Posts about: "Engine surge" [Posts: 41 Pages: 3]

speedbirdconcorde
Yes we did, just a couple if I remember correctly and relatively minor failures at that. (Regular ultrasonic NDT inspections had been instigated to pre-empt these failures from actually occuring). New elevon purchases were rightly seen to be the answer to the problem; the poorly designed trailing edge extension modifications of the late 1970's were as was said before, the source of these failures, due to moisture ingress in the honeycombe structure).

Mr Vortex
As has been posted previously, there was a small NACA duct on the right hand side of the fin, that provided the air source for fuel tank pressurisation. This pressure was controlled to 1.5 PSIG.
This switch operated two valves that would drain out any residual fuel for maintenance (for example, replacing a vent valve); it was not used very often however.

Islander539 and ChristiaanJ
The actions of Airbus at Filton are nothing short of disgusting. By 'removing the insulation' you will need to strip the cabin completely bare (seats, galleys, ceiling panels and all of the side-wall panels). They say that 'Filton was only ever going to be an interim home for Concorde', what total crap !!
The idea is to 'cocoon' the aircraft 'until a permanent home is found'. I hope all readers here realise that this will involve BREAKING UP THE AIRFRAME to make it road transportable. The reasons that scarebus are giving for all this are vague and misleading, but here's my take. There are pressures around from various people and bodies 'to return a British Concorde to flying condition.' Now a lot (NOT ALL) of these people although very well intentioned are not that well informed and their wishes are not reasonably possible. But the pressures exist nonetheless, and scarebus will do anything to prevent this possibility, nomatter how unlikely, from being progressed. So we have G-BOAF, the youngest Concorde in the world, with the lowest airframe hours, in pretty good structural condition (she's suffered from being outside for 7 years, but nothing terminal) and actually in the hands of the dreaded scarebus (who would rather forget that Concorde ever existed, and was almost certainly the reason why they even noe exist). Doesn't take much working out now, does it?

Dude

EXWOK
I think you will find it was 18\xb0 TLA mate (age catching up with us??? . OK I know it has been 7 years).
The 18\xb0 TLA limitation was set to prevent TOD 'pop' surges, due to the resulting large intake ramp angle causing excessive compressor face distortion (the one side effect of the intake 'thin lip' modification).

Best Regards
Dude

Pedalz -

I'm not the best person to reply to your ramp query - he'll be along later! - and it's been largely answered already, but the bare bones are this;

Ramp 1&2 Green system, back up of yellow, 3&4 Blue, backup yellow.

Any continuous surge at supersonic speed would affect the adjacent engine, hence the requirement to close all 4 throttles.

Cheers,

EXWOK

A certain CFI (I think) at BA flying club, High Wycombe, who was also F/O on concorde, showed me some photographs of an engine that had eaten a piece of intake ramp.
I think he said that the adjacent engine had surged and a piece of ramp went out the front and down the other engine. This resulted in a double engine failure mid atlantic. They landed in Shannon with very little fuel left.

A double engine change ensued.

Question, how fast was the ramp going if the A/C was at Mach 2?

Maybe M2dude remembers the occasion?

First time that happened was on prototype 001 in the very early days, when an engine "spit out" the entire ramp (there's a photo in Trubshaw's book).
The ramps and actuators were 'beefed up' considerably afterwards... I didn't know an in-service aircraft had suffered a similar mishap.

Good question.... not being an "engine man" I've always been amazed how a nice steady Mach 2 flow, slowed down to Mach 0.5 at the engine inlet, is capable of totally choking off and even reversing itself in less than a second.... no wonder it's usually accompanied by a big bang!

CJ

PS I have no record of any of the British development aircraft ever having lost a ramp, notwithstanding the number of deliberate engine surges they went hrough. But then maybe I wasn't told....

Hi Guys, quite a few little points here, so here's my angle(s):
Pedalz
Ooo yes. The biggest problems we ever had associated with the ramps themselves were wear in the seals at the sides of the forward ramp. Even a few thou' over the maximum allowable side gap was enough to make the intake unstable and susceptible to surging. (It is quite interesting that the rear ramp side gaps were not in the least bit critical, and if Concorde intake development had continued, the rear ramps were going to be deleted altogether). Other failure factors were control unit malfuntions, rapid sensor drift; all of these causing either ramp/spill door drift or runaway. Primary nozzle misbehaviour could also result in intake surges. Having said all that, the monitoring of the intake system was truly superb, and surface runaways, themselves quite rare, would usually be picked up by the control system monitors causing either a lane switch or if that did not work, a total 'red light' failure with the surfaces frozen. No surge was treated as 'just one of those things', and much midnight oil was burned and hair pulled out (so that's what happened to mine ) to try and find the cause of the surge.
My friend EXWOK perfectly answered the intake hydraulics allocations.
EXWOK was right on the ball here as usual, in fact above Mach 1.6 an interactive surge was more or less guaranteed. The cause of interactive surge had nothing to do with the wing leading edge position, but to the radially generated distortion field coming out of the FRONT of the surging intake, severely distorting the adjascent intakes airflow. It mattered not if the originating surge was an inboard or an outboard intake, the other guy would always go also, above Mach 1.6.
You might want to take a look at 'When Intakes Go Wrong Part 1:
Concorde engine intake "Thrust"
and Parts 2 & 3:
Concorde engine intake "Thrust"
Not to mention Part 3:


