ten years since probe and drogue AAR was first demonstrated by FRL, the RAF finally gained approval for a fleet of tankers. For more than a decade the British had argued over the costs of AAR with little regard to the proven benefits. The 16 Valiants authorised appear laughably inadequate when set against over 1,600 tankers in service or on order for the USAF at the time, but it was a start. Now the RAF had to put flight refuelling theory into practice, indeed the work had already begun.

Part: 9
Valiant Efforts.

As the Air Ministry and the Treasury bickered over the cost and composition of the RAF's AAR tanker requirement the engineers and aircrews worked at developing the hardware. After the Air Staff's initial dismissal of in-flight refuelling as a waste of money and resources, they began to see the light as early as 1952 and gave FRL the go-ahead for initial AAR design studies for the V-Force. Flight trials began again in May 1953 when Canberra B2, WH734, was loaned to FRL. It was equipped with the prototype Mk 16 HDU being developed for use on the Valiant and became the first British jet tanker. In early 1954 the single prototype Vickers Valiant B2 undertook formation trials with the Canberra HDU tanker and in 1955 the first British all jet in-flight refuelling was carried out between the Canberra and one of the Meteor F8s from the 1951 RAF fighter AAR trials. Later that year a Gloster Javelin FAW4 was fitted with a wing-mounted probe and flew some inconclusive dry contact trials with the Canberra tanker. In later trials the Javelin FAW9 was equipped with a longer fuselage-mounted probe that proved more successful. A production Valiant B1, WZ376, was fitted with the Mk 16 HDU in 1956 and began flying a series of AAR trials from A&AEE Boscombe Down. The first Valiant-Valiant trials began in 1957 and by February 1958 the Valiant tanker and receiver were granted an official release to service for day and night AAR.
In December 1957 No 214 Sqn at RAF Marham, equipped with Valiant bombers and commanded by Wg Cdr M J Beetham, was selected to carry out the first series of in-service AAR trials. The squadron F540 recorded that preparations were underway to convert the entire squadron to the tanker role, a gloomy and unpopular prospect for the proud bomber crews. In January 1957 the first crews were detached to A&AEE Boscombe Down and FRL's base at Tarrent Rushton on AAR familiarisation courses and in February two 214 Sqn Valiants, WD869 and WD870, completed initial flights equipped with the Mk 16 HDUs. In March 1958 formal trials began codenamed "Trial Numbers 306 and 306A/B". Trial 306 was to test the compatibility of the Valiants tanker and receiver equipment and Trial 306A/B was to develop the initial AAR RV procedures and techniques. The modifications to 214 Sqn's Valiants included the fitting of a nose mounted refuelling probe, installation of the Mk 16 HDU and a 2,000kg fuel tank in the bomb bay, cockpit modifications to the navigator radar's station for a HDU control panel and fitting external flood lights to give a night AAR capability. The modifications allowed the Valiant tanker to transfer up to 20,000kg of fuel at just under 2,000kg a minute. During the initial period of the trial only dry contacts were attempted and there were a few compatibility problems with the Valiant's radar and electrical equipment. Receiver conversion training of the 214 Sqn crews was carried out by Boscombe Down and Vickers test pilots, including Brian Trubshaw who later piloted the prototype Concorde. FRL engineers gave ground instruction. One of the first recommendations of the trial was the establishment of dedicated RAF flight refuelling ground school at Marham, which later became the Air to Air Refuelling School (AARS).
During September 1958 214 Sqn Valiants gave demonstrations of jet aircraft AAR at Farnborough and a number of "Battle of Britain" airshows around the UK. In October a flight refuelling procedures simulator became fully operational at Marham and AAR crew training began in earnest. By January 1959 214 Sqn was ready to begin wet contact trials. From the 26th January to the end of the month four crews carried out 26 day and 17 night wet contacts to prove the system. In February the first of a series of long duration flights using AAR began when Wg Cdr Beetham's crew completed a 12-hour sortie using AAR. When the final treasury approval for an RAF tanker force of 16 Valiants was granted in April 1959, 214 Sqn began a series of highly publicised record breaking long range flights to East Africa. Valiant tankers prepositioned at Malta, Tripoli in Libya and Nairobi and on the 15th April Wg Cdr Beetham and his crew flew a non-stop sortie from Marham to Salisbury (Harare) Rhodesia (Zimbabwe). The distance of 5,320 miles was covered in a record time of 10 hours and 12 minutes at an average speed of 522 mph. 214 Sqn continued to demonstrate the increased ranges made possible with AAR techniques with record flights by the Beetham crew to Johannesburg in June and to Capetown in July. The London to Capetown flight covered 6,060 miles in 11 hours and 28 minutes at an average speed of 530 mph.
