Model Behaviour: PROSIT takes simulation for a spin
Spinning faster; operating hotter. That's the mantra for tomorrow's more efficient jet engines, driven by the push towards more environmentally benign and energy-efficient aircraft that emit less CO2, NOx, and noise.
‘Jet engine turbine disks operate better at higher temperatures where combustion is more efficient and less fuel is burned, so you're not wasting kerosene,’ explains Aleksandar Stanojevic, Head of Innovation Management at Austrian forging specialists voestalpine BÖHLER Aerospace, and coordinator of Clean Sky's PROSIT project.
‘And the higher the rotation speed of the engine disks, you also get more efficiency. What this means is the disk material has to be able to withstand higher and higher temperatures as well as higher loading forces.’
The disk material has to be able to withstand higher and higher temperatures as well as higher loading forces
Turbine disks operate in extreme conditions, subject to a constant stream of high velocity hot gas (turbine disks are located aft of the engine's combustion chamber, an area where there are high temperatures). To address these challenging demands, engine designers need to have a more granular level of understanding of the performance limits that materials can deliver, as well as methodologies to ensure that the chain of material production steps, from the most raw and elementary state to the finished component, is optimised for uniform and predictable quality.
Clean Sky's PROSIT project aimed to develop a ‘simulation chain model’ that would replace the current time-consuming and costly method of testing whether a specific turbine disk design fits well with a particular metal. A major goal of PROSIT was also to simulate the inherent material properties to safely design turbine disks at the minimum thickness and mass needed to save weight in the aircraft – potentially cutting fuel burn and emissions, in line with the EU’s environmental objectives.
Examining the microstructure
The project investigated the microstructure and material properties of Direct Aged Alloy 718 (also referred to by its trade name, Inconel), the 'go to' high-temperature material used in the turbine disks of modern jet engines. Armed with insights gleaned from simulation it became possible to accurately predict the compatibility of DA718 with any new turbine disk design.
'If you look at the material under a microscope, you'll see its microstructure. What's wanted is a fine microstructure, because if you don't carry out the design correctly, you end up with a mixture of fine and coarse grained microstructure,’ explains Aleksandar Stanojevic, Head of Innovation Management at voestalpine BÖHLER Aerospace.
This mix of different sized grains – also addressed as non-uniform grain structures – will reduce the performance and longevity of the turbine disk. Currently used microstructure models are not capable of predicting non-uniform grain structures, so that the only way to produce a disk with homogenous microstructure is to manufacture it and then use destructive testing to examine the material. This procedure takes several loops of redesign and manufacture to reach the optimal design. The process is lengthy and costly.
With the PROSIT simulation model, the idea was to develop an algorithm that can be used in finite element analysis modelling. This would enable accurate prediction of whether a design would be optimal, given the known microstructure characteristics as applied to the particular design.
The project investigated the effects of local non-uniform grain structures on the mechanical properties of the material when used in a turbine disk – it's essential to be able to guarantee that the mechanical properties on the whole disk are equally distributed. PROSIT also examined how these are influenced by the cooling down procedures associated with the heat treatment.
‘The PROSIT consortium developed a model which feeds into the company's manufacturing reality and processes. Having this model helps engine designers define the real characteristics and the actual properties of the material, and then it helps the designers decide whether to use this material for the disks,’ says Clean Sky project officer Rosario Trillo Rivas, who adds that the project context is Clean Sky's Engines initiative – ‘the result of the project feeds MTU's demonstrator, the Advanced Gear Engine Configuration.’
The beauty of the simulation chain model is that it can predict material imperfections, and that information can be used to ensure that the areas that are more prone to failure, if known, can be designed in such a way within the forging so that the weaker parts of the forging are machined away, leaving the final disk free from defects.
Having this model helps engine designers define the real characteristics and the actual properties of the material, and then it helps the designers decide whether to use this material for the disks
The engine maker's perspective
Roland Schmier of MTU Aero Engines says that ‘twenty years ago turbine disks weighed as much as 25% more than they do today, and during that time we've done a lot of calculations to save weight.’
‘Now it's necessary to know exactly the mechanical behaviour and strengths of the material for the calculation of the lifetime, because the turbine disk is a life-limited part of the engine, which is allowed a certain number of cycles to ensure that it does not fail during its guaranteed lifetime.’
‘If we can calculate what is possible through the simulation, then we can instruct the forging company to produce a certain quality of disk,’ explains Schmier, ‘and the idea is that you can therefore have more control over the design through having a deeper understanding of the material.’
You can have more control over the design through having a deeper understanding of the material
Outcomes and potential next steps
The PROSIT project elicited the development of two material models that can calculate the microstructure’s evolution during the complex forging process of a turbine disk.
One model is able to pinpoint quantitatively those regions where non-uniform microstructure might occur. This is a rapid method of obtaining a first overview of the microstructure of the whole forging and can be used during a high number of finite element (FEM) simulations that are performed during the design process of a disk.
The second model can accurately predict the specific, local properties of a forging. With the use of this model, an exact quantification of the final microstructure (e.g. grain sizes and their fraction) and the corresponding mechanical properties are possible. Therefore, the production route of critical parts can be improved and forgings will be even safer by reducing the occurrence of substandard microstructures.
Both models were implemented successfully into the simulation chain and are already in use following validation. The validation was performed by analysing the microstructure and mechanical properties of a number of forged ‘pancakes’ and turbine disks.
‘One of the most important aspects when you're developing a simulation tool is to ensure that the tool really is accurate with respect to the physical testing results,’ says Clean Sky's Trillo Rivas. ‘The way to validate that accuracy is to develop the model and then compare the results of the generated simulations with data extracted from physical experiments. When they're equal that confirms that the tool accurately mimics reality’, says Clean Sky's Trillo Rivas.
Indeed, results obtained from these tests aligned with the predicted characteristics identified via the model simulations, confirming that the objectives of the project had been met.
The potential for further exploitation of the PROSIT project extends beyond its application with Inconel, according to the project's coordinator.
‘The specific solution we have developed in this project is for use with Alloy 718, but the thought processes and all the learnings we've had can, with certain adjustments, be applicable for other nickel based alloys,’ says Stanojevic.