A more reasonable Requirements process

[pathology_doc]

being far more hard nosed on what to fund and what not would have been far better.

e.g. no TSR2, no Concorde, no V-1000 etc.

The problem with TSR-2 is not that it was funded, it's that the requirement changed so much from when EE were first given the nod. If it hadn't been used as the cudgel with which to amalgamate the industry and if the design had been frozen without being meddled into a mid-range strategic nuclear platform, EE could have given them an in-house winner to go with its other two in-house winners (Canberra and Lightning). Vickers could and should have been second fiddle, not co-manager.

Too many cooks spoiled that broth (and some of them were not good cooks).

 

[zen]

Certainly there is a strong argument for just letting EE or Avro get on with it.
An austere version while they worked out the computer issues, would have been a reasonable solution.

 

[zen]

Certainly there is a strong argument for just letting EE or Avro get on with it.
An austere version while they worked out the computer issues, would have been a reasonable solution.

 

[Archibald]

France attempted to splurge out on hardware - 32 prototypes in 1949 alone (shocked me when I read that stat yesterday!) but the majority turned out to be flawed or went no further. So emphasis on building everything on the drawing board doesn't pay off.

It's a tortured story. The two key years are 1948 and 1952. The former had the public owned SNCAs firing by all tubes in a rather optimistic and disastrous way.
All this because that fellow, while a decent human being, made a very, very bad aviation minister in 1946.

Charles Tillon - Wikipedia

en.wikipedia.org

Tillon in 1946 tried to restart a very ruined French aviation industry with massive orders (of pre-WWII and piston engine aircraft) PLUS pushing very hard to catch back and enter the jet age: going full bore on prototypes, including LW wunderweapons sized in the French occupation zone.

You guess: it didn't worked too well. Take the Arsenal VG-90. At first glance a very sexy swept wing jet powered fighter. What's not to like ? except deeply flawed. Same for the SO-6020 Espadon.
Now compare that to the SO-6000 Triton (newt - I kid you not) and Dassault Ouragan... and you can see what was wrong.

By 1952 the government accepted a mix of private companies entered the picture (Fouga, Dassault, Breguet, Turboméca) and succeeded where 1948 had failed miserably.

Case in point: naval fighters.

In 1948 for PA28 "future carrier" were started three SNCA prototypes
- Arsenal VG-90: two built, two crashes, two pilots dead.
- SNCAC 1080: one built, one crash, one pilot dead
- Nord 2200: no crash, decent aircraft, but completely overweight for its Nene turbojet.
And then PA28 was canned on the drawing board.

End result: a licence was taken by the SNCASE to build Hawker Sea Venoms. In 1952.

Same for transport aircraft: 1948 had the Cormoran disaster, 1952 got the Noratlas staying in service until the 1980's.

Same for fighters: 1948 Espadon vs 1952's Mistral & Ouragan.

Bombers: 1948 SO-4000 was a disaster, 1952 Vautour worked fine.

this guy finally understood where was the problem

Pierre Marie Gallois - Wikipedia

en.wikipedia.org

And fixed much more performance levels and requirements.

End result: 1952 saw the first flights or IOC of such smashing success as Dassault Ouragan / Mystère series; Noratlas transport; Alouette helicopter; Fouga Magister trainer.
These four alone become cornerstones and remained in services for decades thereafter. Performance wasn't shiny or stellar, but they WORKED and were build in numbers.

Bottom line: Gallois applied the KISS principle. Keep it simple, stupid. Dassault, Turboméca, Alouette helicopters: same basic idea.

 

[red admiral]

Certainly there is a strong argument for just letting EE or Avro get on with it.
An austere version while they worked out the computer issues, would have been a reasonable solution.

So still massively expensive but now literally just a Buccaneer that can go supersonic? Which will hopefully get better avionics in a decade after more expense?

I'd prefer 200 Tornadoes and 200 Jaguars over 60ish TSR2s; or 60 F-111Ks which have working avionics and are much cheaper

Anyway this wasn't specifically a TSR2 bashing thread...