dixi188
I can never recall this particular event happening with BA , certainly not as a result of a ramp failure. Although in the near 28 years of operation we had quite a few SNN diversions, none that I can ever recall were as the result of a ramp structural failure. The two major SNN diversions that I can recall were G-BOAF in the early 80s when an LP1 blade failed and resulted in a totally wrecked engine (although a completely contained failure) and G-BOAA in 1991, with another wrecked engine due to running in rotating stall. (Both of these events were covered previously in our thread). ChristiaanJ has mentioned quite rightly the event with A/C 001 spitting a ramp out, and Air France had a ramp failure going into JFK. (Covered previously in our thread, due to certain 'human foul ups'). I am not sure, but I think that this one in JFK DID require a double engine change in JFK. (Usually from SNN a BA aircraft would be 3 engine ferried back to LHR).

ChristiaanJ
Nope, you are quite right, no more French or British development aircraft ever suffered a ramp linkage failure again. The 001 ramp failure was a salutary lesson to the design team, and the intake assembly became tougher than old boots after that, nomatter WHAT you threw at it.


Due to the lateness of the hour (and me being up at 4 ), that will have to do for now guys.

Best regards to all
Dude

There's a description and a picture of such an incident in the RR Heritage Book about Olympus. Happened in "Mach Ally" over the Irish Sea, even though the front face of the Compressor was wreaked it could still run up to 85% without surging. Can't remember which 85% though and the book is 4000 miles away from me at the moment.

Regards
H wie

quote: d putting further Concorde's achievements in terms of stability; the world's only previous large delta winged Mach 2 aircraft, the B58 Hustler, had the slightly awkward feature in the case of an outer engine failure at Mach 2, in that the yaw forces were sufficient to tear the fin off. This happened on more than one occasion during service life of the Hustler, but engine failure (or far more likely a deliberate precautionary shut-down) although hardly a non-event in the case of Concorde, it was routinely dealt with without drama or danger.unquote

To rub it in, a typical double engine surge - they were nearly always double surges as the first surge expelled the ramp shock waves and turned the flow into a pitot with a large standing shock ahead of the intake that screwed up the flow into its neighbour - would produce about 1 degree sideslip and 2 deg bank. There would be a +/- 0.2g variation in normal acceleration and that was it! Through Christiaan's kind offices I am posting the records of such an event.

Hustler pilots eat your heart out!

CliveL

Here are the graphs that CliveL was referring to.

The Mach trim control law




The aircraft response to a double engine surge
Split into two halves (longitudinal and lateral response)






Note the almost immediate rudder response, long before the engine N2 rpm starts to wind down. I'll have something to say about that in a separate post....

CJ

A double engine failure, or even a double engine surge, could lead to a very nasty yaw, faster than the pilot, not necessarily instantly aware of exactly what was happening, could counter.
The designers were, right from the start, aware of this problem.

Hence, the prototypes were equipped with specific "contre automatique" (auto-rudder) computers, that would "kick in" a given rudder deflection as soon as they detected an engine failure (and twice as much in the case of a double failure).
Unfortunately... the manner of detecting an engine failure was based on pressure sensors in the engine, which proved to be notoriously unreliable.
Since the whole system was "fail-passive", in the case of a pressure sensor failure nothing happened, other than that I got the "suspect" computer dumped in my lap every time, since it was easier to swap a computer than test and swap pressure sensors....
In the end, it was always "no fault found", and the engineers had to go and test the sensors to find the failed one.

Already on the pre-production aircraft, this Rube Goldberg system was replaced by a single circuit board 'buried' in the autostab computer.
It used a lateral accelerometer to detect the abrupt yaw of a sudden engine failure or surge, and applied appropriate rudder. Look at the sudden rudder deflection 'peak' on the lateral response graph in the previous post.

Since there was no separate 'auto-rudder engage' control switch (the function was permanently active), and it was only mentioned very much in passing during training, some pilots were not even aware it existed.......

CJ

Christian asked if there was an aerodynamicist in the house - I guess that would be me!

The original question was whether there was any vortex activity in subsonic cruise, but the discussion went on to ask about designing for subsonic drag I think.