In the autumn of 1959 214 Sqn Valiants began refuelling Vulcan B1As, newly equipped with in-flight refuelling probes. 101 Sqn based at RAF Finningley was the first Vulcan unit to carry out receiver training and the first dry contact was made by Flt Lt W S Green in XA910 on the 28th October. Reports on Trial 306 were published on the 30th November 1959 and this is the date when the RAF officially established an operational AAR capability. Trial 306 was completed on the 25th May 1960 when 214 Sqn Valiant WZ390 was flown flight refuelled non-stop from Marham to Singapore, covering 8,120 miles in 15 hours 35 minutes, the longest point to point flight made by an RAF aircraft up to that time. In June the Gloster Javelin FAW9s of 23 Sqn began receiver training, culminating in August with the RAF's first fast jet AAR trail by Javelins to Akrotiri. In October 1960 four 23 Sqn Javelins were trailed from the UK to Butterworth, Malaya, in five days, via Akrotiri, Bahrain, Mauripur and Gan. In December the Javelins of 64 Sqn and Vulcans of 617 Sqn began their receiver qualification.
Planning went ahead for a spectacular demonstration of the RAF's new global AAR capability involving a non-stop UK to Australia flight by a Vulcan B1A. On the 20th June 1961 nine Valiant tankers of 214 Sqn deployed to Cyprus, Pakistan and Singapore to support the record flight from Scampton to Sydney by a 617 Sqn Vulcan flown by Sqn Ldr Michael Beavis. At this time it was also decided to abandon the RAF Mk 6 probe and drogue and standardise on the NATO Mk 8 system, which was used by the USAF and US Navy probe-equipped aircraft. Joint RAF/USAF exercises demonstrated the Valiant tanker's compatibility with USAF TAC B-66, F-100 and F-101 receivers and even allowed the Valiant to refuel from the piston engined KB-50 tanker.
In August 1961 90 Sqn, based at RAF Honington, began training in the tanker role and became as the second operational RAF tanker unit in October. The formation of a third Valiant tanker squadron was proposed to give additional AAR support for the RAF transport force and approval sought from the treasury. However, by this time the operational limitations of the Valiant tanker were becoming apparent. The Valiant's small transferable fuel load and single hose required multiple tankers to support even a single fighter on most trail stages. Also its performance was seen as only marginally compatible with aircraft such as the Lightning and it was thought to have insufficient speed to refuel the TSR2. The crucial shortcoming was that the Valiant's planned fatigue life would only last until 1968 and this alone would dictate its replacement with a more advanced aircraft. Studies began on the conversion of the Handley Page Victor as a tanker and plans for a third tanker squadron were put on hold.
The Valiant tankers of 214 and 90 Sqns continued to support RAF exercises and operations and officially became single role tankers squadrons on the 1st April 1962, finally losing their bomber commitment. The Valiants carried out operational AAR training with 50 Sqn's Vulcans, 57 Sqn's Victors and 56 Sqn's Lightnings as well as the Sea Vixens and Scimitars of the RN FAA. V-Force trials included using the new tanker force to test airborne command post communications with the main bomber force and maintain a continuous airborne alert by Vulcan bombers, which lasted fourteen days! In July 1963 twelve Valiant tankers from 90 and 214 Sqns deployed to Libya, Aden and Gan to support Operation Walkabout, the non-stop flight by three 101 Sqn Vulcan B1As from Waddington to Perth, Australia. In March 1964 tankers from both squadrons trailed four Javelin FAW9Rs from Binbrook to Butterworth, staging via Aden and Gan. The long Indian Ocean legs required very accurate refuelling plans that pushed the capabilities of the single-point Valiants to the limit.
Despite their limitations the Valiant tankers were proving to be an indispensable asset to the RAF and the aircrews of 214 and 90 Sqns completed valuable work in developing the AAR procedures and techniques which are still in use today. All was going well until the discovery in August 1964 of major fatigue cracks in the Valiant's rear spar. At first it was hoped that most of the fleet could be repaired but in early December even bigger cracks were discovered in the front spar and all Valiants were grounded. Investigations showed that 60 out of 61 RAF Valiants suffered from major fatigue damage and it quickly became clear that non of them were safe to fly. The RAF finally announced the withdrawal from service of the Valiant on the 26th January 1965. The crews of 214 Sqn heard that their squadron was to be disbanded on the BBC news as the official signal arrived at RAF Marham after most had gone home. The sudden death of the Valiant force ended seven years of AAR development that had given the RAF a truly global force projection capability for its bombers, fighters and transports. The race was now on to fill the crucial gap left by the withdrawal of the pioneering Valiant tankers. Gary W8man Feb 00

