When do too many minor changes require a completely new design?

[riggerrob]

This question arose during discussions about American Special Forces buying the Polish PZL MC-145B Wiley Coyote, twin-engined, light STOL transport with a bewildering array of sensors and weapons hanging from the nose, belly and wings. The belly pannier already makes it look fat.
By the third nose extension, it makes Cyrano de Bergerac's nose look dainty and petite in comparison.
At what point do engineers say "enough!" stop modifications at bulkhead XYZ and design an entirely new nose????????
I understand that the primary reason that the US military buys DHC-4 Cariboo, DHC-5 Buffalo, DHC-6 Twin Otters, Alenia G 111, Shorts Sherpa, Spartan C-27, etc. is because no American factories want to compete in the light twin STOL market.

As an aside, the US military gained foreign military sales approval for the similar-sized, brand-new, Cessna 408 Sky Courier mere days after it received FAA certification (spring 2022). Does this mean that the Cessna is better-sized for the mission or that the USA wants to "buy American," etc.

For an aside, we can continue the debate about when Boeing made too many minor changes to their 737 Max.

P.S. My professional background is in the parachute industry and I can tell you tales about dozens of minor changes forced upon the FAA to the point that no original parts will fit on the updated parachute harness/container.

 

[Foo Fighter]

I suppose it SHOULD come down to external control of which changes are made and how often, from a safety POV. Internal audit and lack of proper oversight caused the 737 max fiasco.

There are military 737 variants suffering (Or suffered) from the same lack of accountability with tools left in wings for example and toilets that do not fit.

 

[Stovepipe]

There are military 737 variants suffering (Or suffered) from the same lack of accountability with tools left in wings for example and toilets that do not fit.

That kind of thing is down to gross lack of inspection by the inspectorial staff in the hangars where the aircraft is being built or serviced. back when Inspectors meant something, no panel was ever closed until the Inspector had checked inside first. I did some time on 737 overhauls and routinely found swarf and rags and the odd drill bit and discarded rivet shanks. It turned out to be a wrangle between full timers and contractors (I was a Connie but knew a lot of the fulltimers well) about who's job it was to clean up/aircraft cleaners refused to clean out swarf as it was considered by them to be the bashers' job/crew chief wouldn't rein in the abuse being given by his full time mates to the connies. Some of it was just down to laziness and unprofessionalism and the worst kind of Union mentality.

 

[Sabrejet]

Bear in mind too that a new type will require a great deal more in the way of certification than an existing type that has been modified. I suspect this is the driver.

 

[riggerrob]

Bear in mind too that a new type will require a great deal more in the way of certification than an existing type that has been modified. I suspect this is the driver.

Yes. Sometimes the new certification data requirements are based upon more accurate measuring equipment.
For example, back in 1994 I was helping Rigging Innovations do drop tests on their P124A/Aviator Pilot Emergency Parachute system. We could have simply applied for certification as a "minor change" to the existing TSO C23C certification for the Talon series of "Talon" sport parachute harness/containers, but we were also certifying a new parachute canopy and the boss wanted to be able to sell to customers who weigh more than the 254 pound limit (on suspended weight) specified in the old TSO C23C (circa 1985).
Since the newer FAA TSO C23D required measurements of opening shock, we had to partner with another manufacturer who had the specialized instrumentation to measure opening shock hundreds of times per second.

 

[BB1984]

This question arose during discussions about American Special Forces buying the Polish PZL MC-145B Wiley Coyote, twin-engined, light STOL transport with a bewildering array of sensors and weapons hanging from the nose, belly and wings. The belly pannier already makes it look fat.
By the third nose extension, it makes Cyrano de Bergerac's nose look dainty and petite in comparison.
At what point do engineers say "enough!" stop modifications at bulkhead XYZ and design an entirely new nose????????

I can think of two possibilities:

1. Requirements are added incrementally rather than "all from the start." In this kind of situation each change is the minimal modification from where you were after the last change and, after a few iterations, you wind up with something radically different than what you'd have had if the initial design had included everything. It works from the customer end too: they get a budget for each modification that is always enough for the current work but never enough to go back and make big changes for overall efficiency and effectiveness, even if that would be more cost-effective in the long run.
   
