When I started flying on Vulcan’s I heard some of the guys on the squadron talking of a Bomber Command dinner that had been attended by Sir Harry Broadhurst, and at which there had been an undercurrent of ‘hissing’ because of a Vulcan crash at London Airport in 1956.

Letter from Broadhurst Report of the Court of Inquiry Statement of expert witness Diagrams showing trace of flightpath Comment from Mac McConnell-Wood Comments from Archie Swan

The photo above is from the April 2007 edition of Aeroplane; it shows 897 on fire at Heathrow after the crash.

Broadhurst had been part of a flag waving exercise to show off the Vulcan around the world, and the aircraft was planned to land at London on its return.  The mythology was that he had persuaded the pilot to continue an approach to London in bad weather, rather than divert and spoil a media exercise.  The story has been in the back of my mind since hearing it, but while mythology isn’t necessarily ‘truth’, the actual events don’t make much better reading.

Broadhurst had a great career and was the youngest Air Vice Marshal when he took command of the Western Desert Air Force in 1943, he was a brave pilot who had earned the D.S.O. (and bar) and the D.F.C. (and bar), most of his career was on fighters.

In June of 2005 I made one of my regular visits to the PRO, this time with a view to reading the files on the crash of XA897, as well as that of XM645 that crashed on Malta in 1975.

In 1956 the Vulcan was a new aircraft for the RAF, indeed XA897 was due to be the first aircraft for delivery to 230 OCU for the training of other Vulcan aircrews, before it was ‘diverted’ to a Fly the Flag exercise where the aircraft would fly round the world, with Broadhurst, Air Officer Commanding Bomber Command, as co-pilot.

The trip had been a great success, and we should remember that at that time the Vulcan would have been an incredibly impressive aircraft to spectators who were not used to seeing large multi-jet bombers. From an incident point of view it had been an uneventful trip, culminating in the last leg of the journey from RAF Khormaksar in Aden, to London Airport. This last leg would also be uneventful, until the latter stages of the approach.

The Captain of the aircraft was Squadron Leader Howard, and it was to prove unfortunate that he had not previously done a full Ground Controlled Approach (GCA) before in a Vulcan, neither had he practised approaches using the Instrument Landing System (ILS) in that aircraft type.  In other pages describing the aircraft I have made reference to the limited visibility available to the pilots through the rather restricted vision panels on the Vulcan canopy.  It is an essential part of the co-pilot’s role to monitor height and speed for the Captain, and to keep a lookout for the runway lights.  This is particularly important in poor visibility and as the aircraft approaches ‘break off’ height. Flying the approach the pilot in control, in this case Howard, would follow his instruments and the ground controller, only looking up and going visual when he reaches his declared break off height. At that point he will only have a short time to acquire the lights, confirm for himself that he feels confident he can make a landing from that point, and declare to the crew that he has the lights and his decision is ‘to land.’  If at that point, his declared break off height, he is either unable to see the lights, or feels that the aircraft attitude is such that he cannot continue for a safe landing, then his decision at that point should be to overshoot and possibly go round again for another approach. In this case when the London weather was passed to him it was obviously marginal and he said that he would only make one attempt before diverting to Waddington.  So there may have been some personal pressure on him at that point to get in off the first approach.

The break off height should be treated as just that; it is the final point at which a Captain makes the important decision as to whether or not he feels capable of landing the aircraft safely. The whole crew is aware of the importance of break off as a decision point and at least one of the Navigators will normally also monitor height as this point is reached.  The co-pilot is normally briefed by the Captain to give specific height calls prior to break off, such as “100 feet to break off” so that the Captain is ready to look up.  Looking out for lights is also an important function.  For those who have not been in the cockpit of a jet on the final stages of approach in bad weather, they won’t realise how difficult it can be to see the landing lights and confirm that they are the runway and not some other motorway or service area lights.  The London weather passed to the crew indicated only light wind (though heavy rain), so the aircraft would have approached with only a small amount of drift on, and the lights should have been close to the nose, making them easier to acquire. Indeed, when the co-pilot did decide that he had seen the lights, he called to say that they were “fine on the starboard bow” (a bit nautical!). The visibility was very poor, the co-pilot says in the report that he could see the ground below the aircraft but not ahead; a classic situation for close monitoring of the aircraft height in relation to the break off height.

