Optimization of the modular cross member portfolio

(1)

EXAMENSARBETE INOM MASKINTEKNIK,

Innovation och design, högskoleingenjör 15 hp SÖDERTÄLJE, SVERIGE 2014

Optimization of the modular

cross member portfolio

(2)

(3)

Optimization of the modular

cross member portfolio

av

Anton Vendelstrand

Daniel Pettersson

Examensarbete TMT 2014:67 KTH Industriell teknik och management

(4)

(5)

Examensarbete TMT 2014:67

Optimering av det modulära tvärbalkssortimentet Anton Vendelstrand Daniel Pettersson Godkänt 2015-02-05 Examinator KTH Ola Narbrink Handledare KTH Ola Narbrink Uppdragsgivare Volvo Företagskontakt/handledare Nicolas Dela Sammanfattning Volvo Lastvagnar är ett av världens största lastbilstillverkare och producerar över 100 000 enheter i 15 olika länder årligen. Alla de tillverkade lastbilarna har ett chassi som består av tvärbalkar av olika dimensioner och uppgifter. För att kunna minska produktionskostnader vill Volvo modularisera sitt tvärbalkssortiment för att kunna ha så få unika artiklar som möjligt. Produkten blir således mer flexibel och kostar mindre att tillverka. De vill även undersöka huruvida det är lönsamt att öka materialkvalitén på tvärbalkarna för att kunna minska i vikt. Målet med detta examensarbete var att minska antalet artiklar i Volvos sortiment, undersöka om det lönar sig att öka i materialkvalité samt utveckla måttkedjor som kan implementeras på Volvos tvärbalkar. Allt detta utan en allt för omfattande förändring av styvhet, spänningsvärden och vikt. Projektet började med att undersöka en rad olika tillvägagångssätt och metoder för att minska antalet artiklar. Fyra metoder föreslogs och en valdes ut med hjälp av en PUGH‐ matris. Den föreslagna metoden gick ut på att leta trender bland tvärbalkarnas parametrar, så som tillverkningsvolym, tjocklek, hållängd, vidd och så vidare. Med hjälp av Pivot diagram kunde trender lokaliseras och således paras ihop till moduler. Utöver modulariseringen utvecklades även måttkedjor som resulterade i nya potentiella standarder, CAD‐modeller och simulationer gjordes på de nya modulerna. En del konsekvenser uppkom under genomförandet, en av dem var i stort sett omöjlig att lösa medan de andra innebar småändringar i modellerna. Rapporten resulterar i ett kraftigt minskat artikelsortiment, nya nitmönster och jämförelser mellan de befintliga tvärbalkarna och de nya modulerna. Ett antal sätt för Volvo att öka flexibiliteten i sitt tvärbalkssystem samt hur de kan spara miljontals kronor om året i materialkostnader med hjälp av ett hållfastare stå Nyckelord

(6)

(7)

Bachelor of Science Thesis TMT 2014:67

Optimization of the modular cross member portfolio

Anton Vendelstrand Daniel Pettersson Approved 2015-02-05 Examiner KTH Ola Narbrink Supervisor KTH Ola Narbrink Commissioner Volvo

Contact person at company

Nicolas Dela

Abstract

Volvo Trucks is one of the world’s largest truck manufacturer and produces over

100 000 units in 15 different countries annually. Every single truck has a chassis which consists of cross members of different dimensions and purpose. In order to reduce production costs, Volvo wants to modularize their cross member portfolio and get as few articles as possible. Thus the product becomes more flexible and cheaper to manufacture. They also want to investigate whether it is profitable to increase the material quality of the cross member to be able to reduce weight.

The goals of this thesis were to reduce the total number of articles in Volvos cross member portfolio, investigate whether it pays to increase material quality and develop geometrical relations that Volvo can implement on their cross members. All of this, without any drastic changes in terms of stiffness, stress values and weight.

The project started with an investigation of several methods in order to reduce the total number of parts. Four different methods were proposed, but only one was chosen for further investigation through a PUGH-matrix. The proposed method was based on parametrical trends of the cross member, such as: manufacturing volume, thickness, hole length, width and so on. Using Pivot charts, it was possible to locate trends and thus pair parts into modules.

Except the modularization, geometrical relations were developed, resulting in new potential standards. As well as CAD-models and simulations were made on the new modules. Some consequences appeared during the implementation, one of them was impossible to solve, but the others only resulted in minor changes of the parts.

This report resulted in a major reduction of the existing cross member portfolio, new rivet joints and comparisons between the existing cross members and new modules. It also resulted in a number of proposals in order for Volvo to increase their flexibility and finally how they can manage to save millions of SEK per year, just by increasing the material quality.

Key-words

(8)

(9)

Preface

This project was conducted as the final course of our Bachelor of Science in Mechanical Engineering with specialization in Innovation and Design, at the Royal Institute of Technology (KTH).

The thesis was carried out during an eight week period, equivalent to 15 credits, during the summer of 2014 at Volvo Group Trucks Technology - Chassis Structure, Gothenburg headquarters.

We would like to thank our supervisor Nicolas Dela at Volvo Trucks for the guidance and for making this thesis possible. We would also like to thank our university supervisor Ola Narbrink for the great support before, during and after the thesis.

A special thanks goes to Sebastien Ragot, Catarina Darrell and the rest of the employees at Chassis Structure for sharing their knowledge and patience with us.

Mechanical Engineering Innovation and Design KTH Södertälje

September, 2014

(10)

(11)

Nomenclature

CAD Computer Aided Design

CP Center Piece

DiVA Digital scientific archive

FMEA Failure mode and effect analysis

KTH Royal Institute of Technology

SEK Swedish currency (Swedish crowns)

TP Tie Plate

(12)

(13)

1 Introduction

This chapter describes the background to the given problem. It’s followed by a description of the requirements specification, goals, project limitations, methodologies, and finally the expected result.

1.1 Background

The frame ladder of Volvo’s heavy trucks consists of side rails and cross members. The main functions of the cross members are to stabilize the frame ladder of the truck and hold the component parts of the chassis. The cross members are also part of the

suspension and must therefore be flexible enough to cope with the loads that the truck is exposed to. Depending on the forces and moments, different kinds of cross members are used. The majority of the cross members consists of two or four tie plates, a center piece and a number of rivets/bolts depending on size and performance.

In order to lower the costs of product development, production and spare parts management, Volvo wants to modularize the cross members to get as few different articles as possible. This results in a standard assortment where many small parts are combined to a larger number of assemblies. The product is thus more flexible and less expensive to manufacture since it is cheaper to produce large volumes with the same tool than to produce large volumes with lots of different tools. With standard interfaces, it’s also possible to easily replace a cross member with another one, without affecting other components of the truck.