Copyright Feb/00, Author Flt/Lt Gary Weightman

Gary has very kindly given his permission to share the above article with the 214 Squadron website. If you wish to contact the Author regarding this article, please send a note to the site and it will be forwarded on.

By Sqn/Ldr Vic Pheasant MBE

The call came through from Marham Ops - mid afternoon on Friday 2nd May 1975; as this type of call always did. 214 were to put four crews on standby over the weekend for an urgent overseas task. 'Why 214, can't 55 or 57 (the Victor Mk1/1A sister tanker squadrons at Marham) provide a crew or two?' 'They are meeting other tasks and have no crews available' came the reply. Uhha, we thought, typical 55/57. But we scrabbled around and found fours crews, including my own crew: Tony Banfield - captain, Dick Druit - co-pilot, Keith Richards - nav rad, Pete Martin-Smith - nav plotter, and myself, Vic Pheasant - AEO. I was also to be the nominated Detachment Commander.

Thankfully, the weekend passed uneventfully with no callout, but we did learn of the task. A Royal Navy Phanton F4 from the Ark Royal had gone unserviceable at the US Naval Air Station Roosevelt Roads in Puerto Rico. It was now serviceable, but the Ark Royal, on a good-will tour to the Caribbean and South America, was now in Rio de Janeiro and the F4 did not have the legs to fly that far. The task was to send three Victors to 'Rosie Roads' via Goose Bay; two for the task and one standby. If the primary and secondary aircraft got away OK with the F4 from 'Rosie Roads', then the standby aircraft would return to the UK via 'The Goose'. At Rosie Roads, the primary and secondary aircraft would get airborne, followed by the F4 to fly south as a loose formation. To maximise fuel availability (and this was the good bit), both the primary and secondary aircraft would land back at Seawell, Barbados. At an appropriate stage en route, the secondary aircraft would refuel the F4 and then give its remaining available fuel to the primary aircraft. The primary aircraft would then fly on with the F4 giving it a further top up, and cast off the F4 somewhere south of the equator from where it would fly on to Rio - provided its navigation and comms gear was OK. If it was not, then the primary aircraft was to accompany the F4 to Rio. As it turned out - a rather nice task.

Of course, when Monday morning came and 55/57 heard about the task - especially the Barbados and possibly Rio bit, they immediately had crews available to muscle in. It was decided that 55 would provide the secondary aircraft, 57 the standby aircraft and 214 would provide the primary aircraft (my crew) plus a UK standby aircraft. Also, because of seniorities in personnel, 55 would now provide the Detachment Commander. So, later that morning on the 5th May, the three aircraft got airborne independently for Goose Bay. We were initially in XA 938 with Dave Parsons as our crew chief in the sixth seat. But we did not get off to a good start. Sometime in the climb it became apparent that something was wrong with the pressurisation; that turned out to be the 'flood flow' stuck permanently on. We had no choice but to return to base. We radioed ahead, were able to quickly swap to good old XA936, and flew on to The Goose with no further problems. After a night stop on The Goose, the following day the three aircraft flew on independently to Rosie Roads.

As a US Naval Air Station, there was little suitable ground equipment for the Victors, so we had to make do as best as we could. (Our means of climbing in and out of the cockpit was via a large wheeled, dual bottle fire, extinguisher apparatus.) We met up with the F4 crew in the rather spartan BOQs (Bachelor Officers Quarters), and the following morning completed the planning and briefing for the task. Later that morning, both the primary and secondary aircraft, and the F4, got airborne and established in loose formation en route OK, so the standby aircraft returned to The Goose and then back to the UK. The secondary aircraft gave the F4 its first refuelling, and it was now our turn to plug in to take its remaining available fuel. Tony did his usual jokey 'I'm not sure if I can do this tricky thing', to be told 'just get it done'. And of course he plugged it in with his usual smooth dexterity. When we had all of his available fuel, the secondary aircraft then peeled of en route for Barbados, and we continued on with the F4. By now, we were over the Amazon delta. It was absolutely 'gin clear' so we were able to view this magnificent panorama. The film Papillon, starring Steve McQueen and Dustin Hoffman about the escape from the French penal colony, had been released a year or two earlier, and this came to mind as we flew over its Devil's Island location.

Having refuelled him a couple of times, it was coming to the point where we had to cast off the F4 and, as briefed, this could only be so if its nav and comms gear was OK. We enquired how things were, to be told that everything was fine. At which point we declared that we were receiving his comms broken and distorted and we were having difficulty hearing him. (You don't often get opportunities to fly down to Rio!) However, we were soon told, in direct Anglo Saxon, that all was fine and that we could depart pronto. So reluctantly we turned away and headed back up to Barbados. Taxiing into the dispersal, we were discussing what we should do in the evening we had on the island, when I looked up to see that one of the hydraulic pump ammeters had fallen to zero. I switched it off and on a couple of times, but it was clear that we had had a hydraulic pump failure. When I announced this to the crew, there were cries of 'well done sir' and 'great stuff'. I protested that I thought that this was genuine, which was later confirmed by Dave Parsons, the crew chief, after we had shut down. So this meant that a new pump had to be flown out to us in Barbados, and that we had at least one clear day to explore the island. We told the 55 Sqn detachment commander the situation, and he said that they would also be staying as their nav radar had a severe stomach upset. Seemed a bit flaky to us, but we were in the clear.

The hotel in which we were accommodated was one of the more established places on the island, just a little worn at the edges, but on the beach. I found that I had been allocated a room which opened right out on to the beach. Soon we were in the bar, then out on the town to explore the night life, which we found to be very lively and cosmopolitan. Everybody mixed without question. The following morning, having awoken to the gentle sound of lapping waves, we decided to hire two open topped 'mini mokes' to explore the island, which proved to be an excellent idea. It certainly was very delightful and picturesque, quaintly British in some ways, but with much local charm, especially the small but colourful wooden houses that were built so that they could be moved. The palm fringed beaches were just as shown in any travel brochure, the weather warm and dry. The Barbadians were a very friendly and helpful people, and very hospitable as we were to find out. Sometime in the late afternoon we were passing a beachside hotel, when a cool beer seemed to be the order of the day. While sitting in the bar supping our beers and looking out on to the clear blue ocean, we were accosted by a very large, and formidable looking, Barbadian. 'Where are you guys from?' he enquired, 'you're not the crew from the Victor down at the airport?' When we confirmed that we were indeed so, his face broke out into a very broad grin. He said that he had been in the RAF for five years in the UK serving as a fireman. He was now the hotel manager, and would we like another beer on him.....and another.....and another. It was a great ending to the afternoon.

