Design of a cargo fastening device

(1)

Design of a cargo fastening device

with a load indicator for heavy cargo

Konstruktion av lastspänningsanordning med lastindikator för tung last

Niklas Stenqvist

Fakultet: Fakulteten för hälsa, natur- och teknikvetenskap Kurs: Examensarbete för civilingenjörsexamen i maskinteknik Omfattning: 30 hp

Handledare: Anders Gåård Examinator: Jens Bergström 2014-05-23

Serial number

(2)

Abstract

The aim of this thesis was to develop a new turnbuckle for fastening containers on cargo ships. The design was supposed to indicate whether the turnbuckle could be re-used, i.e. it hadn’t plastically deformed. This indication could reduce the risk of a turnbuckle breaking during transport, since turnbuckles that have been exposed to a load above their safe working load would indicate this.

In total 25 different concepts were developed and put through two different elimination matrices which were evaluated against a requirement specification. The final design had calculated stress of 630 MPa at breaking load and 472 MPa proof load, the loads are defined by Germanischer Lloyd [16]. A concept for measurement of the deformation of the turnbuckle was developed, however it is in need of further development and testing before it can be implemented. A suggestion of manufacturing method, material and surface treatment has been given, but prototype testing is required to verify the design and ensure adequate corrosion protection.

(3)

Sammanfattning

M˚alet med detta examensarbete var att utveckla en ny spännare till containrar p˚a last- fartyg. Den färdiga konstruktionen skulle indikera om spännaren kunde användas för ytterligare en resa; d.v.s. den hade inte plastiskt deformerats. Denna typ av indikation kan reducera risken att en spännare g˚ar sönder under transport, eftersom spännare som har utsatts för en last över sträckgränsen indikerar detta.

Totalt 25 koncept togs fram och utvärderades, genom tv˚a olika elimineringsmatriser, mot en kravspecifikation. Den slutgiltiga konstruktionen hade en högsta beräknad spänning p˚a 630 MPa vid ”breaking load” och 472 MPa vid ”proof load”, laster som definieras av Germanischer Lloyd [16]. Ett koncept för mätning av deformationen av spännaren togs fram, men den kräver ytterligare utveckling och testning innan den är redo för användning. Förslag p˚a tillverkningsmetod, material och ytbehandling har presenterats, men testning av en prototyp krävs för att verifiera att konstruktionen är tillfredsställande och att den har adekvat korrosionsskydd.

(4)

Acknowledgements

First and foremost I want to thank my supervisor at Technogarden Engineering, Mr Martin Niklasson, for all the help, guidance and valuable discussions during this thesis work.

I would also like to thank Mr. Anders G˚a˚ard for his guidance on how to write the report and the discussions around the thesis work.

Thanks to Ms. My Andersson, Mr. Alexander Jonsson, Mr. Rickard ˚Akerfalk, Mr.

Johan Sandgren and Mr. Martin ¨Ostberg for their valuable input during concept evaluation and elimination.

Thanks to Mr. Lars Jacobsson for helping me with printing printing a prototype in the 3D-printer.

Lastly I want to thank Mr. Jon H¨ogblad for his input and help with all the calculations in Ansys and for the valuable discussions around the results. Without the help and discussions this thesis work would not have been possible.

iii

(5)

List of Figures

1.1 Schematic drawing of standardized container sizes . . . . 2

1.2 How some of the lashing equipment is used[7]. . . . . 2

1.3 The different movements of a ship[9]. . . . . 3

2.1 The half model used in calculations. . . . 9

2.2 The different bodies of the turnbuckle . . . 10

2.3 Constraints and forces acting on the body and rod . . . 11

3.1 The first concept for the body. . . 15

3.2 The second concept for the body . . . 16

3.3 The first concept for the body. . . 16

3.4 The fourth concept for the body . . . 17

3.5 Concepts for the rodholder . . . 17

3.6 The sliding axis within the cylinder. . . 18

3.7 Concept A, the additional concept . . . 21

3.8 The mesh of the simulated turnbuckle . . . 22

3.9 Stresses in concept A. . . 22

3.10 Stresses in the two different versions of concept A at breaking load . . . . 23

3.11 Deformation along the Y-axis at breaking load . . . 23

3.12 Stresses of the final design at SWL & BL . . . 25

3.13 A 3D printed plastic prototype of the final design . . . 26

3.14 A 3D printed plastic prototype of the final design . . . 27

B.1 Example of elimination matrix based on Pahl and Beitz[15] . . . 42

vii

(9)