(16)

Volvo is also aware of the environmental aspects, which can be improved by decreasing the amount of waste material in the manufacturing process and more efficient logistics. Geometrical changes of the cross member will result in less waste at the cutting process. In some cases it may be better to enhance the material quality in order to reduce the thickness of the cross member, which may result in a lighter product. These changes must be done without losing performance of the cross member and increase costs of the final product.

1.2 Requirements specification

The general requirements of this project were to establish a standard catalog of Volvo's cross members that cover as many articles as possible. Using modularization it’s possible to standardize and reduce the number of items, thus reduce manufacturing costs and increase flexibility. Another requirement was to investigate whether the quality of the material could be increased and thereby reduce the weight of the cross member without compromising the performance.

1.3 Goals

The goal of this thesis is to create an optimal set of modular cross members. The modular cross members will be created from a cost, weight, production and functionality point of view. The new parts will replace some of the older parts, but still meet the requirements listed below:

 The new cross members shall have better or equivalent stiffness and stress values as the replaced cross members. A change of 15% is acceptable

 The new cross members must be within the same weight range as the replaced cross members.

 The total number of parts in the portfolio must be reduced.

 Investigate whether it´s profitable to increase the quality of materials.

 Develop geometrical relations that can be implemented on Volvo’s cross members.

1.4 Project limitations

In the investigation of developing optimized modular cross members, the following limitations will be considered:

 Only a certain amount of cross members will undergo the modularization process. The cross members which are located at the front and rear of the truck, will not be considered. This is because they separate themselves from the other cross

members and it would take more time to find a way to modularize.

 Physical prototypes of the new cross members will not be created since it’s too time consuming. Only simulations using the software Creo Simulate 2.0 will be made to compare the performance between existing and proposed cross members.  The project group will not consider the already standardized hole spacing system

(17)

1.5 Methodologies

This section presents the planned methods that would be used to achieve the goals of the project. The choice of methods have been made out of personal experiences and

counseling from supervisors and Volvo staff members.

Internet research 1.5.1

Web-based search engines and databases were used to find relevant facts related to the task.

DiVA

Is an effective and reliable search engine that works well to find both basic and advanced facts. DiVA is an open archive with research publications and student thesis from 34 different universities. (Enheten för digital publicering, 2000)

Google Scholar

This is a simple tool that searches widely for scholarly literature across a big number of disciplines and sources. (Google , 2011)

Literature 1.5.2

Literature in the form of books and compendiums was used to regain knowledge in some relevant areas, such as modularization and design methodology.

Computer software 1.5.3

The group used the following software in order to achieve the project goals: PTC Creo Parametric 2.0

A powerful software that was used to model the generated concept ideas into 3D. PTC Creo Simulate 2.0

Was used to simulate the generated 3D models and examine how different load cases affected the cross members.

KeyShot 4

A rendering software that was used to convert the 3D models into realistic images. Keyshot took account of materials, lighting, shadows, reflections, environments, among others.

GanttProject

(18)

Microsoft Excel

During the modularization phase, Microsoft Excel was used in order to compare existing cross members. Pivot charts simplified the overview of the components in the search for trends regarding geometry and performance.

Interviews and meetings 1.5.4

In the beginning of the project, main focus was to interview employees with expertise in the cross member area. Among these were supervisors and key employees within the group of chassis structure. This led to a deeper understanding of the problem and existing solutions.

During the project, regular meetings were held to ensure that the project was on track.

Study visits 1.5.5

Study visits were made throughout the project. The visits were made mainly in Volvo Trucks workshop.

Pre-study 1.5.6

Before the concept generating phase began the group analyzed the existing cross member systems which Volvo had developed. This was an instructive way to learn more about the problem.

Evaluation and decision models 1.5.7

In order to evaluate the project and the product, the group used the following evaluation and decision models.

PUGH’s matrix

PUGH´s decision matrix was used to evaluate the method that were suitable for further development. The method was performed twice with different references to obtain the best results.

FMEA

FMEA was used to get a clear picture of what would happen if something went wrong during the project.

Innovation methods 1.5.8

Methods used to generate ideas and innovative concepts. Brainstorming and mind mapping

(19)

Product decomposition

Product decomposition was used to get a better understanding of what components the cross member consisted of, how they were manufactured and how the interfaces

interacted with each other.

1.6 Expected results

The expected result of this thesis was to achieve the goals by using the specified solution methods above (1.5 Methodologies). This resulted in a proposal for a standard portfolio of modular cross members that will help Volvo to reduce the number of articles and thus bring down manufacturing costs. With a smaller article catalog the flexibility of the product increases and thereby reducing the administrative burden.

(20)

(21)

2 Current situation

This chapter will present a description of the company and the environment in which the thesis was made in. The chapter ends with a summary of previous standardizations.

2.1 Volvo Trucks

Volvo Trucks is a Swedish truck manufacturer based in Gothenburg, owned by AB Volvo. It was founded 1928 and today it's one of the world's largest manufacturer of heavy-duty trucks for long-distance transport, regional transport and civil engineering sites. Volvo produces over 100 000 units a year in 15 different countries (See figure 3.1). They offer transportation solutions to customers in over 140 countries and they are well known for their achievements in areas such as; safety, quality and environmental awareness (See figure 3.2). Volvo also has three well-known sister companies called Renault Trucks, Mack Trucks and UD Trucks. (Volvo Trucks AB, u.d.)

The thesis was carried out at Volvo Group Trucks Technology - Chassis Structure at Gothenburg headquarters. The project group had the opportunity to work at the main office along with experts in the field.

2.2 Implemented standardizations

Standardization is a documented knowledge, or to establish common conventions and a way to work in order for the company or organization to operate smoothly. In the current situation, only the hole system on the frame and tie plates can be considered as an

implemented standardization regarding Volvo’s cross member portfolio.

Figure 3.2 – Market shares from year 2011 Figure 3.1 – A map of countries which Volvo

(22)

(23)

3 Theoretical frame of reference

This section contains a description of the theoretical models that was used to solve the task. The chapter ends with a brief description of the relevant subjects gained from the education that were of importance.

3.1 Project process

This thesis was based on KTH's standard process according to Projekthandboken (Tillämpad Maskinteknik, KTH Södertälje, 2011). The project was divided into four different phases in order to get a clear structure and overview of what is covered under the frame of the thesis (See figure 4.1).

Each phase was conducted in order, with a follow-up meeting with the project supervisors between each phase. The purpose of these meetings was that the supervisors could check that the project was on the right track and if it was time to move on to the next phase. A Gantt chart was made in the beginning of the thesis that was followed closely throughout the entire project. Given that this project were to be performed during a shorter period of time than a traditional thesis, great emphasis was put on the Gantt chart. It was followed carefully and constantly kept up to date.

