4 ms·

1. Dual cockpits were a result of slapping a couple 51 bodies together, already mostly engineered. 2. Kept it because it worked for longer missions using dual

1. Dual cockpits were a result of slapping a couple 51 bodies together, already mostly engineered. 2. Kept it because it worked for longer missions using dual controls. 3. Saves money in production, tooling, and engineering to just use existing designs where possible. Same parts are already set up, just run the process another time. 4. Symmetry man.

VLM 9y ago
There's an interesting CS analogy in that aircraft fuselage are extremely tightly coupled making changes difficult. Imagine making an existing plane a foot longer. Well that means you'll need a slightly larger wire to carry the current for the landing lights over a longer distance so thats heavier which requires a larger hole thru the bulkhead which is now too light so you put a strengthening plate on it which makes the auxiliary oil cooler not fit, meanwhile its now too heavy for the original landing gear to handle the weight ... eventually 100000 engineer hours later you have a slightly larger plane. It would have been cheaper to just design a new plane from the ground up... Wings are not as tightly coupled as fuselage systems. Almost plastic model like, you can chop off the ends of two planes wings and stick them together. Its not quite that simple, but at least it doesn't involve redesigning the oil cooling system. Some software is extremely tightly coupled like an aircraft fuselage, some less tightly coupled like an aircraft wing.
- rwmj 9y ago
  Not to disagree with your main point that there's a lot of tight coupling ... but planes are regularly stretched[1][2] so it seems like it's still a lot simpler than designing a whole new plane from scratch. Even new designs are really evolutions of old planes. [1] https://en.wikipedia.org/wiki/Airbus_A340#Variants https://en.wikipedia.org/wiki/Airbus_A340#Variants [2] https://en.wikipedia.org/wiki/Boeing_747#Variants https://en.wikipedia.org/wiki/Boeing_747#Variants
  - dunmalg 9y ago
    Probably a better example from the opposite end of the spectrum would be the design changes between the F/A-18C/D and the F/A-18E/F. The initial plans were to stretch the C/D model to add fuel and improve the abysmal range, but this increased weight which required larger wings which required all sorts of major design changes. In the end, the E/F ended up with only 40% parts commonality, and was now a medium weigh fighter like the F-15, rather than the cheap light fighter like the F-16 it began life as. Engineers I know who worked on the E/F design have told me it would arguably have been cheaper to design a new plane from scratch, but the Navy was so afraid of another failed project that they were only able to sell it to the Navy as an "upgrade" of a plane design they already had. Dubious in an engineering sense, but absolute genius in marketing.
  - geezerjay 9y ago
    > but planes are regularly stretched[1][2] One of the main design requirements for those plane models was, right from the start, to support multiple variants by adopting a modular design. Yet, they are still very tightly coupled. In fact, IIRC the Airbus A340 project is widely known for having experienced a billion-euro delay due to production problems caused by CAD software issues. In short, if my memory doesn't fail me, some design teams used Catia V4 while others migrated to Catia V5, whose CAD engine was rewritten from scratch. Consequently, some slight differences in how the different versions handled geometry definitions resulted in some electric cables designed by some engineering teams to be incompatible with the structural layout designed by other engineering teams. We're talking about differences in centimeters in a plane with a wingspan with around 56m.
- willglynn 9y ago
  An F-82 fuselage is literally a longer P-51H fuselage: > North American Design Chief Edgar Schmued incorporated two P-51H Mustang fuselages lengthened by the addition of a 57 in (145 cm) fuselage plug located behind the cockpit where additional fuel tanks and equipment could be installed.
- mandevil 9y ago
  Yes, things are coupled, but they are still malleable. The F-82 fuselage was longer than a normal P-51 fuselage by almost 5' (57 inches). They stretched the section behind the cockpit, to give it more room for fuel tanks.
tragomaskhalos 9y ago
Meh, symmetry is an engineering cop-out; much more fun is https://en.m.wikipedia.org/wiki/Blohm_%26_Voss_BV_141 https://en.m.wikipedia.org/wiki/Blohm_%26_Voss_BV_141 :)
- robin_reala 9y ago
  See also the Rutan Boomerang: https://en.wikipedia.org/wiki/Rutan_Boomerang https://en.wikipedia.org/wiki/Rutan_Boomerang
- mandevil 9y ago
  The Bf-109Z, which never flew, was the same kind of idea as the F-82 but only had one cockpit (on the left side), for that same sort of asymmetry.
rdiddly 9y ago
So it's a hack, basically. An improvised solution that exploits the shortest path to getting something done rather than taking extra time to do it the "right" way.