The next day, which was a Friday, we were told that the new hydraulic pump would arrive that afternoon. The 55 crew decided that they had run out of excuses and would be returning to the UK via The Goose. We said that we would leave the day after provided we got the hydraulic pump installed OK. That afternoon, Dave Parsons and I set of to find the pump at the airport. The two hydraulic pumps in the Victor are situated laterally side by side on the rear floor of the nose wheel bay. They are not easy to get to, and the space is rather cramped with just about enough room to allow two persons to work. Under Dave's instruction, we got the old pump out and installed the new one. It was very hot and exhausting work, especially due to the weight of the pump. We got to the point where we had to make just one last connection of the hydraulic pump to the main piping. To facilitate getting the pump to the right level, shims are used to slightly raise the pump to the right position. But do you think that we could get the large brass nut on the piping to mate with the pump? We tried all ways, re-shimming a couple of times, but to no avail. Eventually it was decided that we would have to send for a new piece of piping, which was of a very complex shape with a number of bends to facilitate its fixing around the nose wheel bay walls. Of course, when the crew were told of the problem necessitating a further delay, they could barely conceal their joy at this turn of events, and we had yet another clear day to explore the island in the mini-mokes, and swim from the beaches. As it turned out, we were into the weekend and we learnt that the replacement part would not arrive until late on the Sunday night. A whole weekend in Barbados! I shall not elaborate further on this turns of events

The new part arrived on the Sunday night, so it was decided that Dave and I would fit it on the Monday morning. If it all went OK, we would get airborne as soon as possible for The Goose. (It was well appreciated that there might be some cynics at Group, and probably at base, that might have other views on our sojourn on this delightful island.) However, if things did not go well and we still had problems getting the hydraulic connections made, then the go/no go cut off would be midday.

Early on the Monday, Dave and I packed and checked out of the hotel to be driven by one of the crew in the mini-moke to collect the part and then out to the aircraft. The rest of the crew took it more leisurely. For the better part of the morning, in the blistering and humid heat, Dave and I struggled in the nose wheel bay until at last we managed to get the new pump and new piece of pipe connected into the system. Dave checked and wire locked all of the connections, and it was time for power on to the aircraft to check that it all functioned. Just before midday, we were able to say that the aircraft was now fully serviceable and, despite the fact that Dave and I were pretty knackered, it was agreed that we would leave as planned. By this time, the mini mokes had been returned, all of the kit was on board, a flight plan had been filed and we were ready to leave. Within an hour we were airborne, and it was a delight to get out of the humid heat and cool down while getting our liquid levels back to some normality.

Just as we were starting to relax in the climb, when passing something like fifteen thousand feet, we heard that familiar, but most unwelcome, hissing sound - of a door seal failure. This was the last thing that Dave and I needed. The door seal was essentially a rubber tube in the space between the cabin door and the fuselage opening. Its inflation, when airborne, made an airtight seal so that cabin pressure could be maintained, usually at around an eight thousand feet cabin altitude. If it failed, as in this case, then cabin pressure would be lost. The usual two alternatives in this situation were either oxygen masks permanently on, or descend to an altitude where an acceptable cabin pressure could be maintained (depending on the size of the leak) or, if not, at an altitude below ten thousand feet. Tony levelled out while we discussed the options. Problem one was that our oxygen level was already down a bit as we were unable to top it up at Barbados, so returning there (as attractive as that might be) was not really an option. A quick calculation of low altitude fuel usage made it clear that we did not have enough fuel to make The Goose at an altitude where we would not need oxygen, and we were uncertain that we had enough oxygen in the depleted tanks for us to make The Goose at high level. The final decision? - divert to Bermuda, which was almost en route.

On landing at Bermuda they clearly did not know what to do with us as we seemed to be taxiing for an interminable period. Eventually we were given a parking area and closed down. While waiting for someone to turn up with transport, Tony, Dave and I discussed what we should do next. Dave was understandably reluctant to undertake a door seal change in Bermuda, and we were unsure whether we would be able to top up the oxygen at Bermuda. What to do? As the discussion went to and fro, Keith Richards wandered up to the group 'I don't know what all this discussion is about. Why don't we just top the fuel up to the gunwales and go up to Goose Bay at low level?!' The three of us just looked at each other - out of the mouths of babes and nav radars! We spent the night at a rather delightful hotel next to a golf course, and the following morning did just as Keith had suggested. The subsequent flight up to The Goose went smoothly, and the aircraft went into the hanger for the door seal change.

There was a sizeable detachment of mainly engineering personnel at Goose Bay, so there were plenty of hands to assist Dave. But our troubles weren't over yet! The Wing Commander Goose Bay Detachment Commander advised us that he had decided to relieve us of our crew chief, who was also Vulcan qualified, and send him on to the USAF SAC base at Offutt where a Vulcan was stuck unserviceable, with the crew chief sick and unable to take the rectification action. We were, understandably, not very pleased at this turn of events, not least Tony, in his capacity of aircraft captain, who had to be restrained a bit from his protestations. The Goose Bay Det. Com. was not the most endearing of characters, but we knew that he was within his rights. It was pointed out to the Det. Com. that we knew that our aircraft was required back at Marham for a major trail that was to take place within a few days, so we could not wait for Dave to return. But more important to the trail was Dave's return to Marham due to the current Squadron shortage of qualified crew chiefs. With the Det. Com.'s refusal to consider this, we requested permission to signal the Squadron about the action that he was about to undertake. He grumbled a bit about this as he had to approve the signal before it went out. Eventually, he sent a message to the effect that he had changed his mind and that our crew chief would now not be required. We did not enquire what it was that had changed his mind!

