Aircraft Structures – Project Course (TMHL26)General information

This course is aimed at developing skills for product development and/or computational analysis relevant to aircraft structures. Previously acquired skills are consolidated when applied to problems of interest to the aircraft industry. After the course the student should:

be able to apply previously acquired skills in the mechanics of structures to integrated

problems with relevance to the aircraft industry. have consolidated their knowledge in engineering science.

be able to formulate, implement and update a project plan.

have experience from working in a team with product development related to aircraft

design.

The course is carried out as a project with regular project meetings. The result from the project is a written report and an oral presentation.

The course is based on a project that is inspired by a real problem in the aircraft industry. Thestudents work with modeling, design and analysis of aircraft structures. There are also activities in project-specific technology.

Work flow

A group of up to 6 participating members is formed and a supervisor is assigned to the group.The group members write a project plan, which is approved by the supervisor. This project plan includes the objectives of the project, the tasks needed to achieve these objectives, the dissemination activities (oral presentation and written report), and the division of labor between group members. The project plan includes a time-line specifying when each activity is to be initiated and finished. All group members are required to actively contribute to the oral presentation and the written report. All group members are required to attend the oral presentations of other groups following the same course.

Project descriptionBackground

Both during flight and on the ground, the engine of a jet aircraft produces heat. A cooling system is required for maintaining sufficiently low operating temperature for various components within the aircraft. To minimize weight, it is possible to use oil and fuel as coolants. These fluids, in turn, are air-cooled through heat exchangers, as schematically depicted in Fig. 1.

Figure 1: Air-fuel and air-oil heat exchangers for a generic jet aircraft.

During flight, a pressure drop develops between the air intake and the air outlet owing to the velocity of the aircraft. This pressure drives cool air through the heat exchangers to provide a sufficient heat sink. Before take-off, however, it is necessary to pump air across the heat exchangers. This is achieved by the use of a jet ejector pump on the outlet side of the heat-exchangers: hot, compressed gas from the engine is pumped through a nozzle into the air outlet, thus creating suction that drives air across the heat exchangers.

In this project, we will consider one particular realization of the ejector assembly of a jet aircraft. The heat exchangers have already been designed, so that their geometry is fixed. Thesame is true for the air outlet with the ejector pump. It remains to design the air duct connecting the heat exchangers to the ejector pump. The interior geometry of this duct has already been designed to accommodate the predicted air flow, and should thus not be altered. Similarly, the geometry of the inlet and outlet of the duct is already designed to fit the ejector pump and heat exchanger, respectively.

The air duct is subjected to inertial loads, mainly arising from the maneuvering of the aircraft.From a structural point of view, strength and fatigue requirements are to be met. The duct is to be milled (minimum thickness 1.7 mm over large surfaces with local thickness minima down to 1.3 mm) from stainless steel or Titanium or Aluminum alloy.

Tentative design

As a starting point, a 3d CAD model (courtesy of Mats Eriksson, Saab) which includes a tentative design of the air duct and the ejector pump with outlet is supplied. It is indicated in Fig. 2 how this duct should interface with the heat exchangers. Note that the interior geometry of the duct is to be left unchanged. The purpose of this duct is to lead air from the

heat exchangers to the ejector pump, and to provide mechanical support for the ejector pump with air outlet; neither the ejector pump not the outlet has any mount point in the frame of theaircraft.

Figure 2: CAD drawing of the air duct to be designed. The attached ejector pump is includedin this CAD drawing, but should not be altered.

Figure 3: Views of the air duct with structural connections. Here, T1 and T2 are measurementbosses, C and D have radial bearings, while A, B, P and Q have spherical bearings. The positions of A, B, C and D cannot be changed.

The air duct and ejector pump assembly is mounted onto the frame of the aircraft at four points, A, B, C and D, whose positions should not be changed (Fig. 3). Co-axially aligned radial bearings are used at C and D. Two links, AP and BQ, connect points P and Q on the assembly to mount points A and B, respectively. These links have spherical bearings in both ends. The placement of points P and Q on the duct can be varied to improve the design.

The 3d CAD model includes two measurement bosses, T1 and T2, which can be removed in your design; they are only included for testing purposes, and will not be present in the final product. The gasket indicated in Fig. 3 is subjected to negligible stress, and can be regarded as a stress-free surface.