Figure 4.1 – Time line of the project process

3.2 What is a module?

The definition of a module is; a functional building block, with clearly defined interfaces selected for corporate specific reasons. (Imsdahl, u.d.)

3.3 Load cases

Three load cases have been developed and analyzed in Creo Simulate. The goal of the simulations was to get an understanding where the "hot spots" appeared, in other words, where the maximum stresses occurred. Once the "hot spots” was found, improvements could be made in the construction by changing geometry and/or material of the cross

Problem analysis Gathering of facts Implementation Result & documentation Meeting with

supervisor 1 Meeting with supervisor 2 Meeting with supervisor 3

Meeting with supervisor 4 & presentation

(24)

members. Since it’s difficult to perform simulations that correspond to reality with a limited model, only simplified simulations with assumed loads were made. Load cases 2 and 3 are originally carried out simultaneously in reality, but in order to make it easier to simulate and evaluate, they have been divided into two different simulations.

Frame torsion 3.3.1

Frame torsion is a load case that occurs when the truck is driving on rough roads. Depending on which wheels that are exposed to irregularities, either the front or rear wheels will be displaced vertically, causing the frame to twist. In order to simulate this load case, vertical forces were placed on each side of a frame in opposite directions.

Figure 4.2 – Load case 1 - Frame torsion

Lateral force 3.3.2

The frame is exposed to lateral loads when the truck is turning and causing the frame to lean in the same direction as the lateral force (See figure 4.3). Depending on the center of gravity, the frame will lean differently.

Frame displacement 3.3.3

As stated in the beginning of this chapter (4.3 Load cases), both load case 2 and 3 occurs when the truck is turning. Load case 3 is a simplification of the frame displacement that occurs in the z-direction of the frame while the truck is turning.

(25)

3.4 PUGH’s matrix

Pugh’s decision matrix is based on finding the best concept solution to a problem, from a selection of other concept ideas, in a systematic and objective manner. The decision matrix is based on six steps as visualized in the figure below (Figure 4.5). The method is easy to perform and effective when different variants of concepts are compared.

One concept will be chosen as a reference, the remaining concepts will then be compared to it. If the compared concept is better than the reference for a specific criteria, it will be rewarded with a plus (+), if it´s worse it will be given a minus (-), and if the criteria is equivalent with the reference it will be rewarded with a zero (0). When all legends have been deployed, a clear picture is given of which concept is the best, based on the

established criteria. The method provides the best results if all group members perform it independently and then compare it with each other. (Ullman, 2010)

1. The issue

3. Criteria 4. Importance

2. Alternatives

5. Evaluation

6. Results Figure 4.5 – PUGH's decision matrix

(26)

3.5 FMEA

FMEA stands for “failure methods and effects analysis”. The method is used to find risks that can occur during the course of a project. The method also identifies the cause of errors and how they can be prevented in the future. FMEA treats issues such as; what happens if there isn’t enough time to finalize the report or a specific deadline? What happens if the budget is exceeded? FMEA is divided into five different steps that must be completed in order to understand the entire problem.

Step 1 – Identify functions that are affected

The first step of the method is to identify the function or process that could fail because of an error that occurs in the project.

Step 2 – Identify failure modes

There can be many failures for each individual function or process. A failure mode is a description of how a problem could happen. It may be helpful to go back to step one and rewrite the function or process if it´s difficult to think of how the problem occurs.

Step 3 – Identify the effect of failure

What are the effects and consequences if a failure happens? If a failure occurs it may affect the production, customer, system, guidelines and so forth. Therefore it´s important that everyone affected will be considered.

Step 4 – Identify the failure causes or errors

This step identifies what causes errors, how likely they will occur and how easy it is to discover them.

Step 5 – Identify the corrective action

The final step consists of three parts, first off all there has to be a recommended action to prevent or minimize the same error of happening again in the future, the second part identifies the person responsible for the action and the last part describes what action was taken. (Ullman, 2010)

3.6 Design tools

Brainstorming

Brainstorming is an idea generation method that requires a group of people. A

(27)

Mind mapping

Mind mapping is a tool that can be used while brainstorming. It´s a way of representing ideas and concepts graphically. Mind mapping can be used to take notes, make plans, solve problems, present information and much more. The procedure of the tool is to write a headline in the center of a paper, and then write down related subtopics with lines connected to the center. (Passuello, 2007-2014)

Product decomposition

Product decomposition is a method that can be used to gain an understanding of how an existing product is built. The method can be used as a starting point to consider a

redesign. But also to understand characteristics of the components and what their functions are. The product is disassembled and all components are registered in a table with manufacturing methods, materials, part names and quantities.

3.7 Relevant subjects

(28)

(29)

4 What is a cross member?

In this chapter the function and structure of the cross member will be described in detail. This is for the reader to be prepared and get a clear picture about the context of the report. Some of the elements and dimensions in this section are fundamental in order to understand the implementation and result of the thesis (See figure 2.1).

Due to different functions the cross members have different designs. The major drivers for the design are:

 Specific needs for the wheel suspension

 Demand on torsional stiffness of the frame

 Demand on stiffness from components mounted to the frame

 Demand from other systems assembled on or close to the cross member

 Light weight

Center piece _Rivets _{Tie plate}

(30)

4.1 Chassis

The chassis is an internal structure that forms the truck's skeleton. The chassis of a truck consists of a steel frame that basically holds up the rest of the vehicle. Moreover, if the wheels, engine, transmission, driveshaft, differential, and suspension are mounted it’s called a rolling chassis.

4.2 Frame

Two longitudinal U-shaped beams keep the vehicle together while they also work as a part of the suspension (See figure 2.2). The frame has a number of different functions. It serves as a support for the chassis components, the body and it needs to withstand the static and dynamic loads that the vehicle is submitted to. Some examples of loads that the frame is exposed to are:

 Cargo load

 Uneven road surfaces causes vertical and torsional twisting  Transverse lateral forces caused by cornering

 Tensile forces and compression caused by acceleration and breaking  Torque caused by engine and transmission

(31)

As mentioned earlier, the frame functions as a mounting surface to other components and brackets. The longitudinal beams have the same length as the complete truck and the standardized hole system starts after the frame bend in order to fasten cross members. (See figure 2.3)

Inner-liner 4.2.1

The inner-liner is made out of steel and works as a reinforcement which is placed on the inside of the frame. It contributes to a more rigid structure. The inner-liner is 5 mm thick and has the exact same standardized hole system as the frame.