After a couple of nights on The Goose, our aircraft was pronounced serviceable, and we returned to Marham, arriving back some six days later than planned. Tony had a bit of explaining to do, especially about the Bermuda bit. But there was much envy on the Squadron over our delightful sojourn on Barbados - but it would have been nice to have made it to Rio!

Source : Sqn/Ldr Vic Pheasant MBE

By Flight Lieutenant Gordon Carter

The article is a narrative completed by Flight Lieutenant Gordon Carter and his son in 2001 for 'Through Eyes Of Blue', a book written by Wing Commander A E Ross (pages 176 and 177, published by Airlife in 2002. ISBN 1 84037 345 8)

My experience as a Flight Engineer began after six months of intensive training with the emphasis on engines and airframes, at RAF St Athan in South Wales. I graduated in September 1943 as a Sergeant and this was followed by selection for the aircraft type in which you were destined to fly.

This process was very much a hit and miss affair. All the graduates were assembled in a hangar and the numbers required for each aircraft type were read out. There were only a few vacancies for Coastal Command, but as this was considered to be a safer option, there was no shortage of volunteers to step forward. People were nearly killed in the crush. To this day I am convinced that all coastal command engineers made either good rugby players or bouncers.

The next stage was a heavy conversion unit, in my case to Stirlings at a Conversion Unit in Suffolk. Crewing up was akin to a 'slave market'. The pilot, navigator, bomb aimer, wireless operator and gunners were already a crew from a Wellington OCU. Once again the hangar was used. The engineers stood in a group waiting for something to happen. Suddenly a pilot stood in front of you and asked if you would like to join his crew (no documentation - he must have just liked the look of you), and then the serious business of conversion training began.

There was very little ground instruction - no simulators, no mock-up aircraft, but straight down to sorting out the gauges and levers as the aircraft lumbered into the sky. The screen engineer occasionally slapped your hand and encouraged you if he was so minded. The heavy conversion unit standard of maintenance was very poor compared to the high quality experienced later on operational squadrons.

In less than three months as a crew, we had an undercarriage malfunction, narrowly escaping a collapse on one side. On another occasion complete brake failure on landing resulted in an over-run into a field beyond the aerodrome. Oil pressure failure on an inner engine, with a loss of auxiliary systems, nearly wrote off the air traffic tower due to a swing on take-off caused by a hydraulic throttle control leaking. Worst of all, a fractured elevator hinged arm on an aircraft signed up as serviceable: if I hadn't noticed it on pre-flight check we would have crashed for sure.

The fuel system on the Stirling relied on gravity feed, and tanks being used in sequence, the cocks for the fuel tanks not required were turned off to prevent inter-tank feeding. On my first cross-country trip a well kept fuel log was essential, gauges being far from accurate. However, late in the trip when the fuel state was crucial, condensation dripping from my oxygen mask due to lack of heating, covered the paper of my log, ruining all my readings. Without upsetting the crew I turned on all the cocks, balance as well, and prayed for continued aileron control and a safe return.

The Stirling experienced ceased when we were posted to Sculthorpe in Norfolk, to join 100 Group Bomber Support and Radio Counter Measures (RCM) operations. There we were introduced to American Flying Fortress crew members, who had arrived with some tatty war-weary aircraft.

Conversion to these Flying Fortresses was carried out on an entirely friendly basis, and some firm friendships established. The local Norfolk farmers were less impressed as during our solo conversion training we regularly "buzzed them up" as they stood on their haystacks. The aircraft all had American markings and any complaints would have been made to a puzzled American 8th Airforce Headquarters.

The engineer on a USAAF Fortress was also the gunner, and two pilots were always carried to make up the crew. We already had a mid-upper gunner who became the top gunner, and as the policy of the RAF was to have only one pilot, the engineer occupied the co-pilot's position.

Although we were not qualified to be pilots we were compelled to spend hours in the old Link Trainer to practice touching down and flying on to the final approach, and we had to be able to relieve the pilot whilst in flight. Some brave souls achieved reasonable but unofficial touch downs on real flights. My pilot actually fainted at altitude on a training flight, when I brought the aircraft down to a lower level, he recovered in time to land; a worrying event that turned out all right. In event of a pilot "snuffing it" we had been advised by the Flight Commander to attempt a "belly flop" at one of the special airfields Manston, Woodbridge or Carnaby.

The engineer's task on the Fortress was the usual requirement of fuel handling, engine setting and monitor temperatures and pressures.

The early Fortresses had hydraulic waste gate controls for the Turbo Superchargers. Sometimes at altitude this would congeal, and then suddenly one or more of the engines would start to roar as the turbine over-sped, swinging the aircraft off course. The engineer would then fight to bring back control by devious manipulation of the cockpit levers. The later Fortress Gs (Mark 3) had electronic control: one simply dialled a number to set all four engines.

One drawback with the Fortress was that its optimum cruising speed was about ten knots slower than the Lancasters and Halifaxes, so in order to keep up higher power than desirable had to be used. This considerably reduced our endurance. The engineer's part in this was the management of a manual mixture lever labelled "auto lean, auto rich". The power settings were a combination of manifold pressure (boost) and rpm - i.e. above a certain figure "auto rich" was required to keep the cylinder and oil temperatures within limits. A chart kept by the engineer recorded the consumption for all engine settings. This figure had to be continually divided into the remaining fuel to establish the endurance figure. The compilation of the engineer's log was therefore quite important. The clerical work in compiling the log was far from easy, the paperwork was on a small board, the control column was in the way, only a little red light could be used, otherwise the pilot's night vision was impaired and too much light was hardly wise over enemy territory. Added to this the heating appeared non-existent so cold hands did not help.