   
2. Designers are working around re-certification requirements, doing things in a way that is sub-optimal on purpose because the best solution requires so much regulatory work that it would sink the program (i.e., what you and Sabrejet have already gone back and forth about). Might be misremembering, but might even have been on this board that the story was a modernized helicopter was using antiquated machine guns, used no where else in the inventory, because qualifying a new machine gun for helicopter use was so time consuming and difficult that it would have caused program issues and no one wanted to accept the increased program risk and cost to deal with it.

 

[pathology_doc]

Wasn't the Spitfire II intended to incorporate as factory-built all the alterations and improvements the Spitfire I received from introduction to service? (Plus a beefier Merlin to maintain performance.)

 

[Stovepipe]

The Mk1 was probably becoming a bit past it when you compare the 109 E4. I'd say it was tied in to rationalising the accumulated changes, as you say with an eye to future Merlin power changes coming down the track, as well as further refining the airframe for mass production

 

[riggerrob]

Wasn't the Spitfire II intended to incorporate as factory-built all the alterations and improvements the Spitfire I received from introduction to service? (Plus a beefier Merlin to maintain performance.)

Mark I, A Model and 100 series are usually crude, weak and sluggish. Mark I is usually only short production run with numerous updates incorporated in the B Model.
Mark I usually retires after only a few years in service.
My question is more about the Mark VI or Mark VII that adds sensors or weapons that had not been invented yet when the Mark I first flew.

 

[Stovepipe]

Cannon, better supercharger, better armour

 

[pathology_doc]

I'd say it was tied in to rationalising the accumulated changes,

We are trying to say the same thing. You just said it better.

adds sensors or weapons that were not invented yet when they Mark I first flew.

Sensors aren't a thing for (the majority of) single-seat fighters in the Second World War - the main one was still the Mk I eyeball. (Yes, I know some American carrier fighters had a radar pod.) The question of adding sensors was one that tended to plague jets from the second generation onwards.

Weapons are of course their own special problem; we all know that the introduction of cannon to the Spitfire did not initially go smoothly, for example. The problems multiplied when missiles came along that demanded their own particular avionics. Sidewinder is a fairly easy retrofit, at least in its early forms, but I recall reading that the Australians played with the idea of fitting Firestreak to their Avon-powered Sabres, only to find that the only place they could fit the missile's support hardware was the gun bays - and while this was acceptable for a test aircraft, it was ruled as a no-go for combat. The pilots understandably and justifiably wanted to keep their guns.

 

[riggerrob]

The original question was aimed at original factories introducing multiple modifications to the same basic airframe.

For civilian airplanes, the FAA issues Supplementary Type Certificates for major modifications, for example the various STOL kits offered by Robertson, sportsman, Ultimate Wing Beaver, etc.. STC kit designers had to demonstrate no degradation of weight, balance, strength, etc. AND improved performance in at least one corner of the envelope.
At what point does an STC change so many things that the FAA requires a completely new TSO certification?

 

[Dynoman]

AC 21-40A

"However, if we find that the proposed change is so extensive that it will require a substantially complete investigation of compliance with the applicable regulations, then the applicant will have to apply for a new type certificate."

14 CFR Part 21.19

"Each person who proposes to change a product must apply for a new type certificate if the FAA finds that the proposed change in design, power, thrust, or weight is so extensive that a substantially complete investigation of compliance with the applicable regulations is required."

There does not appear to be a percentage of change in weight, performance, or another specification (or a combination of changes) that would necessitate a change from an STC to a TSO. It appears to me that the tipping point for the decision is based on the FAA's ability to ascertain whether the modification are safe or if the changes are significant enough to necessitate operating under an STC or if requires the process for a new TSO.

 

[Stovepipe]

In the case of the Spitfire, it would also involve changes to manufacturing processes in the factory and things like dimensional changes to the airframe and things like changes in riveting and bolting as the aircraft was changed. By the time it got to the Spitfire V, that was about as much could reasonably be extracted from the 1939 production standard airframe and more powerful Merlins, a taller fin, different radios, different gun layouts and so on were coming along. You can imagine the difficulty of keeping up with the drawing requirements alone.

 

[Foo Fighter]

MK V onwards was basically a different aircraft, not the only time this happened. I recall a major change about MK XX and on but would need to check facts.

 

[Stovepipe]

Bulkhead moved for heavier engine, last frame altered for bigger fin and rudder, undercarriage moved outwards, universal wing and multiple minor changes.