I know there have been occasions where the weather has been marginal and I have stood on the ladder between the pilot seats on final approach as an additional pair of eyes, looking for the lights. Then a quick scuttle back to my seat before the landing.

Part of a letter in the file from Air Marshal Sir Harry Broadhurst to Air Chief Marshal Sir Dermot Boyle (C-in-C Bomber Command). When you see the diagrams at the bottom of the page, displaying the vertical path of the aircraft on approach, it is difficult to see how Broadhurst can say that it all seemed normal.

Broadhurst starts by saying how good the trip had been.

“ …and now I find I’ve let the party down.

I just don’t know what went wrong with the let down. It all seemed normal to me … I was monitoring everything and I consider I’m a pretty harsh critic when it comes to aviation.

When the talk down finished there were a few moments when I nearly took off my co-pilots hat and put on my brass one.  It might have saved the day.  However, I had installed myself as a co-pilot and I was determined throughout the tour to behave like one. I think that was right.”

The letter continues with Broadhurst apologising for a ‘dreary’ letter, and says he knows that Boyle will do the right thing and he in turn will react accordingly; he signs off as “Broady”.

The two reports that follow should be read bearing in mind the two diagrams at the bottom of the page, these show the aicraft approach path in relation to the glideslope.

Report of the Court of Inquiry

SECRET

ACCIDENT TO VULCAN B.1 XA897 AT LONDON AIRPORT

ON 1^ST OCTOBER 1956

NARRATIVE, FINDINGS, CONCLUSIONS.