For some areas there's a need for a niche/bespoke design e.g. V-Bombers. For areas where there is more choice then there's a need to be more hard nosed about either not starting designs at all, or setting requirements sufficiently that you can carve out a big enough niche yourself e.g. Jaguar. It also helps if someone else pays the bills e.g. Harrier, Hunter

 

[pathology_doc]

I'd prefer 200 Tornadoes and 200 Jaguars over 60ish TSR2s;

Wasn't TSR-2 the Tornado of its day, at least in terms of avionics and ambitions, and weren't some of the lessons learned applied to the newer aircraft?

 

[red admiral]

Wasn't TSR-2 the Tornado of its day, at least in terms of avionics and ambitions, and weren't some of the lessons learned applied to the newer aircraft?

Far higher cost

Lessons learnt are probably one of the main side venefits from the cancelled projects. But it would be far cheaper to fund R&D of specific subsystems instead - not a great deal of knowledge generated by all the people doing fuselage structural calcs with slide rules.

 

[zen]

So still massively expensive but now literally just a Buccaneer that can go supersonic?

Yes and no.
You have a valid point on the low level supersonic cost issue. Though kinda missed the problem that low level got tried in Vietnam and the US discovered why everyone had gone higher in the first place.
Hence the investment into EW and ultimately stealth.
But Frankly the EE or Avro offerings are more intrinsically multirole and consequently more F4-like.

An "expensive" (but priced in Pounds not precious Dollars) twin engined Thunderchief. Might be a more appropriate description.

Which will hopefully get better avionics in a decade after more expense?

The cost of TSR.2 was born in development.
The cost of the AFVG avionics was born, fed beyond UKVG to MRCA.

I'd prefer 200 Tornadoes and 200 Jaguars over 60ish TSR2s; or 60 F-111Ks which have working avionics and are much chcheaper

228 Tornado IDS and 165 ADV.
200 Jaguar solutions very close to earlier P1154 numbers.
But strictly we could have gotten P.17 numbers in similar scale to Tornado and a lot earlier.

 

[Hood]

Plus, to single out the poor TSR.2 again, requirements change. 1957 to 1969 (when it might have appeared in service) was a long time against the ever changing backdrop of requirements.
Vietnam experience is irrelevant - wrong timing to influence the original requirements and by the time of the bulk of USAF involvement TSR.2 was heading to the scrap merchants. Even if it hadn't been cancelled in 1964 it would have been a dinosaur in the cluster bomb-lugging, laser-guided, missile-carrying era that followed. AVFG was light years away from a hulking great bomber larger than a Lancaster lumbering around with a few 1,000lb dumb bombs.

It is unfair to say the Air Staff of 1957 should have forseen the needs of 1970, they thought a TSR.3 would be coming for their post-1970 needs. The MoS/MoA didn't count on a 15-year R&D programme, how could they from their previous experience? Bucc and Canberra worked out of the box, the V-Bombers took less than a decade from paper to service.
The MoA made a bog up of the organisation and industry couldn't patch it up alone and were in disarray too. The pressure on TSR.2 as the programme to cement the industry together created too much stress and worry which resulted in interference and failure.
In contrast the similar stress to provide the nation with a nuclear deterrent at a time when atomic annihilation seemed possible produced three aircraft that worked well and the weapons (varied successes here). Sure in hindsight it seemed wasteful but it worked, whereas ordering fighters off the drawing board based on lashed together research aircraft (Swift) didn't work. Writing the requirements for a long-distance transport after the design has been chosen (V.1000) in a shocking twist didn't work either (then lying about cancellation due to costs as a cover up).
As for airliners, who knows what was going through BOAC and BEA's minds half the time, I can't work it out. Their thinking resembles an Escher puzzle in the shape of a Boeing.

 

[JFC Fuller]

Plus, to single out the poor TSR.2 again, requirements change. 1957 to 1969 (when it might have appeared in service) was a long time against the ever changing backdrop of requirements.
Vietnam experience is irrelevant - wrong timing to influence the original requirements and by the time of the bulk of USAF involvement TSR.2 was heading to the scrap merchants. Even if it hadn't been cancelled in 1964 it would have been a dinosaur in the cluster bomb-lugging, laser-guided, missile-carrying era that followed. AVFG was light years away from a hulking great bomber larger than a Lancaster lumbering around with a few 1,000lb dumb bombs.