The answer to the first bit is that the vortex flow started in a gentle manner from about 6 or 7 deg AoA and got steadily stronger. Depending on the chosen cruise speed and the aircraft weight, the subsonic cruise AoA would have been in the region of 4.5 to 5 degrees, i.e. below any significant vortex development. 6/7 deg would correspond to something in the range 250 to 280 kts probably (I haven't done the sums)

What we were trying to do for subsonic cruise was to have what is known in the trade as 'leading edge suction' acting on a nice bit of forward facing area so that it tried to drag the aircraft forwards as it were. As you can see from the diagram the prototype aircraft had a much more cambered LE so that both suction and forward facing area were very reasonable. This prototype shape was nicely rounded so that LE separation and top surface vortex generation started at a higher AoA than on the production aircraft. Unfortunately this shape, which featured a rather sharp LE on the undersurface, generated a vortex on the undersurface of the wing in supersonic flight and low AoA (near zero 'g'). This vortex got into the intake and caused the engine to surge, so we had to redesign the LE ahead of the intakes as shown. This cost us a little subsonic drag, so you can see from the diagram what you need to do to keep subsonic cruise drag down.

Hope this answers the questions

CliveL

atakacs

Quote:

Just wondering was that the maximum speed "in" the design ? I understand that "the higher & the colder = the faster" was the key to the performance and that the Mach +/- 2.0 cruise was implied by limiting altitude to FL 600 in order to mitigate cabin depressurization consequences. I guess there where also thermal issues but was, say, Mach 2.2 @ FL700 "warmer" than Mach 2.0 @ FL600 ?

Really an answer for CliveL, but I'll have a go. The short answer to your question is 'oh yeah, big time'. Total temperature varies with the SQUARE of Mach number and static temperature. Depending on the height of the tropopause itself as well as other local factors, there can be little or no significant variation of static temperature between FL600 and FL700. The 400\xb0K (127\xb0C) Tmo limit was imposed for reasons of thermal fatigue life, and equates to Mach 2.0 at ISA +5. (Most of the time the lower than ISA +5 static air temperatures kept us well away from Tmo). In a nutshell, flying higher in the stratosphere gains you very little as far as temperature goes. (Even taking into account the very small positive lapse above FL 650 in a standard atmosphere). As far as the MAX SPEED bit goes, Concorde was as we know flown to a maximum of Mach 2.23 on A/C 101, but with the production intake and 'final' AICU N1 limiter law, the maximum achievable Mach number in level flight is about Mach 2.13. (Also theoretically, somewhere between Mach 2.2 and 2.3, the front few intake shocks would be 'pushed' back beyond the lower lip, the resulting flow distortion causing multiple severe and surges).
On C of A renewal test flights (what I always called the 'fun flights') we DID used to do a 'flat' acceleration to Mach 2.1 quite regularly, as part of the test regime, and the aircraft used to take things in her stride beautifully. (And the intakes themselves were totally un-phased by the zero G pushover that we did at FL630). This to me was an absolute TESTAMENT to the designers achievement with this totally astounding aeroplane , and always made me feel quite in awe of chaps such as CliveL.

Quote:

Also wondering what was the max altitude ? Was high altitude stall (for the lack of a better word) ever experimented during tests or training ?

Well the maximum altitude EVER achieved in testing was I believe by aircraft 102 which achieved 68,000'. As far as the second part of your question goes, not to my knowledge (gulp!!) but perhaps CliveL can confirm.

Shaggy Sheep Driver
So glad you are enjoying the thread, and absolutely loved the description of your flight in OAD and your photo is superb. I don't think it is possible to name a single other arcraft in the world that could be happily flown hands off like this, in a turn with 20\xb0 of bank at Mach 2. (One for you ChristiaanJ; The more observant will notice that we are in MAX CLIMB/MAX CRUISE with the autothrottle cutting in in MACH HOLD. Oh, we are in HDG HOLD too ).
Now for your question A very good question. The anti-skid system used a fixed simulated nose wheel rolling speed Vo signal as soon as the undercarriage was down and locked, this was confirmed by the illumination of the 8 'R' lights on the anti-skid panel. The illumination of these lights confirmed that there was full ant-skid release from the relevant wheel, due to there being of course zero output initially from the main gear tachos but this simulated Vo output from the nose gear tacho. The Vo signal therefore ensured that the aircraft could not be landed 'brakes on' (all the main wheels think they are on full skid) and that there was anti-skid control pending lowering of the nose-wheel. As the main wheels spin up on landing, their tacho outputs now start to back off the Vo signal, and braking can commence. As the nose leg compresses, the Vo signal is removed and the Nose-wheel tachos(their were 2 wired in parallel) spin up, their output will now replace the Vo signal, and full precise anti skid operates.
As far as your air conditioning question goes, you needed an external air conditioning truck to supply cabin air on the ground. Not needed in the hangars of course, but come departure time if these trucks were not working, then the cabin could become very warm/hot place indeed (depending on the time of year). Oh for an APU
Best regards