List of Tables

2.1 The main tasks of the project and the estimated time taken to finish each

task . . . . 6

2.2 The aspects of the different categories . . . . 7

2.3 Detailed concept evaluation . . . . 8

3.1 Requirement specification . . . 14

3.2 A matrix over the different sub-solutions . . . 19

3.3 Elimination matrix . . . 20

3.4 Detailed concept evaluation . . . 21

3.5 Required impact strengths . . . 24

3.6 Mechanical characteristics of untreated 42CrMo4 steel.[19] . . . 25

A.1 Work breakdown structure. . . 40

viii

(10)

Abbreviations

TEU Twenty-foot Equivalent Unit

FFE Forty Foot Equivalent (Also known as FEU) GL Germanischer Lloyd

SWL Safe Working Load PL Proof Load

BL Breaking Load

ix

(11)

Chapter 1 Introduction

1.1 Background

1.1.1 Technogarden Engineering Resources AB

Technogarden Engineering Resources AB is a technical consultant company and the work in this thesis was carried out in their office in Karlstad, Sweden. They are a part of the Norconsult AS group, which has their main office in Norway. Technogarden Engineering Resources AB was founded in 2003 and specializes in technical advising, developing organisations and recruiting.

1.1.2 Problem description

In 2012 about 600 million twenty-foot equivalent unit (TEU¹) of goods were transported on ships [1]. A schematic figure of these containers can be seen in figure1.1. The average value of a forty-foot equivalent (FFE²)container , which is the same as two TEU, was USD 2.678 according to Maersk Line [2]. The new tripple-E class ships of Maersk Line are capable of carrying 18340 TEU, giving a full shipment a value of about 49 million USD [3].

The containers are loaded in port and stacked onto each other, sometimes as high as

1A twenty-foot equivalent unit is 20’ x 8’ x 8’6” (Length x width x height) according to ISO 668:2013.

2A forty-foot equivalent is 40’ x 8’ x 8’6” (Length x width x height) according to ISO 668:2013.

1

(12)

Chapter 1. Introduction and specification of problem 2

Figure 1.1: Schematic drawing of standardized container sizes

Figure 1.2: How some of the lashing equipment is used[7].

10 tiers and over 20 containers in width. To secure the containers onto the ship several different types of lashing equipment is used, a schematic view of how some of the equipment is used can be seen in figure1.2.

According to a small survey conducted by the World Shipping Council among their members, between 350 and 675 containers on average are lost to the sea each year[4].

These accidents happens without the one in charge, the captain, getting a warning that they’re about to happen. Loss of tension in, failure of or too high compressional loads can lead to lost cargo or injuries to the working staff [5], [6].

(13)

Figure 1.3: The different movements of a ship[9].

However, even though safety precautions are taken and the cargo is secured by profes- sionals, the wind force in combination with the ship movements (defined in figure 1.3) can overstress the lashings and lead to failure. In bad conditions a phenomenon called parametric rolling can occur, producing large roll amplitudes [8]. Parametric rolling is a phenomenon were the ship experience large rolling amplitudes, inducing high stresses on both the hull of the ship as well as the equipment securing the containers on deck [10]-[13]. While the ship is moving with either pure head or following seas or at a small angle (<5 degrees), with a low encounter frequency of waves, the buoyancy forces will further develop the rolling instead of dampening it [14]. In some cases the roll angles have been as high as 40 degrees combined with wind speeds of up to 30 m/s, with loss of containers as a consequence [8].

1.1.3 Current equipment

The currently used equipment, which the new design should be compatible with, consist of twist-locks placed in the corner of the containers and lashing rods/bars that are fastened next to the twist-locks. The turnbuckle in this thesis should be compatible with the lashing bars used today and with the threaded corer fitting that connects to the opposite corner of the lashing bar.

(14)

1.2 Project specification

While securing heavy cargo on for instance a ship, a truck or on trains there is a risk that the fastening device might break or lose tension. This can result in loss of cargo, damage to property or, in a worst case scenario, injury to people. There are currently no cargo-fastening devices that indicate if the tension is above the appropriate level or if the tension is lost completely.

1.2.1 Definition of problem

Secure heavy cargo with adequate force and maintain it until it’s time for unloading.

While the cargo is secured, the force should be observable at all times to make sure it is adequate.

1.2.2 Purpose

To find a solution that ensures adequate tension of cargo until unloading.