In order for the distance between the longitudinal beams not to increase 10 mm when the inner-liners are mounted, Volvo have chosen to make the cross members with inner-liner 10 mm shorter. The majority of the produced trucks don’t have an inner-liner.

(32)

4.3 Tie plate

The tie plate is the part that connects the frame with the cross member, advantageously with screws or rivets. Usually four tie plates are required to attach the center piece. Length and height are measured by number of holes (See figure 2.4). Some tie plates have more than one row of holes (hTP>1). This is mainly to reinforce and stabilize the brackets that must be attached to the tie plate. Volvo strives to use one type of tie plate both up and down on the center piece.

Table 2.1 – Descriptions of measurements, units and explanations for the tie plate

Tie Plate (TP)

Measure Unit Explanation

Length (lTP) Number of

holes

The length of the tie plate is measured by the number of holes along the y-direction. The example below has lTP=15

Height (hTP) Number of

holes

The height of the tie plate is measured by the number of holes along the z-direction. The example below has hTP=2

Base (bTP) Millimeter The base of the tie plate is the length between the top of the tie

plate and the bending radius

Thickness (tTP) Millimeter The thickness of the tie plate is the same as the thickness of the

sheet metal Z-direction 1

(z1) Millimeter

z1 is a measure of the projection in the z-direction between the

inner surface of the tie plate to the closest hole center Z-direction 2

(z2) Millimeter

z2 is a measure of the projection in the z-direction between the

outer surface of the tie plate to the closest hole center

Geometry Explanation

Rivet holes The number of rivet holes in the rivet joint. The example below has 7 holes

Tie plate web

(33)

4.4 Center piece

The cross members connects the two longitudinal beams, which means that the shape is similar to a ladder. The center piece should not be completely stiff. The reason for this is that the frame requires some flexibility since it's a part of the suspension.

The center piece is a piece of sheet metal, punched and bent into a U-beam. It’s designed in a weight and cost efficient geometry that meets the performance requirements. On the web of the center piece there is a standardized hole system. The hole system works as a supplement to the frame, and makes it possible to mount brackets and routing on it. As mentioned earlier, there are two width-standards on the cross members depending on whether it's an inner-liner mounted on the frame or not. The cross member needs to be 826 mm wide with an inner-liner and 836 mm wide without. If there is an inner-liner, each rivet joint need to be moved 5 mm towards the center of the center piece. This will not affect the manufacturing particularly much, since the tool only needs to move 5 mm in one direction. The consequence is that Volvo basically gets two identical, but still unique, articles. lTP hTP bTP z2 z1 Rivet holes TP Webb tTP

(34)

Table 2.2 – Descriptions of measurements, units and explanations for the center piece

Center piece (CP)

Measure Unit Explanation

Cross member

width (wXM) Millimeter

The distance between the outsides of the tie plates on the cross member

Center piece

width (wCP) Millimeter The distance between the outsides of the center piece

Length (lCP) Millimeter The length between the web and the opposite edge

Height (hCP) Millimeter Distance between the outer sides of the rivet joints

Thickness (tCP) Millimeter The thickness of the center piece is the same as the thickness of

the sheet metal

Geometry Explanation

Y-radius (1) This is a radius in the y-direction. It is the same on the upper and lower part of the center piece. Mainly to reduce weight.

X-radius (2) This is a radius in the x-direction of the center piece. It is based on space for routing and results from calculations.

CP Web (3) The CP web is the outer surface with the standardized hole system

(35)

wXM hCP lCP wCP 2 3 4 1 tCP

(36)

4.5 Rivets

Riveting is one of the oldest fastening methods there is. A solid rivet consists of a head and a shaft that is plastically deformed when fastened. It is a permanent joint and can´t be removed from the hole unless breaking it. There are different types of fastening tools that can be used to fasten rivets such as a pop rivet plier, hydraulic riveting gun and rivet machines. However, the rivet gun is the tool that is used to fasten the components of the cross members.

The rivets are in most cases manufactured with a weaker material than the components it´s fastened with. Therefore it doesn´t contribute to better performance properties. Rivets have the best performance properties when subjected to shear loads, but are also capable of being exposed by tensile loads even though it´s not what they are intended to do. The performance of the rivet connection is determined mostly by shear and bearing stress.

4.6 Manufacturing method

The majority of Volvo's cross members are neither manufactured nor surface treated in-house. They are purchased from external manufacturing companies in Europe. Assembled cross members are sent on pallets to Volvo´s plant where they are mounted to the frame. The steel used in the majority of the cross members are Domex 460 MC (Nordin, 2014). The manufacturing process starts with a sheet of steel with proper thickness and

dimensions. Then the contours and hole patterns are punched out, although low volume parts might be laser cut. Finally the flat part is shaped, usually by bending. Both center pieces and tie plates undergo the same operations although the tools are different (See figure 2.8).

When the part has been shaped, the surface needs to be treated. These processes contain a cleaning operation (blasting), a pre-treatment, a primer coat and finally a top-coat (paint job). When the whole process is done, the parts are sent to assembly.

Shaft Head

(37)

Contours Hole pattern Shaping Center piece Tie plate Green = Cutting Blue = Bending

(38)

(39)

5 Implementation

This section describes how the project were implemented and methodologically solved.

5.1 Choice of method

Based on idea generation, four different proposals on methods were presented. These were considered to eventually help the project group to achieve the goals that were established in the beginning of the project. No detailed information about the methods is described, only general ideas and suggestions. The reason for this was that no methods would be excluded even though they were less realistic.

1. Create new cross members replacing the existing

This method is based on designing a completely new cross member, including tie plates, center piece and riveting systems. This method generates a new innovative product that replaces the existing cross members with the approximate same stiffness. With a new design, there are opportunities to think outside the box and not be locked into old ways of thinking. However, this method is a bit time

consuming since prior knowledge about how to develop a brand new cross member is needed.

2. Calculate max / average / min stiffness and replace existing cross members

This method proposes that a maximum, medium and minimum stiffness on Volvo's cross member portfolio is calculated. Once these values are identified, three new cross members will be designed and developed to meet the required stiffness values. This can be done several times until a major part of the catalog is covered and the required performance of the cross members are met. In this way it's possible to reduce the number of articles and replace them with a new modular portfolio.

3. Locate trends and combining a solution

(40)

4. Calculate max / min for each section of the chassis then replace each section with a cross member

By dividing the chassis in different sections, it's possible to focus on the specific requirements for a cross member in a particular part of the chassis. Since the geometry and design of the cross member is similar depending on where it's located, the selection is simplified considerably by examining some sections separately. In order to find out the performance of the cross member, calculations on maximum and minimum stiffness should be made. This will result in an interval which the new cross member should cover in that specific section.