VLM 9y ago

There's an interesting CS analogy in that aircraft fuselage are extremely tightly coupled making changes difficult. Imagine making an existing plane a foot longer. Well that means you'll need a slightly larger wire to carry the current for the landing lights over a longer distance so thats heavier which requires a larger hole thru the bulkhead which is now too light so you put a strengthening plate on it which makes the auxiliary oil cooler not fit, meanwhile its now too heavy for the original landing gear to handle the weight ... eventually 100000 engineer hours later you have a slightly larger plane. It would have been cheaper to just design a new plane from the ground up... Wings are not as tightly coupled as fuselage systems. Almost plastic model like, you can chop off the ends of two planes wings and stick them together. Its not quite that simple, but at least it doesn't involve redesigning the oil cooling system. Some software is extremely tightly coupled like an aircraft fuselage, some less tightly coupled like an aircraft wing.

rwmj 9y ago
Not to disagree with your main point that there's a lot of tight coupling ... but planes are regularly stretched[1][2] so it seems like it's still a lot simpler than designing a whole new plane from scratch. Even new designs are really evolutions of old planes. [1] https://en.wikipedia.org/wiki/Airbus_A340#Variants https://en.wikipedia.org/wiki/Airbus_A340#Variants [2] https://en.wikipedia.org/wiki/Boeing_747#Variants https://en.wikipedia.org/wiki/Boeing_747#Variants
- dunmalg 9y ago
  Probably a better example from the opposite end of the spectrum would be the design changes between the F/A-18C/D and the F/A-18E/F. The initial plans were to stretch the C/D model to add fuel and improve the abysmal range, but this increased weight which required larger wings which required all sorts of major design changes. In the end, the E/F ended up with only 40% parts commonality, and was now a medium weigh fighter like the F-15, rather than the cheap light fighter like the F-16 it began life as. Engineers I know who worked on the E/F design have told me it would arguably have been cheaper to design a new plane from scratch, but the Navy was so afraid of another failed project that they were only able to sell it to the Navy as an "upgrade" of a plane design they already had. Dubious in an engineering sense, but absolute genius in marketing.
- geezerjay 9y ago
  > but planes are regularly stretched[1][2] One of the main design requirements for those plane models was, right from the start, to support multiple variants by adopting a modular design. Yet, they are still very tightly coupled. In fact, IIRC the Airbus A340 project is widely known for having experienced a billion-euro delay due to production problems caused by CAD software issues. In short, if my memory doesn't fail me, some design teams used Catia V4 while others migrated to Catia V5, whose CAD engine was rewritten from scratch. Consequently, some slight differences in how the different versions handled geometry definitions resulted in some electric cables designed by some engineering teams to be incompatible with the structural layout designed by other engineering teams. We're talking about differences in centimeters in a plane with a wingspan with around 56m.
willglynn 9y ago
An F-82 fuselage is literally a longer P-51H fuselage: > North American Design Chief Edgar Schmued incorporated two P-51H Mustang fuselages lengthened by the addition of a 57 in (145 cm) fuselage plug located behind the cockpit where additional fuel tanks and equipment could be installed.
mandevil 9y ago
Yes, things are coupled, but they are still malleable. The F-82 fuselage was longer than a normal P-51 fuselage by almost 5' (57 inches). They stretched the section behind the cockpit, to give it more room for fuel tanks.

tragomaskhalos 9y ago

Meh, symmetry is an engineering cop-out; much more fun is https://en.m.wikipedia.org/wiki/Blohm_%26_Voss_BV_141 https://en.m.wikipedia.org/wiki/Blohm_%26_Voss_BV_141 :)

robin_reala 9y ago
See also the Rutan Boomerang: https://en.wikipedia.org/wiki/Rutan_Boomerang https://en.wikipedia.org/wiki/Rutan_Boomerang
mandevil 9y ago
The Bf-109Z, which never flew, was the same kind of idea as the F-82 but only had one cockpit (on the left side), for that same sort of asymmetry.

rdiddly 9y ago

So it's a hack, basically. An improvised solution that exploits the shortest path to getting something done rather than taking extra time to do it the "right" way.