On trip in mid 1944 was to Konigsburg, East Prussia, flying over Sweden at a height of 9000 feet and for a duration of nine hours forty minutes. The aircraft's endurance was stretched to the limit. For such trips the aircraft relied on the use of so-called long-range or Tokyo Tanks: these tanks had no gauges, but relied on cocks being operated around the bomb bay area to turn them on. The fuel from the Tokyo Tanks gravity-fed into the main tanks. There was no indication of fuel flow: all one had to go on was the main tanks appeared to be reducing their usage. All the aircraft on this trip experienced very low tank readings on our return over the North Sea. One engineer kept a small hacksaw in his toolkit and panicked to the degree that he sawed through the hydraulic pipes of the Tokyo operating cocks, as we had been led to believe that pressure held them shut.

One other thing for the engineer to do was keep a good look out: with a hundred or more aircraft with no navigation lights on, all eyes were important.

In conclusion, the Fortress was a much more docile aircraft than the Stirling, more like a four engine Anson from a handling point of view. The low powered, but very reliable 9 cylinder Wright Cyclones were no match for the 12 cylinder Merlins or the 14 cylinder Hercules of British aircraft. However the blower enhanced the performance at altitude.

Our task in RCM operations was to neutralise German defences as much as possible and thus help to cut down our night bomber losses.

We carried devices to jam German Air Interception radar and ground/air communications. These had strange code names such as "Jostle" and "Airborne Cigar". We carried German speaking Wireless Operators as extra signallers. We dropped "Window" to blanket German radar. We carried out spoof raids in which aircraft would head towards a target and hopefully draw the attention of the German defending forces from a genuine raid. We would then drop a "Window" and withdraw under it's cover. We might repeat this manoeuvre several times during one night.

Stirlings of our Group would similarly orbit in wide circles over the North Sea in a manoeuvre known as "Mandril Circle" - also designed to neutralise German radar detection. Bombers would then emerge from behind its cover, hopefully on a course German defences had not anticipated.

During one tour on the Fortress, we were damaged by flak, attacked several times by fighters, and had a minor mid-air collision with an out-going (friendly?) aircraft which damaged our tailplane. The Pilot received the DFC after 39 operations, the Rear Gunner the DFM for shooting down an enemy plane.

We were safer in the Fortress at night because we had four Air Gunners with 0.5 in ammunition. However, our RCM role with "spoof" raids on German cities became more hazardous in early 1945, as the enemy started getting wise to our tactics and aircraft were shot down in the circuit of our final base at Oulton.

Despite these early experiences I went on to "engineer" in many four-engined RAF aircraft until the early 1970's. By then I had just about got the hang of it!!

Source : Flight Lieutenant Gordon Carter

By William Walker

At this time there was no flair path as such, what we had was a vessel similar to a watering can with a long spout. Down the spout was a long Wick and the cans filled with paraffin. This Wick was ignited and the vessels alternated down each side of so-called runway. After a time Jerry used to get in the circuits as aircraft waited to land, and of course they became sitting ducks. They also had great fun picking off the aircraft tradesman whose additional duties consisted of lighting these flair paths, and whose job was to extinguish them during an air raid. In order to extinguish these monsters, one had to remove his field service cap and place it over the wick to douse it. Not content with being used as target practice by Jerry, the next morning, you were charged with destroying public property and made to pay for a new field service cap.
I will give you another example of that, the Wellington was notorious for the air intake catching fire, the reason for this was that as the pilot was instructed to start the engines, one of the ground staff tradesmen had to get behind the prop, open the flap door at the front of the engine and operate a KYgas pump to start the engine. This caused petrol to dump into the air intake and ignite. One then had to block it off, and because someone had already used the plyboard to make a toy for his kiddy one had to use his cap. The air Ministry in this country are obsessed with what parts of the uniform are classed as public property, and no excuse on Earth will be accepted if they get damaged. To them it is justified to allow a plane and all its crew plus the equipment to burnout than to destroy your cap putting the fire out. Total damage 2/6 for cap 1 1/2 pence for Badge.

There goes the cap again !

Some bright spark at the air Ministry decided to destroy the German crops by setting fire to the fields. To achieve this, this bright spark decided to place phosphorus in containers of water, and then when over the target pour it down the flare chute. Of course our friend never gave it a thought that it would stick to the sides of the flair chute, consequently it was high summer, and during the hot daylight hours fires were bursting all over the place, and of course, the poor old ground staffs had to sleep on and under the main planes in order to be on hand to douse the fires. Also they were getting burned because it was sticking to their hands.