RECOMMENDATIONS AND OBSERVATIONS

OF THE COURT OF INQUIRY

NARRATIVE

• On 1^st October, 1956, Vulcan B.1 XA 897, captained by Squadron Leader D.R. Howard, with Air Marshal Sir Harry Broadhurst as co-pilot and a crew of four, left ADEN at 0250 hours G.M.T. with the intention of landing at LONDON AIRPORT at 1008 G.M.T.
• The flight was normal. The forecast landing weather at LONDON AIRPORT was passed to the aircraft when it reached EPSOM at about 0958 hours G.M.T. as; 2/8ths cloud at 300 feet; 7/8ths cloud at 700 feet; main cloud base 5,000 feet; visibility 1,100 yards; heavy rain and little wind.  This proved to be an accurate description of the subsequent weather experienced.
• The Vulcan had ample fuel to divert and the weather at WADDINGTON, which was considered as a diversion, was forecast as; 1/8^th cloud at 600 feet; 3/8ths at 8,000 feet; main base 13,000 feet; visibility 3 nautical miles.
• The Captain after R/T discussion with Bomber Command Operations, and with the agreement of the co-pilot – who emphasised to him that if he was dissatisfied with the weather conditions prevailing he should divert – decided to make one attempt to land at LONDON AIRPORT. The break off height which the captain set himself for this approach was 300 feet indicated above sea level (touch down height at LONDON AIRPORT 80 feet). [This makes the 300 feet ASL only 220 feet above ground level AGL.  J Dillon]
• At 1004 hours G.M.T. the Vulcan was 1,500 feet five miles from touch down on Runway 10 Left, and began its descent under G.C.A. (LONDON TALK DOWN) with a Q.N.H. of 1017 millibars set on both the Captain’s and co-pilot’s altimeters, which was the correct setting.  [Q.N.H. is the pressure at sea level, at that time, so with this set, when the aircraft is on the ground at London the altimeter will read 80 feet.  J Dillon]
• The Captain remembered in the initial stages of the approach going above the glide path by 80 feet, then over correcting and going 100 feet too low.  This latter was in fact only shown as 50 feet low by G.C.A.  Then having established the power required, he believed he remained sensibly on the glide path throughout the descent; the latter part of which was made through well broken cloud, but in extremely heavy rain which, despite the use of windscreen wipers, made forward vision practically nil.
• In fact, as shown by the G.C.A. Talkdown, the Vulcan again went above the glide path, and then at 1¾ miles from touch down had descended to 100 feet below it. The aircraft had, however, regained the glide path at 1½ miles and was advised 80 feet above it a little before ¾ mile from touch down.  The Captain continued his descent under G.C.A. Talkdown and at 1,030 yards from touch down the Vulcan made initial contact with the ground, removing both main undercarriage units.  Neither of the pilots considered that this impact was heavy, though the airspeed during the approach was proved to be about 150 knots.
• At no time after the ¾ mile on the approach did the G.C.A. Talkdown Controller warn the Captain that he was below the glide path.
• Just after the first impact occurred, the phrase “You’re quite clear to land on this approach” was used by G.C.A. Talkdown Controller, which the Captain may have interpreted to be final clearance to look ahead and make his landing, because at that moment he asked the co-pilot for lights.
• The co-pilot, who could see the ground below the aircraft, but nothing ahead through the windscreen, at first saw no lights. He then discerned a glow, and finally advised: “Lights fine on starboard bow”.  The Captain looked out and saw the lights, but could not interpret them; although LONDON AIRPORT’S high intensity lighting was on, and at 100% power. The co-pilot at that point realising that the aircraft was much too low called “Up, up”, or something similar, whereupon the Captain began to pull up and increased power very quickly just as the aircraft touched the ground at about 140 knot. He believed that his altimeter read about 300 feet A.S.L. when the aircraft touched, but just before contact the co-pilot gave the reading of his altimeter as 250 feet A.S.L.
• After impact the aircraft left the ground very quickly and the Captain reverted to instruments.  The co-pilot said “You can still make it”, or words to that effect, but the Captain replied “Negative, I am going round again”.  Full power was on and acceleration had been achieved, apparently beyond 160 knots. The Vulcan began to climb away in a steep attitude.
• During this climb the Captain found he could not control the aircraft, which was trying to roll to the right, though he had the control column hard over to the left. He thought he might manage to check the roll, and believed he had some elevator control.  It soon became apparent to him that he could not regain control of the aircraft because, though the controls felt normal force wise they did not seem to be taking effect.  He felt that the aircraft was both yawing and side-slipping; though he determined this by feel rather than by reference to his instruments.
• The pilot was positive that the aircraft was not stalled, and thought that the initial impact had in fact damaged the controls, though he saw no red warning lights indicating Power Flying Control failure, nor did he notice the position of his Control Surface Indicators at the time.
• He gave the order to abandon the aircraft and himself ejected. The co-pilot took the controls instinctively, but as soon as he pushed the control column forward and applied port aileron he realised that the controls were not responding.  To him the bank appeared to increase to about 75⁰, and the speed to fall off to 150 or 160 knots. He repeated the order to abandon the aircraft and then ejected.
• The ejections, which were within a second or so of each other, probably occurred at about 200 to 300 feet and, according to eyewitnesses, when the aircraft was either in a slight turn to port or in straight and level flight. The aircraft had passed the Runway Caravan and was about 1,000 yards down the runway when the ejections occurred.
• The Runway Controller Under Training in the Runway Caravan was listening to the G.C.A. Talkdown and watching the Vulcan approach at the same time.  He is categoric that the G.C.A. Talkdown Controller gave “400 yards from the runway: talkdown complete” when the Vulcan had passed the Runway Caravan by about 100 yards.
• The Runway Controller in the Caravan was also watching the aircraft and listening to the G.C.A. Talkdown. He confirmed that the last message “400 yards from the runway: talkdown complete” was given late.  In fact, he stated that the last word coincided with the pilots’ ejection which was about 1,000 yards down the runway past the Runway Caravan.
• After the ejections the Vulcan banked steeply to starboard, the nose dropped, and it crashed on the right-hand side of the runway at the far end and burned out.  The remaining crew of the aircraft – Squadron Leader Stroud, Squadron Leader Eames, Squadron Leader Gamble, and Mr. Bassett, were killed instantaneously.