 

[pathology_doc]

I recall a major change about MK XX and on

Mk 21 was the variant designed from the ground up to take the two-stage, two-speed Griffon, and had a new, more rigid wing with much higher aileron reversal speed (the speed at which the ailerons cause twisting of the wing which nullifies their effect); the Mk21's aileron reversal speed was technically 850mph, much faster than ANY Spitfire could ever be expected to go. They also took the opportunity to design it for four 20mm cannon and nothing else. The rest as Stovepipe says.

It was almost called the Supermarine Victor I. Very few of them saw active service and none shot down an enemy aircraft in WW2. The stop-gap Mk 14, based on the Mk VIII airframe, was ready first and served in far greater numbers.

Mk XX is contentious, and I've read various accounts of what this airplane actually was. One says that it was the renumbered Griffon Spitfire IV, when that mark was reassigned to the PR variants. Another says that it was the F.21 with single-stage, two-speed Griffon.

Spitfire marks and their connection to the aircraft's evolution could be a book all by themselves, especially since the aircraft's service history was characterized by the lash-ups (e.g. V, IX, XII, XIV) being built and used in far greater numbers than any of the aircraft in the planned development pathway (III, IV, VIII, 21).

 

[Foo Fighter]

Thanks for the clarification, much appreciated. I for one would buy that book.

I would like to see an archeological dig on the housing estate now located where the runway for the Castle Bromwhich factory was to reclaim the tooling for these later marks. Apparently still in the big holes they dumped everything in PW.

 

[pathology_doc]

I would like to see an archeological dig on the housing estate now located where the runway for the Castle Bromwhich factory was to reclaim the tooling for these later marks.

Translation: "I would like to assemble, using the discarded tooling, the 494mph Spiteful and use it to troll some of the slower jet airliners."
Wouldn't we all, mate... wouldn't we all.

 

[pathology_doc]

I for one would buy that book.

It could be fairly short, but might easily be huge - I need to go reference my huge Spitfire book tonight, actually.

The thumbnail sketch.
Mark I - the original classic.
Mark II - rationalized rebuild with all Mark I changes, more powerful engine.
Mark III - the first true attempt to make alterations/improvements to the airframe, including Merlin XX engine, cannon wing. Not proceeded with.
Mark IV - Initially the first Griffon Spitfire, one example; later reassigned to unarmed PR variants.
Mark V - Evolved Mark II offshoot to get Merlin 40-series engines into the air quickly.
Mark VI - First pressurized variant, single-stage, single-speed Merlin optimized for higher altitudes.
Mark VII - Pressurized airframe development with two-stage, two-speed Merlin.
Mark VIII - Unpressurized equivalent of VII. Supposed to be the definitive Merlin version, but eclipsed in production and service by:
Mark IX - Mark V-derived lash-up with Merlin 60-series engine.
Mark X - Essentially a PR version of Mark VII, limited numbers built
Mark XI - Essentially a PR version of Mark IX
Mark XII - First Griffon variant (single-stage two-speed), derived from Vc.
Mark XIII - Limited production armed PR with cannon removed and Merlin 32 for low level recon.
Mark XIV - First two-stage Griffon variant, derived from Mark VIII airframe.
Mark XV - Not used for Spitfires; was used for first, Mark XII-like Griffon Seafire.
Mark XVI - Designation for Mark IX with Packard Merlin, many had a bubble canopy.
Mark XVII - Not used for Spitfires, was used for a developed Seafire XV with bubble canopy.
Mark XVIII - Rationalized Mark XIV with bubble canopy, slightly strengthened airframe.
Mark XIX - PR variant of Mark XIV.
Mark XX - See above; seems to be the Mark IV renumbered.
Mark 21 - Definitive improved airframe with stronger, 4-cannon wing for Griffon 60-series.
Mark 22 - Bubble canopy variant of 21.
Mark 23 - F.21 variant with high-speed wing section, not proceeded with.
Mark 24 - Basically a 22 with a different electrical system.

 

[Foo Fighter]

Much appreciated. Thank you.

 

[Stovepipe]

where do the two-seaters fit in? are they all IX airframes?

 

[pathology_doc]

where do the two-seaters fit in? are they all IX airframes?

I think some of them are VIII and some are IX. I vaguely recall the Russians doing a local modification to a V, but I couldn't quote you a reference.

 

[steelpillow]

When do too many minor changes require a completely new design? They don't. A completely new design is needed only when minor changes, however cumulative, can no longer deliver on the latest requirement. A case in point is the Spitfire. It ran through 24 marks, many with minor changes to the wing design. But by then the wing had reached its limit and to go any further a clean-sheet design was necessary; this became the Spiteful.