Source : William Walker

by Laddie Lucas

The following is only an extract from the full story as told to Laddie Lucas by Ralph Fallows formerly of 214 Squadron.
INDIAN NIGHTMARE Ralph Fallows, Observer 215 Sqdn [ Winnipeg] ' Before winter of 41/42 we had been opping with 214Sqdn in 3 Group from Stradishall East Anglia We were used to ops against Gernany but we were first heavy bombers to arrive in India' Went on Mission [Long Haul] only map available finished 20 miles to north. Had sextants but NO tables, NO radio fix available, and NO met info. Ploughed on 3 hrs using winds info given at briefing. Suddenly hit horrendous storm with extreme buffeting and blinding flashes. 'We fell out of the sky at an alarming angle when with wild surge of power the engines roared into life' Suddenly Sammy in rear turret shouted over intercom 'This is bloody stupid Lets get to hell out of it' At that we turned but we were hopelessly lost. Flew out of storm Got rid of bombs in a river bed and Jim the pilot did a forced landing in a field which was hair raising after being in the air 8hrs 25 mins and with no fuel left and a bomb hung up on board. Four kites came back after 2 hrs and 1 crew was never heard of again.
Other than being an interesting story and yet another example of how crews had to pay with their lives from the incompetence of their superiors, it is also rare evidence that crews from the 214 had been robbed to staff the newly reformed 215 which was destined for the Middle East.

Source : Laddie Lucas

A short story by Plt/Off Bill Foskett

As far as India was concerned, we were fully trained and operational on Wellingtons at 15 OTU Harwell. Our posting was to India and at that time we were with Squadron Ldr, Packe as our pilot . In keeping with all overseas postings, we were given 9 different jabs (needles) before we left Harwell and told to report to the London Hospital of Tropical Diseases on the way home, to receive the yellow fever jab. That one was in the backside, the others were in both arms and titties. (barbarians - the very thought makes one cringe) When I finally arrived at the Hampstead underground station with two fully loaded kit bags strung round my neck, I had a very high temperature and felt knackered. Tossing my bags onto the train, I still remember to this day the train guard calling to his mate "Fred; here comes Englands last hope" The next three days were spent recovering ! Then over the following six days I received 7 telegrams (in those days it was by a uniformed lad on a bike) and each one was a different posting. Although it was reassuring to feel needed everywhere at once, The RAF's wisdom came seriously under question when I opened the 7th telegram informing me to revert to the status quo. All those jabs were for nothing !!

Source : Plt/Off Bill Foskett

by Flt/Lt A.C. Wallace

The following is an excerpt from a letter Flt/Lt Wallace wrote to Geoff Shattock in 1996 (pilot) in where he describes a raid on Rostock on the 20th of April 1943 when they were approaching the target (Heinkel Works) and nervous a rear gunner in the Stirling in front of them opened fire on them shooting Geoff through the leg and generally ventilating the plane.

"It was on April 20, and we took off at 2157 on 'N for nuts'. Weather was very good and we could get a landfall as we crossed the coast somewhere around Cromer . Got good Gee fixes for a while then took some Polaris shots when we ran out of Gee range. We were routed a long way north, and crossed the Danish coast pretty well on track at nearly 56 degrees N., at the south end of Rinkabing Fjord. There was some flak, and by this time you had come right down to treetop height.
From there on it got quite exciting, probably one of the most exciting times of my life. Having got a good pinpoint at the coast and having no further nav aids, I came up and sat with you prepared to map read us the rest of the way. We could see the tank defences along the coast (mainly big sort of angle iron contraptions). The trees went by in an alarming way, and we could see doors being flung open and even people standing in the light of their homes. Dickie remarked that we could almost tell people we'd been in Denmark rather than over it. We crossed Denmark fairly fast (Jutland , that is), and crossed Fyn, Langland and Lolland. These are what they are called now but I'm not sure they were that on my chart. At one point there was a lot of light flak around and some of it was coming right at us. I should recall that the term `light flak ` refers to the type (Bofors with tracer) rather than to the severity! Hunt was in the front turret and started yelling "Those sons of bitches!", and opened fire on them . Tommy and Dickie joined in and I'm not sure whether we shut them up or just drew attention to ourselves.
As we reached the open Baltic Sea, you climbed to about 3000 ft., and all was clear as day. I could see all the land marks and we turned on about 200 T. to Rostock, soon in sight. However, as we drew closer, it was obvious that Jerry had put up a massive smoke screen over the whole area; in fact, the "Bomber Command War Diaries", a remarkable publication, states that as a result, the bombing was scattered. This was no town blitz- we were after the Heinkel Works. As we started our target run as best we could, I saw a Stirling weaving a bit ahead of us. Then there was a rather unpleasant bang, and you said something like "Oh, Jesus Christ!" Then you said `take that!', referring to the control stick. Then you called "Hunt, come up here quick!" As the only even remotely qualified second pilot, he let the bombs go and came up and took over the controls.
I gave Hunt a course to steer. I told him to steer 015M for 10 minutes and then turn on 315. Then I went and found a flashlight and got your flying boot off. It was full of blood and there was a hole through your leg below the knee. I got the medical kit and put a shell dressing on it, but didn't try any fancy hemostasis because I was sure it would quit bleeding O.K. All this messing around in the dark took a while, about 20-25 minutes, and when I got back up with Hunt he was still steering 015 - either he hadn't heard me or didn't have a watch. I looked out and could see clearly the Copenhagen area right ahead. So we turned on about 270 and headed for home. In answer to your recent comment, I can't imagine that anyone even remotely thought of bailing out.
The trip back was pretty uneventful. You described the amount of pain you had as about equal to a severe toothache. You soon took over the controls and declined any more help. I offered to shoot some codeine into you with some very neat little gadgets in the medical kit, but you were obviously scared it might destroy your razor-sharp awareness. Smitty radioed ahead something like: Pilot wounded; all O.K. When we got to base the Waaf on Flying Control asked: "What is the nature of your wounds?" and you answered that you had a bullet in the calf of your leg. You set it down perfectly and were whisked away to the hospital. Next day the ground crew found that ONE bullet had cut off the cockpit lighting, cut out the hydraulic exactor control of one port engine, gone through your leg, mangled some of my charts, and still had enough zip to make a dent in the armour plating beside my navigation table.
It was a bad night for 3 Group (20/04/1943) - we lost 8 of 80 planes and didn't do a good job. Also a lot of people who got back were shot up with some killed including one gunner from 214."

