Module Learning Outcomes Assessed:

1. Apply fundamental knowledge to investigate new and emerging technologies.

2. Extract, analyse and interpret data pertinent to an unfamiliar problem, and affect its solution using computer based engineering tools when appropriate.

3. Analyse and appraise the capabilities of computer based models for solving problems in engineering, and the ability to assess the limitations of particular cases.

4. Review and appraise engineering workshop and laboratory skills.

Aim

The aim of this coursework is to demonstrate your ability to receive and interpret data to create and evaluate the credibility of an FEA model as well as critically assess the results thereof; including dissemination using technical arguments. In other words, pre-process, solve and post-process the results of FEA. You will also be required to optimise the structure in terms of maintaining or improving structural performance whilst minimising or maximising given attributes.

Introduction

Finite Element Analysis (FEA) is an invaluable tool to modern day engineers, enabling the capability to make design decisions based on a components simulated performance under different loading and environmental conditions as well as validation of load case and load path calculations.

Welcome! You have just landed a job at an exciting new technology company specialising in GPS trackers for high end luxury cars. Your new reporting manager, the Chief Executive Officer (CEO) of G-P-S systems is very excited about you joining the company as an FE expert. As the only engineer with any background knowledge of structural mechanics and FEA the expectations of you are sky-high!

Although your reporting manager, JC, is very friendly and extremely enthusiastic (s)he has no knowledge of structural mechanics what-so-ever, but JC is convinced that you are an FE guru! The future of the company rests on your shoulders as any significant warranty issues (such as poor structural performance; for example cracked cases leading to malfunctioning units - meaning that stolen cars cannot be found) will severely cripple if not bankrupt G-P-S systems; you need to tread carefully.

Notes

N1 G-P-S systems can only afford 1 FE license - they use Altair HyperWorks; consequently no other FE software can be used for the impending tasks, i.e. this Course Work (CW).

N2 There is only 1 geometry model for all G-P-S trackers; a HyperMesh (*.hm) model is available on Moodle.

N3 G-P-S are working on a number of new trackers to suit different vehicles; however they only want you to work on 1 version in this CW. The version and the loading you will be working on is dependent upon your Student ID (SID).The details are as follows:

a. The material of your G-P-S model is based on the last digit of your SID using the key and table below:
0-2: Aluminium
3-6: Steel
7-9: Titanium

b. The magnitude of the loads for the two load scenarios, F (as shall be specified later) are dependent on the second to last digit in your SID using the following equation:

F (N ) = 2^nd last digit of SID . 0.13

For example if your SID is 1234567 F₁ equals 6. 0.13 ≈ 0.78N.

If the second to last digit of your SID is zero use the value 9 instead.

c. In general G-P-S models are made using two different gauge thickness' according to the 3rd last digit of your SID using the below table:

Your reporting manager is very busy - (s)he does not like reading long-winded reports; hence (s)he has created a template for all reporting which you will find in a separate document entitled:

You must use this document for your CW; no other formats will be accepted. Pay close attention to the instructions in this document. If you do not follow these you will be deducted marks.

N4 You may assume that all loading is linear static and that all material behaviour is isotropic.

N5 You should be as succinct and concise as possible in your responses.

N6 Remember that statements without justifications are not credible.

Task 1 - Pre-processing, solving and post-processing

Your reporting manager JC is convinced that the best possible way to protect the precious electronics to be mounted inside the case is to ensure that the deflection magnitude remains absolute zero. JC wants your view on the feasibility of this; please comment and discuss.

JC now wants you to assess the structural performance of the case subject to two load-cases representing the worst case impacts on the G-P-S case, see figure 1.

Figure 1. G-P-S boundary conditions.

There are two load-cases to be applied to assess the structural performance of the case; LC1 and LC2. The magnitude of the load for both LC1 and LC2 equals F as calculated in N3.b of the notes above and applied as defined in Figure 1. Note that these two load-cases should not be applied simultaneously. The Single Point Constraints (SPC) applied for both LC1 and LC2 are identical and as defined in Figure 1.

To avoid any "nasty" warranty issues JC wants to make absolutely sure that the results are correct; unfortunately there is however not enough time to physically build a case and test it before the cases are delivered for installation in the luxury cars. (S)he therefore wants you to complete a mesh convergency study and critically assess the performance of this particular version of the G-P-S case.

Based on this specific task (s)he wants you to recommend a meshing protocol;

i.e. mesh size, shape-functions and element shape for all future FE testing of G-P- S products. This should include a specific list of the most important mesh quality check parameters and their specific values. Don't forget to justify your recommendations! Also remember that "over constraining" your recommendations means a sharp increase in product cost; on the flip side "under constraining" may lead to warranty claims - either way G-P-S may go bankrupt!

As all G-P-S products are to be installed in cars; vibration analysis forms a vital part of the structural performance. Due to the orientation of the electronics inside the case it is important to know the lowest natural frequency of the case in situ (i.e. when installed in the car). You should therefore determine the lowest natural frequency of the box via a "fixed-free" vibration analysis. It is also important for the electronics engineers to know the approximate location of the natural largest excitation of this mode in order to avoid placing delicate electronic parts in this location. You should note that the electronics engineers are based overseas and so their knowledge only rely on your written documentation.

After a review by the electronics engineers it has been determined that the electronics inside the case will not be damaged by vibrations at the natural frequency you determined in point 1.3. They do however estimate that they will be damaged if the frequency is 1% higher than that found in 1.3. They have therefore decided that there is no need for concern as there is a 1% margin; and surely results obtained via FEA is correct. JC concurs with the electronics engineers; after all time is of the essence and there are many other issues to resolve. As the structural engineer what is your view?

Task 2 - Optimisation

JC has been very pleased with your work so far; but unfortunately there has been a rise in sheet material cost equivalent to approximately a £1.5 increase per G-P-S case. As G-P-S' ambitions are to produce millions of cases per year the impact on the business is immense. As material price has gone up so has the price of scrap metal. Business analysts have therefore come to the conclusion that using as little sheet metal as possible will increase profit of the company. JC therefore asks you to optimise, redesign and validate the case with the objective of using as little sheet metal as possible. JC also advises you that:

• The optimised design must have the same structural performance as the current design; i.e.:
o The displacement values determined in 1.2 must not increase by more than 10%.
o The first natural frequency found in 1.3 cannot change by more than ±10%;
i.e. if you found a natural frequency of 100Hz in 1.3 the first natural frequency of your optimised design must be between 90-110 Hz.
• The cost of manufacturing must not rise significantly; i.e. your design must be as simple as possible.
• The external dimensions of the case cannot change.
• The internal void of the current case cannot change.
• The application of boundary conditions as specified in Figure 1 cannot change.
• It is not possible to change the gauge, i.e. thickness of the sheets used.

Based on your instructions from JC you decide to conduct a topology optimisation of the sides of the case only using the FE model you have already created. Remember to validate your new design before you present it to JC.

Whilst your reduced mass design from point 2.1 is being considered it turns out that a number of G-P-S cases seem to be experiencing fatigue failures caused by the stress concentration factors originating from the three sharp corners at the bottom of the case (i.e. the location of the three "Z" constraints in figure 1). JC asks you for your thoughts/recommendations as to how the stresses could be reduced in these areas (by simple shape changes). He then asks you to ensure that the stresses on the "holes / cut-outs" you have introduced via the topology optimisation is minimised (via shape optimisation). He asks you to send the results (complete with instructions of what needs to be changed) to the workshop so a prototype can be made ASAP.

Attachment:- G P S trackers.rar