Criteria 5.1.1

Six criteria were developed to compare the proposed methods in PUGH's decision matrix. Some of the criteria relates to modularization, while others relate to personal knowledge and general conditions within the group. All of them were weighted by relevance. The criteria that were considered during the evaluation of the methods were:

1. Generates the least number of modules

If the portfolio is supposed to decrease, the method should generate as few unique articles as possible.

2. Precision in terms of performance

Some methods lead to a more accurate value on the required performance. These methods, however, could mean that the method also were more time consuming.

3. Realization

The realization was weighted after how likely it was that Volvo would implement this method after the thesis was completed.

4. Compatibility (fit with the existing components)

How well the new components would fit with each other and together with the existing portfolio.

5. Competence

A criteria that were weighted according to how high competence the group members had in relation to the method that were to be performed.

6. Hardware and software

(41)

Selected method 5.1.2

As mentioned earlier, the choice of method was based on PUGH's decision matrix (See appendix A.8). Two matrices were conducted in order to select a method to proceed with. In the first matrix, method number two (See above) was chosen as reference. It came on second place regarding to the evaluation, after method number three.

To see if method number three was the best choice, it was chosen as reference on the second evaluation. Method number two was removed from the matrix since it would have been the inverse of method number three if they were to be evaluated together again. The result of the second evaluation concluded that method number three was preferable for the implementation of this thesis.

5.2 Product decomposition

A cross member was decomposed in Creo Parametric in order to get a better

understanding of how the components interact with each other (See appendix A.1).

5.3 Implementation of method

In the beginning, all the cross members listed in the excel document were studied carefully in order to get a grasp of how they differ from each other, how large volumes they are manufactured in and where they are located on the frame. After this

investigation, 27 cross members were chosen out of Volvos portfolio. The cross members that were chosen had similar designs and were comparable in size. The cross members located at the rear of the frame were excluded from the modularization since they have completely different designs in comparison with the rest of the cross members. Cross members that weighed nearly 100 kg was also removed because they were few and weighed more than twice as much as the cross members which were selected for further development.

When the implementation phase began, the cross members were studied in their entirety with all associated components. It quickly became clear that this was not a good approach to modularize, since there were too many components and data to keep track of at the same time. Hence the components were studied individually.

(42)

All components of the cross members are compiled in an Excel document with important data which Volvo have created. It was difficult to see differences regarding the designs between the components since they were poorly visualized. One way to visualize the components better was to create assemblies with tie plates and center pieces. This made it easier to compare sizes, hole geometries and designs.

Figure 5.1 – Assembly of the chosen tie plates

Pivot chart 5.3.1

In data processing, Pivot tables are used as summary tools. A Pivot table can automatically sort, calculate totals and averages of the stored data. The results are then instantly

summarized and displayed.

The project group used a number of Pivot charts to find trends among the cross member portfolio. This was an easy way to sort out the important parameters that were significant for the existing cross members to be compatible. The main focus was on width, length, thickness, number of rivets per joint, and production volume (no particular order). The choice of parameters was based on that they were consistent and that they affected other parameters. If there were patterns to follow, it was possible to prepare geometrical relations which may later be adopted into standards.

(43)

Breakdown and pairing 5.3.2

In order to get a better overview of the tie plates and center pieces affected by the modularization, all the components were printed. This made it possible to put all the components on a whiteboard and match those that had similar dimensions with each other. It became easier to see which components that had been left out and how they could be included in the module system.

New parts were created with Creo Parametric when the breakdown and pairing was finished and analyzed. The starting point for creating the new components was to use an existing part that was manufactured in large volumes as a template. This way the risk of increasing the weight of a large volume component was minimized. The complete module system can be found in the appendix (see Appendix A.2).

0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 207 57 46 8 2124 4 71 0 216 66 0 65 216 68 23 1 2205 54 70 211 34 38 7 215 46 93 4 2124 8 11 3 219 63 6 98 211 34 38 9 213 54 0 35 2205 9 93 1 208 29 41 6 208 29 41 9 2124 8 11 4 2124 8 11 5 215 46 93 3 215 46 93 6 213 31 11 5 215 47 47 8 (bl an k) 6 8 11 12 8 11 15 9 12 (blank) 6 7 8 (blank) Total

Total number of tie plates in the portfolio with the same weight and length

tTP _l

TP _{TP article number}

Trend

Figure 5.3 – The selected components were placed on a whiteboard and then paired together

(44)

5.4 Models

A design tool called sheetmetal was used to create all components in Creo Parametric. The main function of sheetmetal is that a detail can be bent and straightened. The tool is effective to use when editing bending angles, thickness and flange length of a detail. A feature of sheetmetal is that a visualized image of how the detail will look like, directly after the cutting work, can be obtained. This function can be used to see how much material that will be cut off during production. When a drawing is made on a sheetmetal detail, measurements can be deployed where the bend occurs. Once all the components were modelled an assembly was created, where tie plates, centerpieces and frames were included.

5.5 Motivation for standardization

Below are the four main modularization and optimizations that were proposed to implement on Volvo's cross member portfolio:

1. Reduce the number of articles using symmetric tie plates:

This reduces the product catalog significantly since numerous tie plates are

asymmetrical. Volvo doesn't see this as a major problem since some tie plates are each other's mirror image and therefore no tooling costs needs to be added. The positive aspect of using symmetrical tie plates is that it's easier to combine different parts when all geometries are based on the center of the rivet joint.

2. Compatibility between center pieces, tie plates and modules

This can be done in order to reduce the number of articles, but also to reduce or combine a variety of modules. All trends and similarities among the articles should be examined and paired together into one common piece. It's also possible to do the same thing on finished modules. If they are similar and if no big modifications are required, two modules can be combined into one

Figure 5.4 – A tie plate created in sheetmetal

(45)

3. Standardized number of rivet joints

A determined number of rivets per joint make it possible to combine different items with each other. Moreover, it's possible to specify an absolute position for all joints in the catalog, which makes it more consistent.

4. Standardized measure of web-x distance.

The web-x distance is important in order for the cross member to be replaceable and thus increase the flexibility of the modular system. A lot of components and routing is attached to the standardized hole system on the center piece. If the web x-distance always is the same in the entire portfolio, the components and routing on the center piece will always be independent of the cross members. The reason for this is because the hole system always will be in the same position, regardless which cross member that is used. If the web-x distance however differs, the fitting of the components on the cross member will be problematic.

5.6 Geometrical relations

Trucks are complex designs, with numerous of parts that depend on each other. In order for everything to fit optimally, it's required that there are certain standards and measures to relate to.