Source : Flt/Lt A.C. Wallace

by Flt/Lt Marshall Angus Johnson

This is probably the most comprehensive listing of cigarettes available during WW2 that was ever assembled.

All of these were either sent from Canada to Germany or were traded for amongst POWs. All were available between 1941 and 1945.

Buckingham, Brittish Console, Players, Sweet Caporal, Gold Flake, Winchester, Scotch Blends, Capatan, State Express 444, State Express 333, State Express 555, State Express 777, Senoir Service, Fifteens, Churchman, Royal Oak, Gallahers, Black Cat, Tenners, Rum & Maple, Raleigh, Wild Woodbines, Chesterfield, Camela, Juno, Roy, Diana Martins, Old Gold, Lucky Strike, Pall Mall, Twenty Grand, R6, Canadian Legion, Phillip Morris, Torch, Riverhead Gold, Doux Caporals, Eligant, Wrights, Helas, Plane, Helas, Junak, Haudegen, Domino, Drava, Tenners, Marchorkows, Haudegen, Papastratos

Source : Flt/Lt Marshall Angus Johnson

by Shiela Bryne

Aircrews were put together in a rather haphazard manner. There being no ridged formal assignment, crews were formed largely by picking and choosing amongst themselves. During the training sessions friendships would develop as the pilots, Aimers, Gunners and Navigators trained along side one another, and subsequently spent time in the barracks and local pub together. Over the course of these training and recreational activities, groups would begin to form up based on these friendships and the casual assessments they made of each others skills, personality and sense of humour. In the end, a crew with a common bond and trust in each others skills would emerge. Notwithstanding, adjustments might have to be made, "and were made" if the crew could not function as a team.

Source : Shiela Bryne

By Arthur Skone and Plt/Off Bill Foskett


A typical description of a bomber crew at the time was provided by an Air Ministry publication entitled Bomber Command, which was issued by the Ministry of Aviation in 1941.

"The men of Bomber Command are appointed to fulfill a special mission. Their life is not that of other men - not that even of those in the other branches of the service. Its very physical conditions are different. For them nowadays much of the night is day, much of the day a time for sleep and repose. Discipline is constant yet flexible...Triumph and disaster are met and vanquished together.

The captain and its second pilot do the actual flying, the observer navigates and drops the bombs; the wireless operator helps the navigator and the air gunners do the fighting. The same spirit and practice of co-ordination is required of a bomber crew as of a crew of a racing eight or the members of a football eleven...

The bomber pilot differs in training and environment from his colleague flying a Spitfire or Hurricane. A pilot of the Royal Air Force is subjected at an early stage to a process of selection by which it is determined whether he is better fitted to fly a fighter or a bomber. Both will have to fly aircraft; both will wear pilots wings; but here their ways diverge. The fighter pilot is in action for an hour and a half to two hours at the most, often far less. He is usually led into the fight by his squadron leader.

Very different, but equally important, qualities are required of a bomber pilot. He must be capable of considerable physical and mental endurance, for it may be to fly for the most part of the time over hostile territory or across the unfriendly sea. During much of the flight he may find his aircraft the object of attack by enemy assailants who can break off and renew the assault at any moment. Surprise, that weapon which more then any other wins a fight, is theirs to wield at will. The bomber pilot must fly doggedly on, defending himself with the aid of darkness and cloud outside and with the skill of his crew and their machine guns inside. The bomber pilot must not forget that he is one of a team and that the team is not flying separated from him in another Hurricane or Spitfire, but the same aircraft, crouched over the navigator's table or hunched up in the gun turrets. He must be imaginative, yet not dismayed by his own imagination, brave yet cautious, cool yet daring."

Source : Ministry excerpt by kind permission of Arthur Skone From his publication "A Peaceful View" The Biography of Alan Raymond Collier Skone

The photo above is of Plt/Off Bill Foskett taken at the flying control tower at Udine in North Italy. This photo is one of my personal favourites because it so perfectly depicts that mental image many have of the young, dashing, devil may care, airmen of these times. An image the RAF did much to foster and perpetuate I might add. Boys being boys, those that the RAF Recruiting office couldn't convince to join for the sake of their country, they snagged through subtle propaganda by convincing them that the girls swoon over a man in an RAF Uniform. (I can assure you however that the latter never entered Bills mind.)
A case in point...................

Source : Plt/Off Bill Foskett

© Copyright No. 214 Squadron website and the families and contributors to the site.

All information, pictures and articles on this site, unless otherwise indicated, are the property and copyright of the families of the No. 214 Squadron collectively or of an individual member.

The use of any materials from this site for any commercial purpose whatsoever is strictly forbidden.

Privacy / Data Notice

This site was developed using raphics/w3schools88x31