1.2.3 Goal

The goal of this project is to deliver several concepts and a working design for a new kind of fastening devices for heavy cargo. The device should be fit for trucks as well as trains and ships. The device should also indicate when the tension is higher than intended or when the tension is lost through a sensor.

(15)

Chapter 2 Method

2.1 Project planning

The first step of the project was to identify which work packages should be included and which dependencies there was between them. This work breakdown structure (WBS) can be seen in appendixA. Following this a time estimation of each task was performed in order to fit the allotted time for this thesis, which is 30 ECTS credits or 800 hours of work.

2.1.1 Work Breakdown Structure

The project was divided into the main tasks of a design process, as defined by Johannes- son et. al.(2004) in their book [15]. The main tasks of the WBS can be seen in table2.1 and each of these main tasks are divided into subtasks. Some additional main tasks were identified, such as patent preparation/application and three different presentations, one held for the receiver of the project, Technogarden Engineering Resources AB, and two for the project advisor and examiner at Karlstad University. A quarter of the time of the project was spent on writing the report and the second largest entry, Evaluation of design, included a ’design-loop’ to ensure that the final design met the criteria listed in the requirement specification.

5

(16)

Chapter 2 Method and theory 6

Table 2.1: The main tasks of the project and the estimated time taken to finish each task

Task Name Work

Start-up 36 hrs Pre-study 98 hrs Report 204 hrs Concept development 80 hrs Patent preparations 12 hrs Design phase 72 hrs Half time presentations 12 hrs Finalization of design 84 hrs Preparation for manufacturing 68 hrs Evaluation of design 104 hrs Final presentation 30 hrs Sum 808 hrs

2.2 Problem refinement

The initially identified (see chapter1) problem was reviewed to achieve a more accurate and detailed description of the problem. This was done by first expanding the problem and generalize it to fit all kinds of lashings of goods, ranging from heavy containers on ships to securing of small boats at a pier. It was then divided into two main categories of interest and each of these was further divided into detailed areas. The reason for this was to make sure that the correct problem was solved, as changes in the definition of the problem late in the design phase will be costly compared to changes in the start according to Johannesson et. al.(2004) [15]. This eased the task of identifying potential stakeholders and people to interview that could help narrowing down the problem even further.

2.2.1 Requirement specification

The identified requirements of the part were divided into Demands and Wishes. The wishes were weighted between 1 to 5, where 5 was the most important wish and 1 was the least important. The criterias were also divided into five different categories including Design, Operation, Function, Manufacturing and Life cycle. The different aspects included in each category are presented in table2.2below.

(17)

Table 2.2: The aspects of the different categories

Category Aspects included

Design Mechanical properties, environmental conditions

Operation Customers’ use of product

Function The function of the product

Manufacturing Includes environmental effects and cost

Life cycle Recyclability

2.3 Concept generation

In order to ensure that the final product meets the requirements listed in the requirement specification the concept generation process was divided into five different steps, as suggested by Johannesson et. al.(2004) [15].

• Formulate the problem in a broader, abstract, solution-neutral form

• Conduct a function analysis, which divides the main function of the product into sub-functions

• Seek solutions to the sub-functions

• Combine the alternatives to the sub-functions into a solution

• Sort out potential final solutions

When the sub-problems, which the sub-functions should solve, have been identified and given several solutions all the different sub-problems are placed in a matrix. In this matrix, the first column represents the sub-functions of the final product and the rest of the columns are filled with different proposed solutions to each sub-function. The next step is to combine the different solutions into a potential final solution and to sort out the unreasonable combinations and through that reduce the potential solutions to a smaller number.

The next step is to make use of an elimination matrix and for this thesis Pahl and Beitz elimination matrix was chosen as initial method for sorting and evaluation of concepts [15]. An example of this elimination matrix is shown in appendix B.1. The concepts were then given a plus or minus sign, depending on if they were believed to fulfil the criteria or not. A plus sign gave a value of +1 and a negative sign gave a value of -1

(18)

and all of the concepts with a total value of four or above were put through to the next round for another, more detailed elimination matrix.

The detailed elimination matrix was evaluated with the help of a reference group and an example of the matrix can be seen in table 2.3.

Table 2.3: Detailed concept evaluation

Concept → 1 2 3 (ref) 4 5 6

Demands ↓

Design - robust 0 0

DATE

0 0 0

Design - ergonomic + + 0 - -

Operation 0 + + 0 +

Manufacturing 0 0 0 + +

Function + + 0 - -

Storing - - 0 0 0

Sum + 2 3 1 1 2

Sum 0 3 2 5 3 2

Sum - 1 1 0 2 2

Net value 1 2 0 1 -1 0

Rank 2 1 4 2 6 4

Further development No Yes No No No No

The limit for passing to the next round was set to a net value of 2 by the reference group.