Statement of Captain Hunt on the proceedings

SECRET

STATEMENT BY CAPTAIN V.A.M. HUNT F.R.Ae.S. F.I.N.

DIRECTOR OF CONTROL AND NAVIGATION.  M.T.C.A.

OBSERVER AT THE R.A.F. COURT OF INQUIRY.

• I attended all the meetings of the Court of Inquiry at which evidence was taken and also had informal discussions with the Court during its proceedings.  I should like to record my grateful thanks for the courteous way I was granted freedom by the Court to question witnesses.
• I have now given careful thought to the findings, conclusions and recommendations of the Court in relation to my knowledge of the evidence given and it is with the greatest regret that I have to record that I am at variance with them on a number of points.  I believe that undue prominence has been given to the failings of the G.C.A. system and that too little consideration has been given in the Conclusions to the findings referring to the errors by the pilots, particularly when it is recalled that this accident took place at a civil aerodrome where a large number of transport aircraft and passengers were likely to be collected and where, but for the grace of God, this pilot less jet bomber would have crashed on the Central Area or the North Apron, with considerable loss of life.
• In view of the gravity of the above statement, I have thought it fit to outline below my reasons for dissent from the report as it now stands.
• During the early findings the Court refers to the undulations of the approach in elevation, but a study of the curve of the rates of descent against time, prepared from the reconstructed plot, shows that there must have been definite large scale undulations which, up to a mile and a half from touchdown, would have ranged from less than 200 feet per minute to 2000 feet per minute.  This is not inconsistent with the final descent rate of over 2500 feet per minute for the last few seconds of the aircraft’s flight. This is significant, as the G.C.A. team and pilot between them would only have had about five seconds after the aircraft had passed downward through the glidepath for the last time to pass a message and for a correction to be made before the plane had crashed.  To my mind there is considerable doubt whether, in fact, a warning at this rate of descent and at this time would have avoided the disaster, since from experiments in other aircraft it is known that something in the order of five seconds elapses after an instruction has been received in the aircraft before a change of 10 feet from the original path can be effected.
• In paragraph (e) of the findings it is stated that the last message “400 yards from the runway, talkdown complete” was about half a mile late. From a study of the evidence available and a calculation of the speeds involved I conclude that the Court’s figure must be considered an absolute maximum and that the probable delay was less and of the order of seven seconds.
• In elevation there is unfortunately insufficient evidence to conclude positively whether the Talkdown Controller was unaware that the aircraft had gone below the glidepath or that he had failed to pass the information, and I think therefore that it is not possible to apportion responsibility for this error to either member of the team.  It should be remembered that during this talkdown the approach and descent speeds were well above those normally met with at civil aerodromes.
• I am surprised to see that no reference is made by the Court to the part played by the second pilot during the crucial stage of the approach.  From the evidence available I believe that the second pilot, although being asked to keep a good lookout for lights and also to pass heights during the descent, did not spend enough time in looking out for the final cloud break, the lights, and checking the visibility. Nor did he remind the pilot that he was approaching his breakoff height.
• Reference is made to the fact that the pilot had not performed a full G.C.A. in a Vulcan.  In addition I believe that it should be recorded that he had not practised I.L.S. in a Vulcan either. I think it would be unwise to consider that carrying out a few ‘director controlled’ approaches from Ringway into Woodford is any substitute for practising a ‘precision’ talkdown.
• In the light of the above I believe that the first three conclusions should be redrafted and expanded on the following lines:
• In view of the lack of recent instrument approach practice with any radio or radar aid, the Captain was not justified in making an attempt to land at London Airport in the prevailing circumstances.
• The Captain made an error of judgement in setting himself a breakoff heigh of three hundred feet and committed a serious breach of flying technique in going below that height.
• The co-pilot failed to give early enough warning to the Captain of the aircraft of the proximity of the ground and lack of forward visibility.
• The omission of the G.C.A. to warn the pilot that he was going below the glidepath failed to prevent the accident and this must therefore be considered as a contributory cause of this disaster.
• These conclusions confirm my view that there were faults both in the air and on the ground and it is probable that no one fault by itself would have led to the accident.