Distance between mounting holes 5.6.1

Volvo has implemented a standardized interface for the mounting holes in the frame. This means that the mounting holes on the tie plate webs must have a certain distance from each other, in order to make it possible to fit the cross member to the frame. When

dimensioning a cross member that has one unique tie plate there are two dimensions that needs to be considered; hCP and tTP. These variables need to measure 180 mm together, in order to fit into the standardized hole interface of the frame (See figure 5.6)

One of the goals was to make the cross members more flexible by combining different types of tie plates to a center piece. By doing so there are two more dimensions that have to be considered; z1 and z2, which results in the following equation:

This geometrical relation can be useful when developing new cross members or combining different tie plates with a center piece.

[5:1]

[5:2] 180 = hCP + tTP

(46)

Web x-distance 5.6.2

Center piece web x-distance is a measure of the projection in the x-direction. This projection is the distance between the CP web to the nearest hole center on the tie plate (see figure 5.8).

When a cross member is mounted with the frame it has to be attached from the inside. This means that the center piece can’t be too close to a fastening hole, which otherwise will result in that the screw can´t be inserted (see figure 5.7). The bolt requires a radius of 15 in order to be insertable (see Figure 5.8). To enable all screws to be insertable, the CP web must be positioned between two fastening holes.

Figure 5.7 – Screw clashing with the center piece

Figure 5.6 – The figure shows the front view of the cross member with important variables.

Center piece (web)

(47)

Figure 5.8 – The side of a cross member with a standardized distance

To ensure that the web x-distance is the same for all cross members, a standardized distance was introduced. The distance is measured from the center of the rivet joint of the center piece to the CP web. As of now, the standardized distance is 73 mm, meaning that web x-distance will always be 23 mm. (See figure 5.8). An explanation to why the

standardized distance was chosen will be presented later in this chapter. The selected center pieces from Volvos cross member portfolio have a range of

thicknesses between 4-7 mm. When the thickness of a center piece increases, it increases inwards (see Figure 5.9). It is important that the standardized distance from the center of the rivet joint of the center piece to CP web is large enough, so that the increased

thickness of the center piece does not prevent the mounting of the screw.

Tie plate

Center of rivet joint (center piece)

Center piece Area that is required for fastening

(48)

For an example; if the thickness of the center piece is 6 mm, the distance x will be 23-6 = 17 mm, which means that the screw will be insertable because 17 > 15 mm. The result of this will be that the thickness of the center piece can´t exceed 8 mm, since the distance x will be 23-8 = 15 mm (See figure 5.8). Centerpieces with thicknesses of 8 mm and above can be compatible if the standardized distance (73 mm) increases. However, this means that lCP must increase since the rivet joint must be repositioned further out towards the edge. Otherwise, the rivet joint holes will be too close to the edge, which will result in weaknesses in terms of performance.

Due to geometrical reasons, there has to be an odd number of mounting holes in the tie plates (see figure 5.11). If there is an even number of holes and the tie plate is

symmetrical, the side of the CP web will be placed right in front of a mounting hole, making sure no screw can be inserted (see figure 5.12). This problem can be solved by using asymmetrical tie plates, which means that the center of the rivet joint is displaced (see figure 5.13).

CP Web

Rivet holes

x = 23 - t [5:3]

Figure 5.9 – When thickness of the center piece is changed, it increases inwards

Figure 5.10 – Even mounting holes in the tie plate

(49)

It´s not preferable to manufacture asymmetrical tie plates since two different articles has to be created. And since one of the goals was to decrease the number of articles of the modular cross member portfolio, it is preferable to keep the number as low as possible.

Standardized rivet joints 5.6.3

The benefit of standardizing the rivet joints is that components from different modules can be combined with each other. What´s important when standardizing is that there needs to be an optimal set of rivet joints that replaces the previous ones, which still

withstand the highest stresses. Currently there are several different types of rivet patterns in Volvos cross member portfolio.

When the standardization of the rivet joints commenced, an investigation was made in order to find out which hole pattern that was utilized the most. It turned out that the patterns that consisted of three or six rivets per joint was used the most. The benefit of using an extra rivet is that it relieves forces from the other rivets. It also makes it possible to modularize the components easier.

Asymmetrical tie plate

Symmetrical tie plate

Not possible to insert a screw

Side of CP web

Figure 5.12 – Side of CP web in front of a mounting hole (symmetrical tie plate)

Figure 5.13 – Side of CP web between two mounting holes (asymmetrical tie plate)

(50)

As of now, the rivet joints of the tie plates and center pieces are sketched from scratch when a new model is created. To simplify the CAD work of the rivet joints in the modelling phase, standardized rivet joint sketches were created. Two rivet joint sketches with three and seven holes were made. The idea is that the standardized rivet joint sketches will be used in as many components as possible in Volvos cross member portfolio. When a new tie plate or center piece is created, the required rivet joint can be selected from a folder and then be inserted into the model. This means that the people creating a new

component doesn´t need to spend unnecessary time on drawing holes and put dimensions on them.

Cross member width 5.6.4

In order to combine different types of components from different modules, some parameters need to be standardized. The first dimension (L1) is; from TP web to the center of the rivet joint. The second dimension (L2) is; from the middle of the center piece to the center of the rivet joint. The figure below illustrates the dimensions (see figure 5.18). These dimensions must add up to either 826/2 mm or 836/2 mm, depending on whether the cross member is intended to have an inner-liner or not. If these parameters are not fulfilled the cross member will be too big or too small to fit inside the frame.

Figure 5.15 – Rivet joint with three holes

Figure 5.16 – One of the former rivet joints with 6 holes

Figure 5.17 – The new rivet joint with 7 holes

(51)

Since L1 and L2 must add up to 826/2 or 836/2, four dimensions needs to be standardized. This is because the rivet joints containing three and seven rivets have a different center positions. Descriptions of the distances are listed below:

 LI7 = Length from edge of center piece to the center of the seven rivet joint, with inner-liner

 LN7 = Length from edge of center piece to the center of the seven rivet joint, with no inner-liner

 LI3 = Length from edge of center piece to the center of the three rivet joint, with inner-liner

 LN3 = Length from edge of center piece to the center of the three rivet joint, with no inner-liner

Table 5.1 – Standardized values of LI and LN

Rivets LI LN

3 61 mm 56 mm

7 75 mm 70 mm

These dimensions are not optimal, but the thought is to bring out the idea of

implementing this kind of standardization. This results in an easier combination between different components and modules. When optimal dimensions are found for this

standardization, L1 and L2 can easily be calculated with the following formulas:

𝐿

₂

=

𝑤𝑖𝑑𝑡ℎ 𝐶𝑃

2

− 𝐿

𝐼3

, 𝐿

𝐼7

, 𝐿

𝑁3

, 𝐿

𝑁7

𝐿

₁

=

𝑤𝑖𝑑𝑡ℎ 𝐶𝑃₂

− 𝐿

₂

[5:4]

[5:5]

Figure 5.19 – Distance LI7 between the edge of the center piece to the center

(52)

5.7 FE-analysis

Creo Simulate treats assemblies as rigid bodies, which makes it difficult to investigate how the rivet joint can withstand stresses. In the subsequent simulations, only center pieces have been examined.

