At its heart, a jet engine operates on a remarkably simple principle. Air is drawn in, compressed, mixed with fuel and ignited, with the hot expanding gases producing thrust. (A process memorably and succinctly described by one wag many years ago as "Suck — Squeeze — Bang — Blow".)
However, wringing the greatest possible performance out of such a powerplant is anything but simple. In fact, it is one of the most complex undertakings in the manufacturing industry.
Every year, millions of manhours and billions of dollars are poured into improving turbofans, the type of jet engine used in airliners; the price of fuel and the vast amounts of Jet A-1 consumed by the world’s airlines mean that even a one per cent improvement in fuel burn is regarded as a worthwhile advance. (In more recent times, the concomitant reduction in the emission of greenhouse gases has also become a major factor.)
Get the formula right in making an engine more efficient and huge sales can be forthcoming. Both Boeing and Airbus have backlogs of several thousand aircraft on their order books, with new models of the fast-selling 737 and A320 families due to appear in the next few years and extending the production runs of the types well into the 2020s. Many modern airliners can be produced with a choice of engines, depending on customer preference. If your engine is a few per cent more efficient than your competitor’s, major orders are likely to flow your way.
And, with the air transportation sector predicted to grow at three to four per cent annually for the foreseeable future, it’s a market that will bring successful engine manufacturers (and their suppliers, such as GKN Aerospace Engine Systems) an enduring revenue stream.
Simulation as a critical path
Since the 1960s, new airliners have shown steady improvements in reduction of fuel burn, emissions and noise. In recent years this positive performance trend has started to flatten, requiring more and more effort to maintain progress.
Simulation is an increasingly important tool being used by GKN Aerospace in this constant process of refining the performance of jet engines. By using simulation as a critical path in developing jet engine components, timescales — and thus costs — can be trimmed, to allow the best possible use of personnel and machinery in delivering products to a customer.
Allied with this, the consistently high price of fuel means that the requirement for lighter, and thus more fuel-efficient, aircraft is greater than ever; lightweight technology has a key role to play in cutting fuel burn and simulation is increasingly important in developing lighter components.
Making the greatest use of simulation, says Henrik Runnemalm, director of research and technology at GKN Aerospace, means running design and manufacturing simulation in a closed loop process. Design simulation involves the creation of the proposed components "virtually"; manufacturing simulation tools are used to predict optimised factory logistics, machine tool and robot movements, component deformations and specific manufacturing process physics.
Results from the manufacturing simulations are then fed back into the design process. “This ‘loop’ is what we’re trying to build up here,” he says. “You don’t want to find it’s not possible to produce a part, or that it’s creating too-high stress levels which could force the part in-service time to be reduced.” Finding a balance between the technical ideal, producibility and cost is vital.
“The key here is that the design capability simulation needs to be totally integrated with the manufacturing side, which also has its part to play in simulating what’s happening with the ‘lifing’ of the product” — that is, the length of time a component can survive in the engine before being removed for maintenance or replacement.
“When we design a part that is fabricated by welding, then the stresses and deformation created during manufacturing are actually part of the lifing,” explains Runnemalm. Simulation of these manufacturing stresses can reliably predict the effect of manufacturing processes such as welding on a product. Welding, for example, can result in unwanted deformation and stresses within the material.
“If you’re trying to design an apple and don’t include all the manufacturing stresses you tend to end up with a pear,” is how he describes it.
Linking design and manufacturing
In producing any new engine component, manufacturing simulations link design and manufacturing during product development and act as a tool for designers and manufacturing engineers to evaluate different concepts or manufacturing processes.
Runnemalm divides the design aspect of the product development process into three stages — concept design, preliminary design and detailed design. Similarly, he divides the manufacturing part of product development into three sections that track their design counterparts — inventory of known methods, preliminary preparation and detailed preparation.
Simulation is used to help with some of the most basic aspects of development, such as aerodynamic strength and vibration dynamics, before heading into manufacturing territory through simulation of processes such as machining or heat treatment.
Stresses on components being welded can be reduced by careful sequencing of individual welds. Such sequencing does not always follow the pattern that might be expected. Once that sequence has been determined, simulation is also used to programme the welding robot to perform the necessary manoeuvres to follow that sequence.
On a legacy product, such as the turbine exhaust case (TEC) of the PW2000, which powers the 757, the aim is to optimise existing processes such as the weld sequence, says Runnemalm.
The TEC requires about 200 welds and, at one point in the production of the engine, problems arose with geometrical tolerances in the engine.
Tolerances between components in a modern turbofan are very tight and meeting these tolerance criteria can be difficult because of internal stresses created in the component by processes such as heat treatment. Simulation can identify the best changes in the production process to improve those tolerances.
In the case of the TEC, several welding sequence concepts were investigated to meet these tolerances. Welding simulations showed that residual stresses could be lowered by using a different welding sequence. Moreover, simulations also concluded that to avoid problems with tolerances, a pre-deformation should be given to the product before welding.
On the GEnx, the Dreamliner engine that was created from a "blank sheet of paper" design, and for which GKN Aerospace manufactures the turbine rear frame, “We use the tool to say ‘We need a weld in this position because it’s creating less stress’.”
Simulation can also be used as an investigative tool and allows examination of welding sequences that had previously been too complex and costly to explore.
But does simulation always provide the correct answers? “That’s really the key of GKN Aerospace’s capability,” says Runnemalm. “We’ve been working really hard proving that our simulation tools are giving us the right answer, so we can trust them.”
Simulation can bring its own problems. In the virtual world, edges of components can be designed to be infinitely sharp. In practice, however, there has to be a balance between that aerodynamic ideal and the limitations of the manufacturing process.
To support its work, GKN Aerospace collaborates with a group of universities and other institutions in fields such as aero performance, solid mechanics and material characterisation and model building. Some partners carry out specialised manufacturing-related work — for example, how a manufacturing process is described to a computer in terms that it can understand and act upon.
Today, more than ever, time is money. Companies have to focus on getting it right first time, both when producing a new product, or modifying an existing one to work more efficiently.
Decreasing costs, time and risk by increasing the information available about a product and its manufacturing processes will help a company achieve a better market position and improve its competitiveness.