V.A.M. HUNT

16/10/56

Also on file at Kew is a statement to the House from the Secretary of State for Air, of which this is an extract.

“We are agreed that there was an error of judgement on the part of the pilot in selecting a break off height of 300 feet and going below it and also that the G.C.A. Controller did not give adequate guidance on elevation during the descent and, in particular, that he was at fault in the concluding stages in not warning the pilot that he was below the glide path and therefore dangerously close to the gound.”

The statement concludes by saying that it was not possible to define the degree of responsibility between Vulcan crew and ground Talk Down staff.

In file AIR 20/12396 there are comments that the controller was felt to have given inadequate height information to the pilot in the last few seconds, but the point was made that the safety of the aircraft is the responsibility of the pilot, it is his decision whether or not it is safe to descend below the break-off height, even if the talk down is felt to be poor.  The controller was working from a dual display that had a height trace on one ‘tube’ and an azimuth trace on the other. Because of deviations in azimuth the controller was felt to have spent too much time on that, to the detriment of the height information.

Diagrams showing aircraft flightpath.

This shows the trace from 6nm out to touchdown.  The glideslope is from point ‘X’ on the right, to the Touchdown point on the left. Height variation is shown against time, i.e. at 4nm with a time of 10.04 and 36 seconds, the aircraft is on the glideslope.

A similar trace but from 2nm to Touchdawn point. The red line shows the flight path before impact, with the green line as the estimated path after impact. Glideslope id from ‘Y’ to ‘X’.

I had an email from Mac McConnell-Wood who was crew-chief  for XM657.

“I heard from an inhouse AVRO engineer that the main cause of the London accident was that the inertia switch tripped on the first impact. This was a safety feature designed to reduce the possibility of electrical fires and cut all electrical power when abnormal impact G was encountered. Unfortunately it also cut the power to the PFCU's making the A/C uncontrollable from that point on. This was one of those cunning devices that Avro was keen on like the explosive fire supression system in the fuel tanks!”

The following was emailed to me by Archie Swan, ex Missile Systems Fitter [or MisFit as he called it].

Switching to the end of your article, you've that comment from McConnell-Wood. Mac was on the Mk2's, 897 was a Mk1 - there was a heck of a difference in the electrics, as you're probably well aware. But his "inhouse AVRO engineer", provided that Mac reported his wording correctly, was somewhat adrift.

Yes, there was a possibility that an inertia switch might have been a part of the cause, but it may not have been the main cause. On the original Mk1's that we had, there was a "safety connection" made
through the Nosewheel micro-switch. If/when that connection was severed, it resulted in the loss of the PFCs.  We'd actually had some arguments with some of the engineers & designers etc. at Woodford about that when we were doing our original Vulcan courses there. But we were assured that it was a needed safety factor.

Now - back to the touchdown area item. In that field, after the crash - and during the site examination - they found that Nosewheel, with ripped out cable. That, in itself, could have been the cause of the loss of the PFCs.

We ended up, very quickly, modifying all the Mk1's to get rid of that feature. It was during that wash-up that Gus Walker took over as 1 Gp's AOC - interestingly in a recent book about Sir Gus, it makes a mention of his involvement. He had us all, Groundcrew and Ground School - plus Jock Hannah who'd taken over Fred Bassett's place as the AVRO Rep, going over all the PFC circuitry etc. Fortunately, Gus was well versed in Engineering. That was the first time I'd been hunkered down on the floor with an A.V.M.

Anyone else got any comments on PFC loss on this incident?