These views get stated frequently on this forum and elsewhere but I am not really sure they stand up to scrutiny. Yes, TSR-2 was a big aircraft, but its intended radius/payload performance was also significantly greater than both AFVG and MRCA (crudely; 1,000nm, 600nm, and 400nm respectively), the size difference doesn't seem especially disproportionate. The OR for TSR-2 didn't change that much either, it remained a low altitude theatre bomber with a significant reconnaissance capability. Air-to-surface guided weapons (to be Martel ultimately) were one of the requirements in OR.343, no reason it couldn't have received an LGB capability later too. In my interpretation, industry were given a challenging but actually quite static requirement to meet. Had it have been pulled off, with no major flaws, it would probably have had a long and illustrious career like the F-111 did with the USAF.

In relation to the topic of this thread the real questions are where did the requirements come from? e.g. why was 1,000nm radius so important? The means of operation, low altitude flight using SLAR, INS, Doppler, and Forward Looking Radar almost seem lifted from the OR.324 studies of the early-to-mid 1950s. My suspicion is that the reconnaissance capability was a direct product of Sandys-proofing; when the RAF had told him in 1957 that they needed manned bombers for "flexibility" he had retorted that deterrence involved hitting fixed targets so no such flexibility was necessary, after some flailing the RAF had managed to hold the line that reconnaissance required a manned platform. Had they have turned up with an OR for a manned TS-2 instead of a TSR-2 they may have been told to come back with an OR for a ballistic missile.

 

[Hood]

Yes, TSR-2 was a big aircraft, but its intended radius/payload performance was also significantly greater than both AFVG and MRCA (crudely; 1,000nm, 600nm, and 400nm respectively), the size difference doesn't seem especially disproportionate.

This is true, although the F-111 was also smaller, though the A-5 Vigilante does similarly show that fast and far meant a big airframe in the late 1950s. Aerodynamics evolved quickly, there was no chance of VG for example forming part of the solution for GOR.339 yet two years later OR.346 had Vickers VG all over it. And the OR.346 designs were smaller, in part due to naval dimensional restrictions.
In some respects given the overlaps in the ORs (OR.346 radius 1,000nm with 6,000lb, 200nm low-level, Mach 2.5, reconnaissance role) it is surprising that there were no moves to merge both types like TFX in the USA.

Air-to-surface guided weapons (to be Martel ultimately) were one of the requirements in OR.343, no reason it couldn't have received an LGB capability later too.

Maybe, but TSR.2's load capacity was puny, 6,000lb in the bay and 4,000lb external - each F-111 pylon could carry 5,000lb! External carriage was vital for GW and LGBs, the TSR.2 with four 1,000lb pylons and a small wing to boot wouldn't be best suited for that.
Even if we assume TSR.2's in-built ECM negates the need for an AN/ALQ-101 pod, a AN/AVQ-23E Pave Spike means only three Paveways.

In relation to the topic of this thread the real questions are where did the requirements come from? e.g. why was 1,000nm radius so important? The means of operation, low altitude flight using SLAR, INS, Doppler, and Forward Looking Radar almost seem lifted from the OR.324 studies of the early-to-mid 1950s.

Radius was vital for the interdiction role. For example if you look at the airports ~1,000nm away from Robin Hood Doncaster (RAF Finningley):
USSR: Tallinn, Rivine, Minsk, Pskov
Yugoslavia: Belgrade, Dubrovnik
Romania: Timişoara, Cluj-Napoca
Hungry: Debrecen
You could just about get to North Africa. Of course in terms of penetrating heavy defences the 600nm low-level radius is more likely; which gets you to Dresden, West Pomerania, Prague and almost to Poznan if you sipped fuel or got a tanker top-up.
So a UK-based TSR.2 force could interdict most supply lines to the frontlines, a Cyprus-based contingent would give an overlap too of southern targets.

Yes I agree on the OR.324 influence, accuracy in navigation was considered vital to weapons accuracy. Avionics moved on rapidly too.

My suspicion is that the reconnaissance capability was a direct product of Sandys-proofing

Not necessarily, there was Canberra PR capability that needed replacing and Valiant BPRKs and SLAR equipped Victors could hardly be expected to provide enough tactical reconnaissance and would be vulnerable over Warpac airspace. Besides they had enough on their hands to provide targeting data for the V-Force. Tactical recon was also inbuilt into the P.1154 too. Fighter PR could only penetrate so deep. You have to remember that accurate maps and radar surveys behind the curtain were not necessarily available and a lot of targeting data was required. No use sending 617 Sqn in their TSRs to bomb the railway junction at Chumpnitz if you had to guess the coordinates to punch into the nav system tapes or for the nav to spot the target on his SLAR scope when he had nothing to compare the returns against. Against mobile Warpac formations and reinforcements the need for real-time data would have been critical.