Quality of material 5.7.1

Steel was used as material for all components during the simulations. This was necessary in order to compare with different material qualities in the result phase.

Mesh 5.7.2

A mesh is created after all components have been assigned a material. AutoGEM generates a mesh that is suitable for the model to be analyzed. A simulation provide more accurate results if the mesh is dense and compact in areas that is affected by stress concentrations. By having a dense mesh the computer requires more memory, resulting in that more time is needed to perform the simulation. The mesh in the figure below was created in the thick version of module 1 (see figure 5.20), it consists of 17606 tetrahedrons.

Constrains and loads 5.7.3

Load case 1

In load case 1 the cross member is subjected to torsion. The right frame is locked in all directions (x, y, z). The left frame is fixed in the x-direction. This is because the stresses will be distributed more evenly on the center piece, which makes it easier to find the maximum stress of the graph that are created during the simulation.

Two opposite vertical loads are placed on the ends of the frame to the left. Since the loads are placed on lines, it may result in stress concentrations when the simulations are being made. Therefore a bit of the ends of the frame are excluded with the tool "isolate for exclusion".

All constraints and loads are visualized in the figure on the next page (See figure 5.22).

(53)

Load case 2

In load case 2 the cross member is subjected to lateral loads. In this simulation, only the right frame will be fixed in all directions (x, y, z).

Two opposite loads are placed on the edges. Since the loads are well distributed along the frame, no exclusion will be made. All constraints and loads are visualized in the figure below (See figure 5.23).

Load case 3

In load case 3, the frame is subjected to a load which displaces the frame in the z-direction. The right frame is locked in all directions (x, y, z). The left frame however, is locked in x- and y-direction. Otherwise the center piece will be displaced obliquely in the x- and y- direction. The horizontal load is placed at the far end of the left frame. All constraints and loads are visualized in the figure below (See figure 5.24)

Loads

Constraints

Figure 5.22 – Load case 1 with associated constraints and loads

Constraints Loads

(54)

Load

Constraints

Simulation 5.7.4

Three load cases will be simulated to find out which one of them that is exposed to the highest stresses. The load case that results in the highest stresses will be the basis for further dimensioning. A full analysis of all load cases with all modules had been optimal to do. But since all modules have basically the same design, the load cases will have the same spread.

Load case 1

The frame torsion causes strains along the y-radius of the center piece. The stresses are evenly distributed on both sides but are low in the middle of the center piece (See figure 5.26). The inner side of the bend of the tie plates are exposed to stresses as well (See figure 5.25). High stress concentrations occurred in the upper edge of the center piece where the tie plates overlap, these stress concentration were neglected.

Stress concentrations

Figure 5.24 – Load case 3 with associated constraints and loads

Figure 5.25 – Stresses on the inner sides of the bends

Figure 5.26 – Stress concentrations developed at the upper edge of the center piece

(55)

Load case 2

The center piece is very stiff when it´s subjected to lateral loads in the x-direction. This results in relatively low stresses in the cross member. The highest stresses occur in the corners of the inner bends of the tie plates (See figure 5.29). Rivets and screws are subjected to tensile stresses (See figure 5.30).

Figure 5.27 – A graph of the stresses that occurs on the y-radius of the center piece. X-axis = length of curve, Y-axis = stress.

Figure 5.29 – The figure shows the stresses in the bends

(56)

Load case 3

The results of load case 3 resembles with load case 1. Although the last mentioned is subjected to 10 times more stress. The highest stress occurs on the y-radius of the center piece (See figure 5.31).

The results of the simulations are listed in the table below (See table 5.2). The values were standardized since the load cases were only meant to be compared to each other. The simulations made it clear that load case 1 was the extreme case, since the other load cases were exposed to 10 times less stress. The other load cases may be affected differently when using other loads. After discussions with the supervisor at Volvo, it was concluded that further simulations will be made with load case 1.

Table 5.2 – Standardized values of stress, calculated in Creo Simulate

Load cases Standardized values of stress

Load case 1 100 Load case 2 9,6 Load case 3 8,2

Simulations with load case 1

The next step of the implementation of the method was to investigate how much the stiffness, weight and stress differs from the existing cross members. The goal with the newly developed cross members was that they would be more flexible and lightweight than the existing ones, and still be comparable in terms of performance.

There are a total of 15 cross members in the new module system, both with and without inner-liner included. It would take too much time to simulate them all, the only thing that differs between an inner-liner compared to a non-inner-liner version is that the rivet joint is displaced 5 mm. Three modules that had undergone the most drastic changes from the current cross members were chosen for further simulations. They are listed in the table below with specific reasons to why they were chosen (See table 5.3).

Figure 5.31 – Stresses on the top of the cross member

(57)

Table 5.3 – Modules that were chosen for further simulations

Modules chosen for further simulations

Reasons of why the modules were chosen

Module 1  WCP of the centerpiece was shortened by 14 mm to enable compatibility with module 2.

 Tie plate bTP have increased with 7 mm.

 This is the lightweight cross member in the module system

Module 3  WCP of the center piece was shortened by 75 mm to enable compatibility with module 4 and 5.

 Tie plate bTP length have increased with 23 mm

Module 6  This is the heaviest cross member in the module system.

 Due to the new rivet joint, bTP of the tie plate have increased by 5,5 mm

5.8 Change of material and thickness

One requirement was to investigate whether it´s worth increasing the material quality. When improving the material quality it´s possible to reduce the thickness of the

components and thereby also weight. By doing this change the yield limit increases but the stiffness decreases. There are ways of compensating loss of stiffness but during this investigation only the thickness will be changed. The following materials were chosen for investigation; DOMEX 460 (original) and DOMEX 550 (new material). According to SSAB the price increases about 10% when the material quality is improved. Specifications for the materials can be found at SSAB´s website (SSAB, u.d.).

As discussed in geometrical relations the distance between mounting holes is affected by change of thickness (see chapter 5.6.1). Therefore the height of the center pieces (hCP) need to increase since tie plates have been reduced in thickness.

The thickness reduction can be seen in the table below (see table 5.4). All values are in millimeters

Table 5.4 – Change of thickness and hCP

Modules _Center

(58)

(59)

6 Results

This chapter presents the data that were collected and analyzed.

