September 8, 2023, ©. Leeham News: We described the Preliminary design phase of an airliner development program over the last weeks. Now our project is transitioning into Detailed design.
It’s the most challenging part of the project as we now go from perhaps a thousand people involved at the OEM into tens of thousands and even more people at consultancies and suppliers.
The Preliminary design phase sets the project up for its most demanding phase, Detailed design.
We wrote that an airliner typically has three to four million parts. While many of these are standard parts or minor variations thereof (rivets, fasteners, bolts, nuts, pipes, connectors, cables, chips and components in systems and avionics..) a million at least are custom parts. All these have to be designed to the precise requirements of the aircraft.
This is a HUGHE task, that engages not only most of the engineers of an OEM but the OEM will also contract consultancy companies to do part of the OEM design tasks. A large portion of the OEM’s organization will prepare the contracts, information flow, and rules on how to manage sub-suppliers and how to involve and govern these in the project.
We wrote about the way a Preliminary design phase can be more precise in the definition of an aircraft and leave a minimum to interpretation. Mathematical models as specifications of functions are preferable over written specifications. Ideally, large parts of the aircraft exist in interacting mathematical models in a so-called Digital Twin.
Equally important is that the state of the overall design of the aircraft stays fixed. Let’s take an example;
The increase of the authority of the aileron could stem from the aircraft mass increasing and thus its inertia in the roll axis but also from a changed requirement for allowable crosswind during hot and high takeoff and landings.
Whatever the cause, it will send a shockwave of changes through the organization, which in turn increases the mass, which in turn increases the engine thrust demand, which in turn increases the engine size, which in turn beefs up the pylon, which in turn increases the powerplant stress on the wing, which in turn increases the wing mass, which in turn….
This chain reaction is called the aircraft project’s spiral factor. An increase of one kilo of mass somewhere leads to an increase of the aircraft’s empty weight by two kilos, which can result in an increase of the Maximum Takeoff Weight by three kilos to keep the range unchanged.
Each project and project part has its spiral factors; the above is made for illustration. I have been in projects with spiral factors above three.
The above illustrates a couple of things around Detailed design:
There are more things that need discussion around Detailed design. We will continue the discussion next week and then also discuss what changes in the traditional way Detailed design has been made could lead to faster and more efficient projects.
It is very hard and very important to freeze the interfaces between systems and teams. From then on they cannot change the interface between teams and they have to stay within geometric, mass, power, software interface and heat boundaries to finalize their design with its performance and cost.
Don’t forget the critical step in the process where before moving from PD to detail design company executives decide its better not to invest in a new airplane and spend that money on stock buybacks instead.
Thank you for this!
It has to be a fantastic, to be part of a team, that designs a new airliner
Hard work, creativity, ingenuity, vision, imagination, satisfaction…
Helping to create something complex, and useful!
Yes, this is a wonderful look “under the hood”.
Thanks, Mr. Fehrm.
Well presented!
One aspect I am pondering, that being there is a system of blue prints that would be between digital and written (unless blue prints are intended as written?)
I rue the day that we did not move into digital, blue prints were good ref for our post build work, but they got torn up, lost and even the so call As Built were often wrong.
I corrected a lot of controls prints for one builing where I got the sepia, but building prints were not accessible to me.
An aspect that does get involved is figuring out ways to do the same job better, you can balance things or even save weight.
That of course can go to far, the case of the wing join on the 787 where (if my memory is right) they shaved off structure to save weight and went too far.
Each mfg balances the factors out as best they can obviously.
The build difference between the 787 and A350 would be an outstanding example of two different approaches that work (spun fuselage vs frame and panels).
How can a aircraft manufacturer, design a new Airliner, without knowing how the future is going to unfold? There are so many unknowns!
Is instantaneous Satellite communication the next step?
Will Satellite communication enable global connectivity?
Are Infrared sensors going to be installed in all new Aircrafts?
Will advanced Lidar sensors, map the terrain, obstacles, and weather?
Its like anything, you do the best with what you have and none of that affects a core design and if its a future tech you can leave add in capability.
The ICE is going to be with us for a long time as are aircraft designed now or a latter.
Very much like ADSB, you add it to older aircraft as needed. Not clean and tidy but its not a clean tidy world.
You replace an old aircraft family doing the same routes at the same speed but consuming less fuel, give more comforft and safety and generate less noice. Then you hope the market stays the same and don’t get too late to market to experience the A380-800 vs. 777-300ER situation where the 777-300ER had more modern engines, good cargo and 2 engines with ETOPS rules changed to allow more non-stop flights.
Claes:
Well put.
There was huge skepticism on the A380 and clearly it was justified. But that bird was a we want to prove we are superior to Boeing not good market analysis.
The A340 made more sense but with a 777 competitor it was doomed.
The 787 clearl6y was based on excellent market analysis as well as tech and only failed in management execution (Round 3 or 4 now).
“The Operation was a complete success; however, the Patient died.”
Airbus got the A380 x % better than the 747-400 but late and not the SYD-LHR non stop range it was initially designed for. The 777-300ER became much better than most thought and killed the 747-400/-8 as well. 747-400 was produced 694 ea, the 777-300ER has been made around 837ea. It will be interesting to see the RR Ultrafan once it is flying on its A380 test aircraft, maybe that is the engine it should have needed from the start?