Same with colonial 'policing actions', it would take a lot of recon to find the enemy insurgents and react accordingly. That the TSR.2 was even considered for such jobs is baffling but there you go.

 

[zen]

OR.346 was inclusive of the RAF, and RAF variants of designs produced.

Tactical Laydown would at best be hitting 2 targets per flight. It was extremely optimistic to imagine a second flight could be achieved and getting into the realms of fantasy to imagine beyond that.

Conventional bombing. It was felt it was more important to hit the target than to carpet bomb.
It was also extremely expensive to field guided weapons. TSR.2 kept the guidance on the plane.
Even during the first Gulf War the majority of ordinance was unguided.
PGM was reserved for critical targets.

 

[pathology_doc]

Yes the argument of Project Cancelled is perversely contradictory "boohoo look at all the taxpayers money wasted on these duds... boohoo if only the government had ordered all our crazy dud ideas we'd have been ok. PS boohoo our duds wouldn't have been duds if the government hadn't shackled us with stupid things like requirements and budgets."

The book doesn't read as well to me after many years as it did on the first go around, but I think you're being a bit on the harsh side. No, not everything could have been built; but IMHO Wood's point is that if you look at any particular cancelled project in isolation, there were very good reasons for continuing a promising line of development instead of strangling it in the cradle the way things actually went down.

The fate of the SR.177 and thin-wing Javelin, production jigs in existence and the first aircraft already on the line when the axe falls, was particularly cruel. If you're going to argue they should never have been built, fine - but then you must also argue that they should never have been allowed to start.

 

[JFC Fuller]

OR.346 had the advantage of being a paper study so we don't really know how realistic the given performance figures were, I do have some skepticism that the Vickers 581 would have achieved TSR-2 performance whilst lugging around the structural weight required for an internal weapon bay and a couple of trim engines (which is what Vickers claimed it would do). There is a fascinating story to be told about (lack of) jointness in UK combat aircraft design, winding through the NA.39, P.177 (a rare happy example), OR.346, P.1154 and OR.355. It is a tale of paranoia, inter-service tribalism, stubbornness, political edicts, and big egos. Various civilian suggestions at jointness were resisted in the most extreme manner by both the RN (specifically Naval Air Division) and the RAF at various times.

TSR-2s external load capacity was 4,000lbs on each inner wing pylon and 1,000lb on each outer wing pylon for 10,000lbs external (though obviously it wouldn't be going 1,000nm with that!) so twin Paveway carriers could have been possible for those inner wing pylons. It is also possible a single Paveway could have been carried in the bomb bay, or that space could have been used for fuel to offset the drag from the external stores. All conjecture but rather fun and demonstrative of the fact that the aircraft was able to be a heavy hitter at long range.

I can think of lots of places within 1,000nm from various RAF accessible bases that the RAF might want to bomb. What I was, not very clearly, trying to say was that I have yet to find a specific stated rationale for 1,000nm. In submissions to various committees that figure was often tied to providing reconnaissance for the Army (who don't seem to have been as convinced as the RAF were that they needed that). Of course there was a need for reconnaissance, mostly tactical (a different requirement to strategic intelligence, e.g. what are the coordinates of a railway junction, which is something that should be known before war broke out), but it became a major component of the programme in a way I don't think it would have had it not been for the 1957-58 discussions on the role of manned bombers.

 

[Hood]

The book doesn't read as well to me after many years as it did on the first go around, but I think you're being a bit on the harsh side. No, not everything could have been built; but IMHO Wood's point is that if you look at any particular cancelled project in isolation, there were very good reasons for continuing a promising line of development instead of strangling it in the cradle the way things actually went down.

It is still an interesting read (and indeed nothing has quite covered the same breadth of projects in one volume) but it pays to be critical of Wood's hypothesis (or 'Wood’s paradigm' as Matthew Uttley coined it).

TSR-2s external load capacity was 4,000lbs on each inner wing pylon and 1,000lb on each outer wing pylon for 10,000lbs external (though obviously it wouldn't be going 1,000nm with that!)