Dude

As usual Dude you beat me to it! I really must give up having another life

As Dude says, the 'cruise' condition was set by the aircraft specification for transatlantic range on an 85% (ISA +5) day and the chosen mach Number was 2.0 (of which more anon). This gives a Total Temperature of 400.1 deg K. [Dude, I know your pipe-smoking thermodynamicist and he was having you on - he is quite capable of memorising the square/square root of 407.6 or whatever!]
To give margins for sudden changes in ambient temperature (we had to cater for a 21 deg change in one mile) the Mmo was set at 2.04 which matches 400 degK at ISA +1. In theory then we could have flown faster than our chose Mmo at anything colder than this, but there are two limits:

1) The object is not to fly as fast as you can but to fly with minimum miles/gallon. If you have a nice cold day and enough thrust to go either faster or higher which do you choose? For best specific range you go higher every time.
2) The thing that everyone forgets is that civil aircraft have to have margins around their authorised envelope. In Concorde's case these were set principally by the intake limits and engine surge.

Dude also says quite correctly that 101 flew to 2.23M but the production aircraft was limited to 2.13M. Now you may not believe this, but 101 could fly faster than the production aircraft because she (101) leaked like a sieve!.

I doubt I will get away with that without some explanation

Once you get past a certain Mach Number the airflow into the intake is fixed. The performance (intake pressure recovery and engine face flow distortion) then depends on how this air is shared between the engine and the throat 'bleed'. This bleed was ducted over the engine as cooling air and then exhausted (in principle) throught the annulus formed between the expanding primary jet and the fixed walls of the con-di nozzle. But if you took, or tried to take, more bleed air the intake pressure recovery went up and the primary jet pipe pressure went up with it. This meant that the primary jet expanded more and squeezed the available annulus area which restricted the amount of bleed air one could take.

Obviously if there are alternative exit paths between intake and final nozzle then you can take more bleed air off and the engine face flow distortions will benefit along with the surge margin. 101 was fairly 'leaky' in this respect, particularly around the thrust reverser buckets on the original nozzle design. This meant that 101's intake distortions were lower than the production aircraft so she could fly faster without surge - at least with the first attempt at intake control 'laws'. We managed to tweak most of the margin back eventually. Engine bay leaks were good for surge margin but VERY bad news for m.p.g.!

Here are a couple of diagrams to show what I mean. the first shows the surge lines for the various aircraft variants and also the N1 limiter Dude was talking about. NB: the X-axis is LOCAL Mach Number not freestream. The difference comes from the compression of the underwing flow by the bit of the wing ahead of the intake. Mmo + 0.2 is shown

">The next shows the surge free boundaries in sideslip and normal acceleration. You can see the zero 'g' capability Dude was enthusing over. ">

As for 'high speed stall', I don't think we ever contemplated trying it! Our requirements in 'g' capability were defined and that was it. Besides, the aircraft would fly like the proverbial stone-built outbuilding at those sorts of conditions so I don't think one would have been able to get anywhere near a stall in the conventional sense. Stall as commonly defined for subsonics (deterrent buffet) might have been another matter, but I don't remember anything.

Cheers

No, it is just the opposite. It delays the formation of vortices so that one can develop some leading edge suction on forward facing surfaces and keep the drag down in subsonic cruise. It is there inboard and not outboard because inboard there is some wing thickness which allows one to get some decent forward facing area whereas out board the wing is too thin to make it worthwhile.

The prototype had even more 'droop' in front of the intakes, but that produced a vortex at low incidence (near zero 'g') that went down the intakes and provoked surge.

Normally in subsonic aircraft yes, but in this case the reason is that since the tips are well behind the CG the washout makes the tips give an effective nose-up pitch which helps trim the aft movement of lift as you go supersonic.

Cheers

Clive

PS: Everyone seems to be adding their favourite Concorde photograph so I thought I would be different and add my LEAST favourite

">

Hello skyhawkmatthew!

M2dude gave a good answer on your question in post #1085, so I think I may quote this here again.

Re : 9min to mach 2.

Not sure you can get CG back that quickly.

In the (restored) Sim with a lightweight fuel load that will not get you anywhere and not bothering about the CG, the absolute minimum time to Mach 2 at 50,000ft on a pretty constant VMO chase is just over 15mins, so really unlikely that this was possible in real life....but will stand corrected if someone says other wise.

The A/C had diverted to cardiff as they had suffered a engine surge due to a double intake lane failure and had to slow to subsonic early. That coupled with additional time with engines running at JFK meant they were just not comfortable about coming to London and possibly declaring a fuel emergency.