6.1 Reduction of components in Volvos cross member portfolio

By implementing the method described in chapter 5, the selected components from

Volvo’s cross member portfolio were reduced from 31 to 18. 15 tie plates were reduced to 6, and 16 center pieces were reduced to 12. The conclusion of these results is that Volvo has opportunities of reducing their components, and by doing so establish more flexible cross members. The complete pairing can be found in appendix (see appendix A.2)

6.2 Consequences of optimization

When combining technical articles, it’s likely that there will be consequences with other parts. Some of them are easy to fix, while others require more challenging work in order to fulfill the demanded functions.

The simplest way to reduce the number of articles is to make the tie plates symmetrical. This causes a change of position of the web, which results in that all the components mounted on the cross member must be moved according to the new web position. In the future, symmetrical cross members can be negotiated with other members within the chassis group, and thus implement a standardized way to make the mounted parts less dependent of the cross member.

Note that some existing parts are made smaller and some are made larger. This is to pair them together in an optimal way. Often, it’s the parts that are manufactured in low volumes that are subjected to changes. These changes may contribute to a stronger or weaker part.

Also note that all cross members are made with either three or seven rivets at the rivet joint. This is to get two consistent and durable rivet joints to choose from. By having fewer rivet joints it results in easier change of components if there has to be an exchange. There were a total of 5 different rivet joints from cross members that were selected in the pairing process.

One of the major consequences occurred with module 4, and was not discovered until the new components had been created and assembled. The original cross member that was modified for the creation of module 4 consisted of a center piece with three different tie plates. Two of the tie plates were asymmetrical with 9 holes, and the other was

(60)

The problem was solved by displacing the lower and upper rivet joints of the center piece (see figure 7.2). This required more material since the lower rivet joints had to be

displaced further out towards the edge.

The supervisor at Volvo clarified that asymmetrical tie plates were more preferable (even though the new module decreased the portfolio with one component) than having

displaced rivet joints of the center piece. The reason for this is that it may cause problems in terms of performance, and the cross member becomes heavier because of the increase of material.

Figure 7.2 – Symmetrical tie plates with even and uneven holes. Center of the rivet joint is displaced on the center piece.

Module 2

After meetings with the supervisor and employees at Volvo it was declared that module 2 will never be implemented, since the changes of the tie plates resulted in that the rear suspension has to be reconstructed from scratch. Module 2 is still included in the report as it shows the train of thought of how it is possible to modularize (see appendix A.2).

Description of the consequences that the modularization resulted in can be found in appendix (See Appendix A.3)

(61)

6.3 Comparison - Volvo’s existing portfolio

In order to determine whether the new cross members were improved, a comparison of the original portfolio was made. Stiffness, max stress, margin until yield point and weight of the new modules was specified as a percentage change of the original cross member. Each module is compared to its original parts. Stiffness and weight calculations can be found in appendix (see appendix A.4-A.6)

Module 1 6.3.1

Thick

The shortening of the center piece resulted in a lower weight than the original since the tie plate have the same rivet joint as before. The shortening of the center piece also resulted in a 4% higher stiffness. It turned out that the new cross member also had a 3% higher “margin until yield point”, which means that it´s stronger in terms of performance than the original cross member.

Thin

This cross member had 16% less stiffness than the original version which is a bit too much. According to Volvo, the stiffness and stress can be 10-15% above/ below the original cross member. Otherwise, investigations must be made since it's too big of a change from the original version. As the components were 1 millimeter thinner, it resulted in 21% less weight. The cross member is however 14% under “margin until yield point”, which means that it would break before the original.

Table 6.7 – Standardized values of stiffness, max stress, yield point and weight Cross

members

Stiffness Max stress (center piece) Margin until yield point (REL) Weight (without rivets) Stiffness ≈ 100 = Good Max stress < 100 = Good REL > 100 = Good Weight < 100 = Good Original (existing) 100 100 100 100 Module 1 (Thick) 104 98 103 98,5 Module 1 (Thin) 84 130 86 79 Module 3 6.3.2 Thick

The existing cross member was originally 75 mm wider in wCP. The result of this change was that the new cross member became stiffer. The idea was that the new cross member would be placed at both boogie and after transmission.

(62)

These changes made the new cross member more comparable with the original cross member but it was still a bit too stiff.

Thin

The thin version of module 3 showed good results in terms of weight, since it decreased with 16% in comparison with the original cross member. It´s also 4% over the “margin until yield point”. The reason why it became strong, yet light, was because of the decrease in wCP which increased stiffness and lowered the stresses.

members

Stiffness Max stress (center piece) Margin until yield point (REL) Weight (without rivets) Stiffness ≈ 100 = Good Max stress < 100 = Good REL > 100 = Good Weight < 100 = Good Original (existing) 100 100 100 100 Module 3 (Thick) 120 93 106 101 Module 3 (Thin) 103 114 104 84 Module 6 6.3.3 Thick

The cross member became heavier than the original version due to added material in the new tie plate, even though material was removed from the center piece. The modifications made the center piece stiffer but more harmful to stress, which appeared in the

simulations. Thin

The thin version of module 6 showed better results in both weight and performance properties than the thick version did. 21% lighter and 1,5% under the “margin until yield point” which is acceptable.

members

Optimization of the modular cross member portfolio

Optimization of the modular

cross member portfolio

Optimization of the modular

cross member portfolio

av

Anton Vendelstrand

Daniel Pettersson

Preface

Nomenclature

Table of contents

1 Introduction

1.1 Background

1.2 Requirements specification

1.3 Goals

1.4 Project limitations

1.5 Methodologies

1.6 Expected results

2 Current situation

2.1 Volvo Trucks

2.2 Implemented standardizations

3 Theoretical frame of reference

3.1 Project process

3.2 What is a module?

3.3 Load cases

3.4 PUGH’s matrix

3.5 FMEA

3.6 Design tools

3.7 Relevant subjects

4 What is a cross member?

4.1 Chassis

4.2 Frame

4.3 Tie plate

4.4 Center piece

4.5 Rivets

4.6 Manufacturing method

5 Implementation

5.1 Choice of method

5.2 Product decomposition

5.3 Implementation of method

5.4 Models

5.5 Motivation for standardization

5.6 Geometrical relations

𝐿

=

− 𝐿

, 𝐿

, 𝐿

, 𝐿

𝐿

=

− 𝐿

5.7 FE-analysis

5.8 Change of material and thickness

6 Results

6.1 Reduction of components in Volvos cross member portfolio

6.2 Consequences of optimization

6.3 Comparison - Volvo’s existing portfolio