• Hi Guest! Forum rules have been updated. All users please read here.

Barnaby Wainfan's 7 sins of aircraft design

apparition13

I really should change my personal text
Joined
Jan 27, 2017
Messages
165
Reaction score
204
The 7 deadly sins of airplane design Barnaby Wainfan (Facetmobile builder, Northrop engineer)
https://www.eaatogether.org/live/forums/6174385951001
The presentation is definitely worth a watch.

I'm not sure if this is the right forum or not, but most of the examples he gives are post war failures. I think it might be illuminating to try and fit failures of projects we talk about to these sins to see how useful the list is. I've created a summary below.

Sin 1: invalid assumptions in specifications
- airplane is a tool
- ConOps defines the job and how it is to be done
- engineers design plane to meet the requirements of the conops
::: examples:
- B-58 only useful for penetration with nuke, not other missions after cold war thawing in late 60s.
- Multi-seat WW-2 fighters for anti-bomber role: Bell SFM-1 Airacuda, Boulton Paul Defiant (no forward guns),
- Failed business model: extrapolate last trend one step too far: SST: speed, Brabazon: comfort, Princess: runways built during WW2 eliminated need for seaplanes, A380: ETOPS change eliminated hub and spoke long haul system.

Sin 2: overconstrained
- too many constraints damage ability to perform mission
- Mission related: restrictive operating conditions, too many missions, unrealistic goals
::: Example: F-111
- non mission related: hangar size, configuration, specific size or other geometric constraints
::: Examples: Short Sterling: hangar size, standard RN hangar led to lack of wingspan which reduced operating altitude to below AA ceiling, US hardened shelters are 44 feet wide.

Sin 3. Dependence on new or immature tech: can give advantage, but: project can fail if tech doesn't work, overoptimism common, need to have a plan B.
::: Examples: Rockwell XFV-12,
JSF: both bet on new tech:
Boeing: low risk on propulsion (Harrier); high risk on thermoplastic wing skins to save weight
Lockheed: high risk on lift fan; low risk on structure
Both bet on new tech, the one that made the new tech work won.

Sin 4. Success of airplane depends on success of engine: engine must perform for aircraft to perform, erosion of engine performance hurts plane's performance (thrust, weight, SFC, reliability)
- If airplane is designed to limit of projected engine performance, airframe can out grow engine: weight increase, drag increase.
::: Example: Eclipse 500 very light jet, Williams EJ-22 turbofan: 500 lbs thrust/$40,000, engine did not produce thrust on time to support Eclipse production schedule, Eclipse bankrupt.

Sin 5. One airplane, many innovations (hobbyshop syndrome): all critical systems must work for vehicle to be successful, working out bugs in many new systems is much more difficult than doing one at a time, one significant failure can doom program.
::: Example: X-33, linear aerospike engine, composite crygenic LH2 tanks, metallic thermal protection system, new lifting body aerodynamic configuration; LH2 tank failed in testing, configuration changed dramatically as aero stability and control issues emerged.

Sin 6. Lack of margin and fallback options: design margins necessary to accommodate shortfalls as design becomes real, weight growth, thrust reduction, higher fuel consumption, higher drag, payload growth.
- Highly integrated design may not have the flexibility to adjust, design margin and fallback options must be designed in from the beginning.
The conventional wing-body-tail configuration works because of the ability to adapt the configuration during design.
::: Example: A-12 GD/MD; weight growth caused by poor load paths and difficulty making large composite parts, approach and launch speeds increased requiring wind over deck, no way to increase wing area of CLmax, span constrained by carrier suitability, sweep constrained by observables, tailless configuration cannot trim flaps.