For the french speaking (or reading) people here, I just found a mine of very interesting informations about Concorde on this website:
Accueil
This site has a database of thousand of concorde flights with the following datas: Date and time of the flight, airframe used, technical and commercial crews, guests, departure/arrival airports and flight type (regular, charter world tour...).
On top of that, many infos and stories around Concorde can also be found there.
I can't resist to translate one of those stories (I'm far from being a native english speaker or a professional translator; so forgive me for the misspellings and other translation mistakes). It is a report about one of the biggest incident that happened to the prototype 001 during the flight tests:

Shock of shockwaves
We were flying with Concorde at Mach 2 since 3 month already on both side of the Channel. The prototype 001 did outstrip 002 which was supposed to be the first to reach Mach 2.
Unfortunately, a technical issue delayed 002 and Brian Trubshaw fairly let Andr\xe9 Turcat be the first to reach Mach 2 with the 001 which was ready to go.
The flight tests were progressing fast and we were discovering a part of the atmosphere that military aircrafts hardly reached before. With Concorde, we were able to stay there for hours although limited by the huge fuel consumption of the prototypes.
The Olympus engines did not reached their nominal performance yet and, most of the time, we had to turn on the reheat in supersonic cruise to maintain Mach 2.
The reheat is what we call afterburner on military aircrafts. Fuel is injected between the last compressor stage of the low pressure turbine and the first exhaust nozzle. This increases the thrust for the whole engine and its nozzle.
The 4 reheats, one for each engine, are controlled by the piano switches behind the thrust leavers on the center pedestal between the two pilots. Air was fed into the engines through 4 air intakes, one for each engine, attached 2 by 2 to the 2 engine nacelle, one under each wing. The advantage in terms of drag reduction was obvious.
However, tests in wind tunnel showed that, at supersonic speed, if a problem happens on one engine, there was a great chance for the adjacent engine to be affected as well by the shockwave interference from one air intake to the other despite the presence the dividing wall between the two intakes. So we knew that an engine failure at mach 2 would result in the loss of 2 engines on the same side, resulting in a lateral movement leading to a strong sideslip that would likely impact the 2 remaining engines and transform the aircraft into the fastest glider in the world.
This is why an automatic anti sideslip device was developed and installed on the aircrafts.
The air intakes are very sophisticated. At mach 2, it creates a system of shockwaves that slows down the air from 600 m/sec in front of the aircraft to 200 m/sec in front of the engine while maintaining a very good thermodynamic performance. In supersonic cruise, the engines, operating at full capacity all the time, were sensitive to any perturbation and reacted violently with engine surge: the engine refusing the incoming air.
Stopping suddenly a flow of almost 200kg of air per second traveling at 600m/sec causes a few problems. As a result, a spill door was installed under the air intake and automatically opened in such event.
To control the system of shockwaves and obtain an efficiency of 0,96 in compression in the air intake, 2 articulated ramps, controlled by hydraulic jacks, are installed on the top of the air intakes in front of the engines. Each ramp is roughly the size of a big dining room table, and the 2 ramps, mechanically synchronized, move up or down following the instruction of an highly sophisticated computer that adapts the ramp position according to the mach number, the engine rating and other parameters such as skidding.
At that time, it was the less known part of the aircraft, almost only designed through calculation since no simulator, no wind tunnel, did allow a full scale test of the system.
The control of the system was analog and very complex but it was not easy to tune and we were moving ahead with a lot of caution in our test at mach 2.
On the 26th of January 1971, we were doing a nearly routine flight to measure the effect of a new engine setting supposed to enhance the engine efficiency at mach 2. It was a small increase of the rotation speed of the low pressure turbine increasing the air flow and, as a result, the thrust.
The flight test crews now regularly alternate their participation and their position in the cockpit for the pilots.
Today, Gilbert Defer is on the left side, myself on the right side, Michel R\xe9tif is the flight engineer, Claude Durand is the main flight engineer and Jean Conche is the engine flight engineer. With them is an official representative of the flight test centre, Hubert Guyonnet, seated in the cockpit's jump seat, he is in charge of radio testing.
We took off from Toulouse, accelerated to supersonic speed over the Atlantic near Arcachon continuing up to the north west of Ireland.
Two reheats, the 1 and the 3, are left on because the air temperature does not allow to maintain mach 2 without them.
Everything goes fine. During the previous flight, the crew experienced some strong turbulence, quite rare in the stratosphere and warned us about this. No problem was found on the aircraft.
We are on our way back to Toulouse off the coast of Ireland. Our program includes subsonic tests and we have to decelerate.