The air duct has an operating temperature of 224°C. It is to be milled from stainless steel (15-5PH H1025 AMS 5659) or a Titanium alloy (Ti-6Al-4V AMS 4911). The inertial load cases and fatigue requirements of the assembly, as well as some material data is attached as a supplement (Supplement A). The 3d CAD model can be downloaded from the course documents on the Lisam page of TMHL26.

This description of the ejector assembly was developed in collaboration with Jesper Selestam and Mats Eriksson from Saab, Linköping.

Problem

Suggest at least two designs for the air duct connecting the ejector pump to the heat exchangers: one for stainless steel and one for Titanium alloy. Keep the mass of the duct as small as possible. Demonstrate that your designs meets the strength and fatigue requirements,and compute their respective masses.

List of supplements

A. Material data and load cases for the air duct-ejector pump assembly.

Supplement A: Material data and load cases

Stainless steel: 15-5PH H1025

Rm (ultimate stress)= 935 MPa (at T=224oC)

Rp02 (yield stress)=855 MPa (at T=224oC)

Density: 7.83 kg/dm3

Titanium alloy: Ti 6Al 4V

Rm (ultimate stress)= 681 MPa (Vid T=224oC)

Rp02 (yield stress)=571 MPa (Vid T=224oC)

Density: 4.43 kg/dm3

Ultimate load (UL):

At the ultimate load, the cohesion of the part should be maintained, and it should not fail. The von Mises stress should be compared to the ultimate stress (Rm). The part is, however, allowed to

plasticize and to change its shape.

Acceleration (nx, ny and nz): nUL=25*1.5*g=37.5*g in any direction (one direction at a time, not all

at once, g=9.81 m/s2).

Inner overpressure: pUL=0.552*1.5*1.5 =1.242 MPa

Elastic limit load:

At this load there should be no plastic deformation of the part. The von Mises stress should be compared to the yield stress (Rp02).

Acceleration (nx, ny and nz): nPL=25*1.15*g =28.75*g in any direction (one direction at a time, not

all at once, g=9.81 m/s2).

Inner overpressure: pPL=0.552*1.5 =0.828 MPa

Fatigue:

Stress concentration factor:

Consider a region of a part which is designed to include a geometrical feature, say a drilled hole, that will create a stress concentration. Let the nominal stress be the largest principal stress in this region when modeled without the hole, and let the maximum stress be the maximum of the largest principal stress in the region when the model includes the hole. Then, the stress concentration factoris defined as

Kt = (maximum stress) / (nominal stress).

Different geometrical features have different stress concentration factors, depending on their propensity for focusing stresses. You are encourage to refresh your knowledge on this topic.

Target stresses:

Load spectra measured for the aircraft have been converted into limiting static nominal loads (the load observed in the absence of a stress-focusing feature). These limiting nominal loads have been compiled in Table 1 (for stainless steel; it will be extended with data for Titanium alloy later).

nz=15.95*g p=0.552 MPa

Kt=1 617 MPa 720 MPa

Kt =2 380 MPa 476 MPa

Kt =3 296 MPa 386 MPa

Table 1: Limits of the largest principal stress for 15-5PH H1025 (target stresses for Titanium will be added later)

The data in the table have been computed for an acceleration field of 15.95g i the x-, y- och z-directions (not to be applied simultaneously) as well as for an internal pressure. If the stress level does not exceed the nz limit in Table 1, the part will not fail within the first 8,000 hours of flight. This is conservative since the spectra for nx and ny are more favourable than that of nz.

If you find that a region of the part is subjected to high levels of stress, you must make sure to include all stress-focusing geometrical features in that region in your FE model. You will then obtain the maximum stress from the simulation, located at a particular geometrical feature. Suppose this feature is judged to have Kt = 2. Then compute the nominal stress ( = maximum stress / 2) and compare it to the limiting stress tabulated at Kt = 2 in Table 1 (380 Mpa for stainless steel). The most conservative choice for Kt is Kt=1, which you need to use whenever your are in doubt.

Similar limiting values for the nominal stress have been computed and listed in Table 1 for the overpressure load.

How to ensure that the fatigue reqirelemnt are met in this project:

Perform these steps:

1) Apply acceleration in the z-direction and the overpressure p simultaneously, and find the maximum principal stress in the solution.

2) Ensure that this maximum principal stress is less than the stress limit for Kt=1 and nz=15.95*g. That is, the maximum, principal stress should be less than 617MPa for stainless steel.

Repeat the same procedure for the x- and y-directions.

The above is the most conservative application of the fatigue requirements. Explain why in your report.

Bolts

The bolts connecting the different parts of the duct can withstand 5 kN tensile load.