2.4 Standards

For the design of the turnbuckle standard ISO-3874 and the Germanischer Lloyd (GL) classification has been used [16]. Both ISO-3874 and GL has explicit loads to be used in calculations regarding the strength of the turnbuckle, however the loads from GL are higher and therefore they’ve been used in calculations in this report.

(19)

2.5 Calculations

To ensure structural integrity of the final design, a Finite Element Analysis was carried out in Ansys. Two parts were included in the analysis, the turnbuckle and a represen- tational model of a lashing bar. To allow for a finer mesh, while keeping the computing time at a low level, a symmetrical model was used. The entire model was divided along a plane intersecting the centre axis, as can be seen in figure2.1.

Figure 2.1: The half model used in calculations

2.5.1 Mesh

The use of a plane-symmetric method resulted in 231943 elements and 936235 nodes.

The different colors in figure 2.1 represent the sliced bodies used for meshing. Each body has its own meshing strategy, to allow the mesh to be optimized for each part of the body since they have different difficulties while meshing. In total there are thirteen different bodies, which can be seen in figure2.2. The different techniques used are:

1. Swept with 20 elements in the Z-direction, 2mm face sizing 2. Swept with 10 elements in the X-direction, 3mm face sizing 3. Body sizing of 1mm, hex dominated mesh

4. Body sizing of 2mm, hex dominated mesh 5. Body sizing of 3mm, hex dominated mesh

(20)

6. Body sizing of 3mm, hex dominated mesh 7. Body sizing of 1mm, hex dominated mesh

8. Swept with 10 elements in the Z-direction, 4mm face sizing 9. Body sizing of 1mm, hex dominated mesh

10. Body sizing of 1mm, hex dominated mesh

11. Swept with 10 elements in the Z-direction, 1mm face sizing 12. Swept with 10 elements in the Z-direction, 1mm face sizing 13. Body sizing of 3mm, hex dominated mesh

Figure 2.2: The different bodies of the turnbuckle

2.5.2 Contacts

Two different setups were used regarding the contacts of the parts. In the first setup all contacts were modelled as bonded, meaning that they’re treated as a single body.

In the second setup, all those bodies which are part of the turnbuckle were modelled as bonded. The contacts between the rod and turnbuckle, however, were modelled as frictionless contacts, with augmented lagrange formulation and the interface treatment was set to adjust to touch.

(21)

Chapter 2 Method and theory 11 2.5.3 Constraints and applied forces

During the simulations the force was applied on the end surface of the rod, 3 in figure 2.3. It was scaled to half of the real value since the model is symmetrical. To ensure that the force was only acting in the Z-direction a remote displacement was put on the entire surface area of the rod. The remote displacement locked the rod in all directions and all rotations except for the Z-direction. A fixed support was added to the threaded part of the turnbuckle, 1 in figure2.3, acting as the connection between the turnbuckle and the container.

Figure 2.3: Constraints and forces acting on the body and rod

(22)

Chapter 3 Results

3.1 Problem Identification

During the pre-study and literature study phase of the project the problem was slightly altered and re-redefined from the previously presented. It was then defined as:

Identification of turnbuckles which have been exposed to a load above their yield strength.

3.1.1 Identified sub-problems

As mentioned in chapter2, the problem was divided into smaller, manageable problems.

These were identified from evaluating the current lashing equipment and through the study of incident reports regarding container ships. The sub-problems that were found are presented below:

• A body designed to withstand tensional loads

• Transfer load from lashing rod to turnbuckle

• Transfer load from turnbuckle to corner fitting of container

• Measure pre-tensioning force

• Measure if the SWL (Safe working load) has been exceeded 12

(23)

Chapter 3. Results 13

These were then concretized and described with two words, a predicate and a subject, as described by Johannesson et. al.(2004) [15]. The two last sub-problems were then combined into one single problem, since they can make use of the same solution. The two-word problems are listed below:

• Withstand load

• Transmit load (turnbuckle to rod)

• Transmit load (turnbuckle to container corner)

• Measure force

3.1.2 Requirement specification

The requirement specification, divided into the five categories mentioned in chapter 2, totalled 22 criteria. The requirement specification, and the weights of the wishes, is presented in table3.1.