Gilbert is piloting the aircraft. Michel and the engineers notify us that everything is normal and ready for the deceleration and the descent.
We are at FL500 at mach 2 with an IAS of 530 kt, the maximum dynamic pressure in normal use.
On Concorde, the right hand seat is the place offering the less possibility to operate the systems. But here, we get busy by helping the others to follow the program and the checklists and by manipulating the secondary commands such as the landing gear, the droop nose, the radio navigation, comms, and some essential engine settings apart from the thrust leavers such as the reheat switches.
The normal procedure consists in stopping the reheat before lowering the throttle.
Gilbert asks me to do it. After, he will slowly reduce the throttle to avoid temporary heckler. Note that he did advise us during the training on the air intake to avoid to move the thrust leaver in case of engine surge.
As a safety measure, I shut down the reheat one by one, checking that everything goes fine for each one. Thus I switch off the reheat 1 with the light shock marking the thrust reduction. Then the 3\x85
Instantly, we are thrown in a crazy situation.
Deafening noise like a canon firing 300 times a minute next to us. Terrible shake. The cockpit, that looked like a submarine with the metallic and totally opaque visor obviously in the upper position, is shaken at a frequency of 5 oscillation a second and a crazy amplitude of about 4 to 5 G. To the point that we cannot see anymore, our eyes not being able to follow the movements.
Gilbert has a test pilot reaction, we have to get out of the maximum kinetic energy zone as fast as possible and to reduce speed immediately. He then moves the throttle to idle without any useless care.
During that time, I try, we all try to answer the question: what is going on? What is the cause of this and what can we do to stop it?
Suspecting an issue with the engines, I try to read the indicators on the centre control panel through the mist of my disturbed vision and in the middle of a rain of electric indicators falling from the roof. We cannot speak to each other through the intercom.
I vaguely see that the engines 3 and 4 seem to run slower than the 2 others, especially the 4. We have to do something. Gilbert is piloting the plane and is already busy. I have a stupid reaction dictated by the idea that I have to do something to stop that, while I can only reach a few commands that may be linked to the problem.
I first try to increase the thrust on number 4 engine. No effect so I reduce frankly and definitively. I desperately look for something to do from my right hand seat with a terrible feeling of being helpless and useless.
Then everything stops as suddenly as it started. How long did it last, 30 seconds, one minute?
By looking at the flight data records afterward, we saw that it only last\x85 12 seconds!
However, I have the feeling that I had time to think about tons of things, to do a lot of reasoning, assumption and to have searched and searched and searched\x85! It looked like my brain suddenly switched to a fastest mod of thinking. But, above all, it's the feeling of failure, the fact that I was not able to do anything and that I did not understand anything that remains stuck in my mind forever.
To comfort me, I have to say that nobody among the crew did understand anything either and was able to do anything, apart from Gilbert.
The aircraft slows down and the engine 3 that seemed to have shut down restart thanks to the auto ignition system. But the 4 is off indeed.
Michel makes a check of his instruments. He also notes that the engine 4 has shut down but the 4 air intakes work normally, which makes us feel better. After discussing together, we start to think that we probably faced some stratospheric turbulence of very high intensity, our experience in this altitude range being quite limited at that time. But nobody really believes in this explanation. Finally, at subsonic speed, mach 0.9, with all instruments looking normal, we try to restart engine 4 since we still have a long way to go to fly back to Toulouse.
Michel launches the process to restart the engine. It restarts, remains at a medium rotation speed and shuts down after 20 seconds, leaving us puzzled and a bit worried despite the fact that the instrument indicators are normal.
Gilbert then decide to give up and won't try to restart this engine anymore and Claude leaves his engineer station to have a look in a device installed on the prototype to inspect the landing gear and the engines when needed: an hypo-scope, a kind of periscope going out through the floor and not through the roof.
After a few seconds, we can hear him on the intercom:
"Shit! (stuttering) we have lost the intake number 4."
He then describes a wide opening in the air intake, the ramp seems to be missing and he can see some structural damages on the nacelle.
Gilbert reacts rapidly by further reducing the speed to limit even more the dynamic pressure.
But we don't know exactly the extent of the damage. Are the wing and the control surfaces damaged? What about engine 3?
We decide to fly back at a speed of 250 kts at a lower altitude and to divert toward Fairford where our british colleagues and the 002 are based. I inform everybody about the problem on the radio and tell them our intentions. However, I add that if no other problems occur, we will try to reach Toulouse since we still have enough fuel.
Flying off Fairford, since nothing unusual happened, we decide to go on toward Toulouse. All the possible diversion airport on the way have been informed by the flight test centre who follows us on their radar.