When do they need a complete rebranding? Ah, that is a different question with a different answer; when it is politically expedient. One should not confuse the two (though it is often politically expedient to do so).

 

[steelpillow]

In the case of the Spitfire, ... You can imagine the difficulty of keeping up with the drawing requirements alone.

One way they managed this was for different factories to stick to different variants, and sometimes even to do the detail design and development work. Most of the Marks manufactured at Castle Bromwich and Westland were not made anywhere else.

Mk 21 was the variant designed from the ground up to take the two-stage, two-speed Griffon, and had a new, more rigid wing ...
It was almost called the Supermarine Victor I.

This wing was still an incremental change from the previous wing. It still fitted the same fuselage attachment points with the same main spar and leading-edge torsion box, and housed the same ancillaries in the same places. There was still a high degree of parts commonality. But it was the end of the line. The wing that really was designed from the ground up became the Spiteful.

Spitfire marks and their connection to the aircraft's evolution could be a book all by themselves

I have several such books

 

[pathology_doc]

I have several such books

Can you give me a few titles so that I can begin the process of brutalizing my wallet?
(I can't guarantee that I don't already have a few.)

 

[steelpillow]

Just sticking to the ones on my shelf. I cannot speak for later editions. You can find plenty more referenced at the end of the various Wikipedia articles on the Spit and its derivatives. Most also cover operations to at least some extent, but that's life:

• Alfred Price; The Spitfire Story, Arms and Armour Press 1986. (Previously published by Jane's). Probably the definitive classic.
• Alfred Price; Spitfire: A Complete Fighting History, Promotional Reprint Co. 1975. Compendium of the two-volume Spitfire at War previously published by Ian Allan. Don't be put off by the title. This is really a collection of essays by many famous names; it is more edited than written by Price. It covers both development and operations from all sorts of angles that you normally never get to understand. Not so much a book detailing the marks, as a book infused by them and their reasons for being.
• G.N.M. Gingell (Ed.); The Supermarine Spitfire - 40 Years On, Royal Aeronautical Society, 1976. More essays, these by those involved in its development. Includes comprehensive comparison tables between all the different Marks.
• Spitfire, Aeroplane Icons (Bookazine), Kelsey Media, 2013.

 

[T. A. Gardner]

I'd say it was tied in to rationalising the accumulated changes,

We are trying to say the same thing. You just said it better.

adds sensors or weapons that were not invented yet when they Mark I first flew.

Sensors aren't a thing for (the majority of) single-seat fighters in the Second World War - the main one was still the Mk I eyeball. (Yes, I know some American carrier fighters had a radar pod.) The question of adding sensors was one that tended to plague jets from the second generation onwards.

By late WW 2 most US single seat fighters had an APS-13 tail warning radar set fitted.



Another common sensor, albeit passive, was the gyro gunsights fitted to most US aircraft late war.

 

[Mark Nankivil]

When do too many minor changes require a completely new design? They don't. A completely new design is needed only when minor changes, however cumulative, can no longer deliver on the latest requirement. A case in point is the Spitfire. It ran through 24 marks, many with minor changes to the wing design. But by then the wing had reached its limit and to go any further a clean-sheet design was necessary; this became the Spiteful.

When do they need a complete rebranding? Ah, that is a different question with a different answer; when it is politically expedient. One should not confuse the two (though it is often politically expedient to do so).

The US Navy in the '50s was good at that - F9F-9 had nothing to do with any previous F9Fs, finally becaome F11F. F4D-2 later become F5D-1, etc... USAF pulled that stunt with the F-84 too and kept it.

 

[T. A. Gardner]

When do too many minor changes require a completely new design? They don't. A completely new design is needed only when minor changes, however cumulative, can no longer deliver on the latest requirement. A case in point is the Spitfire. It ran through 24 marks, many with minor changes to the wing design. But by then the wing had reached its limit and to go any further a clean-sheet design was necessary; this became the Spiteful.

When do they need a complete rebranding? Ah, that is a different question with a different answer; when it is politically expedient. One should not confuse the two (though it is often politically expedient to do so).

The US Navy in the '50s was good at that - F9F-9 had nothing to do with any previous F9Fs, finally becaome F11F. F4D-2 later become F5D-1, etc... USAF pulled that stunt with the F-84 too and kept it.