I had overlooked that inner weight capacity, mea culpa there.

trying to say was that I have yet to find a specific stated rationale for 1,000nm.

Agreed, I have a feeling it was like the 40-city target list, something that sounded good on paper but that nobody really knew what the logic was if indeed any was applied. It just might have sounded like a nice well-rounded figure. As I say given most of those targets are deep behind belts of SAMs and MiG/Su/Yaks it seems unlikely in practice to get that far without going lo-lo-lo for reduced range.

OR.346 had the advantage of being a paper study so we don't really know how realistic the given performance figures were, I do have some skepticism that the Vickers 581 would have achieved TSR-2 performance whilst lugging around the structural weight required for an internal weapon bay and a couple of trim engines (which is what Vickers claimed it would do). There is a fascinating story to be told about jointness in UK aircraft design, winding through the NA.39, P.177 (a rare happy example), OR.346, P.1154 and OR.355. It is a tale of paranoia, inter-service tribalism, stubbornness, political edicts, and big egos. Various civilian suggestions at jointness were resisted in the most fulsome manner by both the RN (specifically Naval Air Division) and the RAF at various times.

True OR.346 was highly ambitious and unlikely to succeed. I think we know enough to take the studies with a dose of salt. Few of them look attractive as a practical proposition and Vickers in reality had a lot of work on VG to do.

The ideal Whitehall air weapon would have been a curved trident so the RAF, RN and AAC could stab each other in the back!

 

[zen]

Just on the RN side.
There is a certain coherence to the mid 50's choices.
NA.39
F.177
And yet the missing aircraft is the twin engined fighter using the same powerplant as those two.
Probably they never got to define that before events hit home.

I think it certainly vexing and confusing the RAF plucked 1,000nm RoA with TMB. Apparently it was based on operations from France and West Germany not the UK.

Again of the OR.346 designs, it's the DH offering that actually is based on a practical design.

The logical reconasense ranges are the sort covered by NMBR.3 and OR.330.
OR.339 sort of falls between the stalls here.

 

[red admiral]

When you look at geography I'd be pretty sure 1,000nm radius was driven by a specific fly to somewhere and drop bomb rather than plucked as a nice round number. It doesn't necessarily have to be driven by day one operations so maybe those SAMs and fighters are a bit sparser later in the war before the tactical nukes come out.

 

[pathology_doc]

It is still an interesting read (and indeed nothing has quite covered the same breadth of projects in one volume) but it pays to be critical of Wood's hypothesis (or 'Wood’s paradigm' as Matthew Uttley coined it).

I see the British Secret Projects books as being its inheritors, albeit from a significantly more neutral viewpoint (though I can't help but agree with Tony Buttler's expressed wish that the Delta 3 and P.1121 could have gotten at least to prototype test flight stage. I feel the same way about the XF-103).

 

[zen]

When you look at geography I'd be pretty sure 1,000nm radius was driven by a specific fly to somewhere and drop bomb rather than plucked as a nice round number. It doesn't necessarily have to be driven by day one operations so maybe those SAMs and fighters are a bit sparser later in the war before the tactical nukes come out.

Probably informed by the realisation there are practical bottlenecks in logistics transport from the USSR to Germany.
The same which had afflicted the Germans going east.
And I seem to recall must have come up when contemplating continuing the war only against the USSR in 1945.
Mostly rail I vaguely reccal.

 

[Siberia]

Maybe, but TSR-2's load capacity was puny, 6,000 lb in the bay and 4,000 lb external - each F-111 pylon could carry 5,000 lb! External carriage was vital for guided weapons and laser-guided bombs, the TSR-2 with four 1,000 lb pylons and a small wing to boot wouldn't be best suited for that.

Even if we assume TSR-2's in-built ECM negates the need for an AN/ALQ-101 pod, a AN/AVQ-23E Pave Spike means only three Paveways.

Could they not use the bomb bay similar to the F-111 and Pave Tack? It would likely need modification to fit/work but it would keep the exterior pylons free.

I have been wondering lately about the disparity, almost double, between the TSR-2's and F-111's conventional bomb carrying capacity. Whether splitting the TSR-2 role into a regional bomber via enlargement or like the proposed F-111H at the cost of some performance, and a tactical bomber based on the Buccaneer or one of its variants might not have worked better.