At low speed, knowing what happened to us and having nothing else to do but to wait for us, time passes slowly, very slowly and we don't talk much, each one of us thinking and trying to understand what happened. However, we keep watching closely after engine 3.
Personally, I remember the funny story of the poor guy who sees his house collapse when he flushes his toilets. I feel in the same situation.
Gilbert makes a precautionary landing since we don't rely much on engine 3 anymore. But everything goes fine.
At the parking, there is a lot of people waiting for us and, as soon as the engines stop, we can see a big rush toward the nacelles of the right hand side engines.
Gilbert and myself are the first to get off the plane and we are welcomed down the stairs by Andr\xe9 Turcat and Jean Franchi who came out from the crowd watching at the right hand side nacelle.
They both behave the same way, with a slow pace attitude, the same look, a mix of disbelief and frustration.
Andr\xe9 is the first to speak: "I can't believe we were not on this flight, really unlucky\x85". Yes, this flight was supposed to be just a routine flight\x85!
The condition of the nacelle is impressive. We come closer and everybody move aside for us with a look of disbelief and respect as if we were hell survivors.
The ramps of the intake 4, those 2 "dining tables", have completely disappeared leaving a hole where we can see the hydraulic jacks and the stub rod where the ramps were attached.
Indeed, only the ramps were missing, apparently ejected forward which was unbelievable knowing how fast we were flying. The ramp slipped under the nacelle causing some damages on it and on the hood of one of the elevon's servo control. Fortunately, the control did not suffer any damage.
What is left of the rear ramp seems to be blocked down inside the intake in front of the engine and we can see behind it the first blades of the compressor, or what is left of it, not much.
The engine swallowed a huge amount of metal but no vital parts of the aircraft has been damaged, no hydraulic leaks, no fuel leaks. I remembered at that time the stories of some B58 Hustler accident where the loss of an engine at mach 2 almost certainly ended with the complete loss of the aircraft. Our Concorde has only been shaken. This incident strengthened the trust I had in this plane. And I was not unhappy to have experienced this ordeal, especially when I saw the frustration on the face of Andr\xe9 Turcat and Jean Franchi.
But we had to understand what happened and how; and also why the ramp's fixing broke.
It didn't take much time to get the answers.
I unintentionally triggered the problem when shutting down the reheat of engine 3. The sudden stop of the fuel flow did of course stop the combustion and the back pressure behind the low pressure turbine. But, probably because of the modification made on the engine before the flight, the stop of the reheat has not been followed by the normal closing movement of the primary nozzle to compensate the pressure drop. So the low pressure turbine ran out of control, dragging down the low pressure compressor which reacts by surging.
Despite the opening of the spill door, the engine surge led to a sudden movement of the shockwaves in the air intake creating a surge in the intake itself. A similar surge happened in the adjacent intake 4 followed by a surge of the corresponding engine. This caused an excessive pressure above the ramps and the fixings of the intake 4 did not hold.
Since it was the first time we experienced a surge in the air intake, we had little knowledge of the stress it would create on the ramps. This led to miscalculation of the strength of the ramps's frames and they did brake.
Another mistake: instead of installing the motion detectors on the ramp itself, to make the production easier, they have been placed on the arms of the hydraulic jacks. This is why Michel R\xe9tif thought that the position of the ramps were correct. The hydraulic jacks did not suffer any damage and were still working normally even if the ramps were missing.
All the data recorded during this event helped us in redesigning the air intakes and the flight test program resumed three month later.
After this, we deliberately created dozen and dozen of air intake surge to fine tune the way to regulate them with digital calculator this time.
From now on, even if it was still very impressive, it was safe and their intensity was not comparable with what we experienced with the missing ramps.
However, a french president may kept a lasting memory of this, much later, during a flight back from Saudi Arabia. This time, I was on the left side, Gilbert on the right and Michel was still in the third seat\x85 But that's another story.
For me, the lasting impression of failing and being helpless during this incident made me wonder what a commercial pilot would have done in this situation. This plane was designed to be handled by standard commercial pilots and not only by the flight test pilots.
At that time, I was interested in taking in charge the management of a training center for the pilots of the future Airbus's clients. This event pushed me that way and I made it clear that I wanted to add the flight training on Concorde in this project. This has been agreed and I did it.
And the Concorde training program now covers the air intake surges and how to deal with them.
Jean PINET
Former test pilot
Member and former president of the Air and Space Academy