Sin 7. concept driven design: concept believed to offer some vital advantage, concept becomes more important than suitability of the configuration, advantage may not be real, or may be offset by inherent disadvantages in the concept, design compromises to keep concept pure and fix problems it causes.
- Magic configurations: triplanes, aft tail pusher props, turboprop canards, BWB (maybe for very large aircraft, jury is still out).
::: Example: Beech starship: turboprop canard, all composite, pusher engines; there was no performance advantage over conventional layout to justify price.
The first thing to come to my mind was an example of Sin 4: the Allison T40 engine's reliability leading to the failure of all the aircraft built around it. The Skyshark and Tradewind especially were promising aircraft.

How would you categorize some other projects? Or what projects do you think failed for reasons not listed in these sins?
 
Last edited:

Fluff

CLEARANCE: Secret
Joined
Sep 9, 2019
Messages
316
Reaction score
144
Where would you put say the Nimrod MRA4?
 

Zootycoon

CLEARANCE: Secret
Joined
May 27, 2008
Messages
354
Reaction score
364
Nimrod MRA4 - Sins 1 & 7

Sin 1 Invalid Assumption in Spec, -
-Four engines was the only acceptable solution. Of course the job is being done today by a twin.

-The engineerIng resource ramp up was known to be invalid by both parties as the contract was being signed. = 2 years of delay (Original CDr 1999 slipped to 2001)

-The Russian submarine threat all but disappeared during the achieved development timescale;- hence the MR4 fleet sized dropped to unsustainable levels. = No business case

Sin 7 Concept Driven design with perceived advantage -
-Four engines (again) forced the reuse of existing airframe. Initially seen as a cost advantage but as time progressed nearly everything needed was new and costly, exceeding the ability of manufacturing to produce the vast number of small component batches = 3years of delay (First flight Dec2000 slips to Aug 2004)

<No the wings did fit = no delay despite the bull...t>

- A “make it with last century technology because FBW is known to be expensive” wish lead to cable operated ailerons which BAESYSTEMS struggled to deliver. There was nobody left who could do clean sheet design of cable flight controls = 4 years of delays from redesigns <yes 3 redesigns>, retesting, recertifications, and many associated repeat flight tests (18 months of flight testing takes 5years)

The ongoing Liliam scam - All seven Sins I reckon
 
Last edited:

helmutkohl

CLEARANCE: Secret
Joined
Nov 29, 2010
Messages
206
Reaction score
200
no examples for Sin 4?
I'm sure there are quite a few planes that had good designs but were hampered by poor engine tech..
i'm guessing a few of the PRC designs

as for the XFV-12.. had they resolved some tech issues. could that fueselage carry much? it looks like there's few ideal places for pylons
 

riggerrob

I really should change my personal text
Joined
Mar 11, 2012
Messages
967
Reaction score
461
Sin 5: Canadian Avro Arrow tried to develop a new air frame, new engines, new missiles and fire control system at the same time.
All systems were progressing, but at slow paces and high costs. The Arrow prototype flew with American-made engines because Orenda Iroquois engines were not ready in time,

This - late 1950s dilemma - reflects a bigger political problem with generals and admirals maintaining grandeous dreams in the face of decreasing defence budgets. Meanwhile, the Canadian Army wasted time and money developing the abortive Bobcat armoured personnel carrier and the Royal Canadian Navy wasted money on HMCS Brador hydrofoil even after they had decided to concentrate on flying helicopters from DDEs. Fierce inter-service rivalry say admirals and generals competing for shrinking budgets.
 

AeroFranz

CLEARANCE: Top Secret
Senior Member
Joined
May 4, 2008
Messages
2,250
Reaction score
149
no examples for Sin 4?
I'm sure there are quite a few planes that had good designs but were hampered by poor engine tech..
i'm guessing a few of the PRC designs

as for the XFV-12.. had they resolved some tech issues. could that fueselage carry much? it looks like there's few ideal places for pylons
And exactly ZERO volume for fuel in the wings...
 

apparition13

I really should change my personal text
Joined
Jan 27, 2017
Messages
165
Reaction score
204
no examples for Sin 4?
I'm sure there are quite a few planes that had good designs but were hampered by poor engine tech..
i'm guessing a few of the PRC designs

as for the XFV-12.. had they resolved some tech issues. could that fueselage carry much? it looks like there's few ideal places for pylons
I mentioned the Allison T40 which, other than a handful of Tradewinds that saw service, killed the Skyshark, Super Savage, and everything else it touched. There are a lot of designs hampered by poor engines, most of the early 50s designs come to mind, but that's a bit different than failing because the aircraft is built around an engine that doesn't appear on time or at all.