That was often because of how Congress was funding projects. If new programs were not being funded, but funding was available for existing programs, then the various services played a name game with equipment to keep funding coming. The FJ Fury is another example of this with the Navy for example.

 

[steelpillow]

That was often because of how Congress was funding projects. If new programs were not being funded, but funding was available for existing programs, then the various services played a name game with equipment to keep funding coming.

Yet another great British invention exported to the USA. Back in 1910 the Army Balloon factory at Farnborough took on Geoffrey de Havilland and his homebuilt plane. He was given little funding for new planes but could scrounge extra for unscheduled maintenance, such as rebuilding after a minor accident or to make improvements. The majority of his early designs were therefore passed off as rework of old wrecks.

 

[Schneiderman]

That was often because of how Congress was funding projects. If new programs were not being funded, but funding was available for existing programs, then the various services played a name game with equipment to keep funding coming.

Yet another great British invention exported to the USA. Back in 1910 the Army Balloon factory at Farnborough took on Geoffrey de Havilland and his homebuilt plane. He was given little funding for new planes but could scrounge extra for unscheduled maintenance, such as rebuilding after a minor accident or to make improvements. The majority of his early designs were therefore passed off as rework of old wrecks.

Another ploy was to refer to experimental aircraft as mailplanes to get them funded as civil projects rather than military. The Fairey long range monoplane is one example, named as Postal Monoplane on the early blueprints.

 

[Rhinocrates]

For what it's worth, this recalls the old problem of the Ship of Theseus. It was even referenced in WandaVision. Essentially,

If the ship of Theseus were kept in a harbor and every part on the ship were replaced one at a time, would it then be a new ship?

Ship of Theseus - Wikipedia

en.wikipedia.org

Take that a step further and suppose that each new timber was slightly different from the one it replaced.

Ask Porsche about the 911 or Land Rover about the Defender.

On the other hand, consider the Ise Grand Shrine.

Ise Grand Shrine - Wikipedia

en.wikipedia.org

Every twenty years the shrine is destroyed and rebuilt and because it is embedded within a theology that emphasizes renewal, it's considered to be the same object. The issue then seems to be purely metaphysical, and that's pretty arbitrary.

 

[T. A. Gardner]

The US Navy in the late 1800's did something like that with several Civil War era monitors. They "reconstructed" / "repaired" them because Congress wouldn't fund new ships. What happened in reality was these ships were entirely new construction by the time they rejoined the fleet. These were ships like the Amphitrite and Puritan and the "repairs" took more than a decade to complete.

 

[riggerrob]

Sometimes the definition of "new" versus "repaired" can be driven by import duties.
Back during the 1970s (?) Mexican import duties were so high that it was cheaper to ship a crashed Cessna 180 back to the Cessna factory in Kansas, for "repairs." Since the crash had rolled the airframe into a ball of scrap aluminum, the Cessna repair shop installed dozens of new parts (bulkheads, stringers, skins, etc.). When you have all the patterns and tools, it is often easier to pick a new piece off the (production) shelf than to rivet a patch onto a cracked old piece. When it returned to Mexico - 90 percent new by weight - the owner only had to pay minor import duties ... way less expensive than buying a new-production Cessna 180.

I used to sew in new flaps - rather than patch the old flap - when I worked at Rigging Innovations (parachute harness/container factory). Since all my repairs had to pass a Technical Standard Order inspector - the same as new production - it was often easier to rebuild to like-new standard.

 

[publiusr]

For an aside, we can continue the debate about when Boeing made too many minor changes to their 737 Max.

Had that been a tri-jet...heavier wing mount engines moved a bit forward could have been counter balanced by a longer tail and more seats, right?

 

[T. A. Gardner]

Puritan as originally designed and started:



Puritan (BM-1) as completed after the "repair."

 

[Michel Van]

Another case is the US Navy Douglas A-3 Skywarrior conversion into USAF B-66 Destroyer
Allot modification and radical changes were made here.

 

[riggerrob]

Have you tried to count all the antennas on a Beechcraft King Air modified for the US Army electronic warfare role? They sprouted antennas from the top of the fuselage, from the bottom of the fuselage, from tip tanks, etc.
It looks like a Russian bear that bit a Ukrainian porcupine in the @$$!
Hah! Hah!
Apparently some even carried complete "spoof" cell phone networks.
The Taliban soon learned to ban cell-phones from meetings.

 

[publiusr]

The Red Lectroid look.