ruddman


...Being that the Concorde looks like a slippery sob, how were the descents planned?...

The distance required to decel/descend from M2.0 in cruise/climb down to 3,000 at 250 kts was obtained from a checklist chart. Entering with the (expected) FL at Top of Descent and then correcting for the average wind component expected in the descent and also for the temperature deviation from ISA gave the required track miles. It wasn\x92t used a lot, because generally the more critical descent requirement was to decelerate so as to be (just) below M1.0 at a specified point on the arrival route, for noise reasons, to avoid booming land.

There was a second chart, utilised in the same way as the first, which provided this information. Sometimes this distance might need to be increased a little, as, if a subsonic cruise was expected before continuing the approach, the engines were \x93warmed\x94 up at M0.97 and after passing FL410, by the application of power, for one minute, by the Flight Engineer.



...Did you just pull the throttles back to flight idle?...

Only if you were willing to run the risk four pop surges from the engines and the near certainty of a clip round the ear from your Flight Engineer.

Usually the pilots handled the throttles from \x93Power Up to Gear Up\x94 and from \x93Gear Down to Shut Down\x94. The Flight Engineer generally did all the rest, which, thankfully, left all the tricky drills and procedures as his responsibility.



...Or was there a little more engine management and more gradual handling of the engines and descent?...

On a normal decel/descent, the handling pilot would select ALT HOLD and then ask the Flight Engineer to reduce power to 18\xba TLA (Throttle Lever Angle). The speed would decay to 350 kts IAS (Indicated Air Speed) IAS HOLD was engaged and the descent flown at 350 kts IAS. The next power reduction (32\xba TLA) came when, still flying at 350 kts IAS, the Mach number reduced through M1.50.


...And I'm guessing the approach speeds were fairly high so hitting the touchdown zone was pretty important?...

In terms of not running off the end of the runway, touching down in the correct spot was as important on Concorde as on other aircraft types.

However, due to the geometry of Concorde on landing, the tail, engine pods and reverser buckets were already fairly close to the runway. Add in a \x93firm\x94 touchdown, or if the wings are not completely level, and ground clearance becomes marginal, so a prolonged flare and floated landing, with an increasing aircraft attitude, was not acceptable. The risk of a pod, tail or a reverser bucket scrape on Concorde was greater than on most conventional jet aircraft.



... So if things got out of shape a little, and a G/A was required, how do you handle what looks like 4 rockets on the wings and applying the right amount of power?...

• Disconnect the autothrottles.
• Apply FULL power without reheat.
• Rotate to 15\xba and level the wings.
• Check for Positive Climb then call for the Gear Up.
• Maintain 15\xba and accelerate (you will accelerate!)
• Passing around 210 kts, reduce power to 95% N2.
• Approaching 250 kts, engage Autothrottles for 250 kts
• Reduce Pitch Attitude, aiming to achieve 2,000 fpm RoC.
• Do not miss the level off altitude for the GA profile.

@EXWOK


There was a certification requirement for descent time from FL600 down to FL100 if I recall correctly. Can't remember the value though. In flight reverse was developed to trim some fraction of a minute off the time to get inside the requirement


@ a_q

Not sure what you mean by a "leaky" intake. At about 2.2M the first shock would hit the intake lower lip and from that point on the total intake mass flow was frozen. Increased engine mass flow could only be obtained by reducing bleed flow and that gave higher engine face flow distortions driving the engine towards surge and lower intake recovery. So engine mass flow was effectively fixed also.
Then the amount of "dry" fuel which could be added was limited because the higher Mach number increased the engine entry temperature but the maximum turbine entry temperature was fixed.
You could add thrust by using reheat, but you would not get as much as you would like because the final nozzle, being designed for 2.0M would be too small for optimum efficiency at higher Mach numbers.
Overall, IIRC we got to 2.23M in flight test. If you pushed me I would say it might be possible with reheat etc to get to 2.25 or 2.26M, but it would be a blind guess!

Bit of a hypothetical question requiring a judgemental response!
My short answer would be not much more than the certified limits - at least not without significant modifications.
FL680 was achieved at the end of a zoom climb, so the Mach No was a lot less than 2.0
M2.23 was in a shallow dive. The object was to demonstrate sufficient margin to avoid surge following the worst temperature transient specified in the TSS regulations. To that end both the intake laws and engine operating lines were tweaked as functions of Mach No to minimise intake flow distortions and maximise surge margin. The result was a long way from the performance optimum one would need for steady cruise.
The power plant was being pushed to its limits at this Mach No.
(As an aside, the subsonic rules make no mention of temperature transients as a cause of Mach exceedences. Some recent incidents suggest this could usefully be reviewed)
The altitude limit could perhaps be more readily expamded. The aircraft normally flew a cruise climb bcause at Concorde cruising altitudes there was no ATC conflict. The altitude was very sensitive to ambient temperature and aircraft weight. FL600 would be associated with end of cruise on a coolish day.
To usefully increase cruise altitude would require more engine thrust, but this could only be obtained by increasing engine TET which would screw the engine fatigue life.
Increasing Mmo from 2.04 would need an increase in Tmo (400 deg K) at any temperature above (from memory) ISA. This in turn would affect the airframe fatigue life unless the structural material were changed. Even then, there were a lot of nonmetallic bits (seals etc) that would also have needed replacement.
Sorry if this is a gloomy assessment, but that is the way I see it!