I suppose the L1011 might fit sin 4, since by the time RR and Lockheed got bailed out and had the engine working in the L1011 the DC-10 had already captured the three engine widebody market.

I've seen drawings of the XFV-12 with a pair of sidewinders on each wingtip and a couple sparrows on the fuselage, but no bombs. I think Spangenberg said as designed it had negative range, which isn't terribly useful.

riggerrob: I'm not so sure about the Arrow, since an argument frequently put forth about it and the TSR.2 is political issues killed them. If the "customer" changes their mind and becomes hostile to the program, or changes the conops during development (Sandys white paper perhaps), I don't think I would call that a design sin.

And I'm not sure what to call overoptimistic cost estimates: we can build it for this much - no we can't *cost overruns*. I don't know if that's a sales issue by overpromising to get the contract, a management issue by deliberately underbidding, a design issue by assuming problems will be easy to overcome, or a customer issue because the customer rejects realistic estimates (N/G A-12) because they want something at an unrealistic price (GD/MD A-12) and get trapped in a money pit because of it. Whatever it is, it punishes accurate assessments and rewards optimistic ones.
 

Zootycoon

CLEARANCE: Secret
Joined
May 27, 2008
Messages
354
Reaction score
364
And I'm not sure what to call over optimistic cost estimates: we can build it for this much - no we can't *cost overruns*. I don't know if that's a sales issue by overpromising to get the contract, a management issue by deliberately underbidding, a design issue by assuming problems will be easy to overcome, or a customer issue because the customer rejects realistic estimates (N/G A-12) because they want something at an unrealistic price (GD/MD A-12) and get trapped in a money pit because of it. Whatever it is, it punishes accurate assessments and rewards optimistic ones.
Very true.

Frequently now in the defence industry, the business model is based on a very profitable after sales, with the trick being not to make too big a loss to deliver the basic platform. Where this goes badly is when there’s long delivery delays;- the contractor incurs extra cost, goes into bigger debt and can’t survive until the after sales. They know a government needs the capability, so will not let them go bust and this might affect the original bid price.

This new age with significant capital projects coming once every twenty years, the pre bid to entry into service being 10-35 years (big complex projects JSF configuration studies late 80’s, F35 EIS 2015 ) making money by a simple sales transaction just doesn’t work anymore.
 

Hanz2k

The truth is grey...
Joined
Feb 17, 2012
Messages
3
Reaction score
0
Sin4 - isn’t this happening on our eyes with B777X program. Russian Myasishchev M-4 Molot.
 

riggerrob

I really should change my personal text
Joined
Mar 11, 2012
Messages
967
Reaction score
461
... riggerrob: I'm not so sure about the Arrow, since an argument frequently put forth about it and the TSR.2 is political issues killed them. If the "customer" changes their mind and becomes hostile to the program, or changes the conops during development (Sandys white paper perhaps), I don't think I would call that a design sin.... [/QUOTE]

Agreed!
Avro Arrow may not have been an engineering failure, but she was definitely a management failure. Engineers were too slow in developing all the complex new technology requested by overly-ambitious RCAF generals under a Liberal government.
As soon as John Deifenbaker and his Conservative Party won a federal election in 1958, they cancelled a few "Liberal" programs, most noteably the Avro Arrow (1959). When Deifenbaker waffled over purchasing Bomark missiles, the Conservative Party lost the next federal election.
 
Top