(24)

Table3.1:Requirementspecification Criteriano.CategoryCriteriaDemand=DWish=W 1DesignCorrosionresistantD 2DesignMaintainadequatemechanicalpropertiesfrom−20◦ Cto+60◦ CD 3DesignComplywithstandardISO3874D 4DesignComplywithGermanischerLloyddemandsD 5OperationWeightbelow15kg[17]W,5 6OperationEasytofastenusingbothhandsW,5 7OperationExchangable’Weakest-link’-partW,4 8OperationAlifetimeof5millionload-cyclesD 9OperationWithstandafallof2mD 10OperationCompatiblewithcurrentlashingequipmentD 11OperationErgonomichandlefortensioningW,3 12OperationTensioningtobedonewithoutreleasingthedeviceW,2 13OperationNotbulkyW,4 14FunctionMeasuretensioninequipmentD 15FunctionMeasurecompressioninequipmentW,5 16FunctionMeasurementofsmallforcesD 17ManufacturingLowcostformanufacturingD 18ManufacturingUseenvironmentallyfriendlymethodsW,5 19ManufacturingDesignsuitableformass-productionD 20LifecycleAtleast60%oftheproductshouldberecyclableD 21LifecycleAtleast85%oftheproductshouldberecyclableW,4

(25)

3.2 Concepts

Following the methodology described in section2.3and the identified sub-problems from section3.1.1several different concepts were generated for each problem. They were then combined to create a working concept for the main problem.

3.2.1 The body

Four different concepts were created for the body of the turnbuckle, each with its own pros and cons and they are all presented below.

3.2.1.1 Concept 1

The first concept for the body of the turnbuckle is an elliptical, egg-shaped, body designed with the force measurement in mind. The idea was to be able to measure the decreasing distance between the two sides and therefore know the elongation in the turnbuckle. A model of this can be seen in figure3.1below.

Figure 3.1: The first concept for the body

(26)

3.2.1.2 Concept 2

The second concept is of a closed, rectangular shape. This simple geometry ensures easier manufacturing compared to concept 1 and it’s easy to adjust according to each customers’ wishes. A figure of this design is shown in figure3.2below.

Figure 3.2: The second concept for the body

3.2.1.3 Concept 3

Concept 3 is similar to concept 2, with the difference that it’s thinner and instead has wider parts on the sides to ensure enough stiffness is achieved. The wider parts also ensures that a good grip can be achieved while tensioning the turnbuckle. It’s shown in figure3.3.

Figure 3.3: The first concept for the body

(27)

3.2.1.4 Concept 4

The fourth and last concept of the body is a combination of the first and third concepts, it has an elliptical shape with thin (relatively), rectangular segments connecting the two ends of the turnbuckle. There are 3 segments connecting the both ends, as seen in figure 3.4, and it’s possible to use two of these to place the force measuring devices on.

Figure 3.4: The fourth concept for the body

3.2.2 Concepts for rod holder

In order to achieve tension between the rod and the opposing corner of the container, a part of the turnbuckle has to hold the rod in place. Three different concepts for this part of the turnbuckle are presented below in figure 3.5. The three concepts are somewhat similar, but with a difference in how to connect the rod.

The first concept for the rod holder has four connecting half-circles on one side of the turnbuckle, which ensures easy fitting of the rod to the turnbuckle. The second concept

(a) Concept A (b) Concept B (c) Concept C

Figure 3.5: Concepts for the rodholder

(28)

has two half-circles on each side of the turnbuckle and four in total. This will make sure that the turnbuckle stays in place without the users having to hold it in place. The third concept has every other half-circle on one side and every other on the other side.

This complicates the mounting of the rod to the turnbuckle, but when it’s mounted it has the best support out of the three concepts.

3.2.3 Concepts for measuring of the force

Two different concepts for the measurement of the force in the turnbuckle have been developed. One is pure mechanical and the other one is based on electronics and strain gauges. The idea behind the mechanical concept is to get rid of the sensitive electric equipment to make the final product more robust. The mechanical concept is based on an axis sliding within a cylinder, which moves as the body deforms. This requires a good knowledge of the stiffness of the design, whereas the use of strain gauges measure the actual strain in the body. The axis will have three different color markings (see figure 3.6) to tell the user when the turnbuckle isn’t pre-tensioned enough, a part where the pre-tensioning force is adequate and a part where it indicates that the SWL has been exceeded. The colors chosen are according to Arbetsmilj¨overkets regulations, to ensure that they’re easy for the user to understand[18].

Figure 3.6: The sliding axis within the cylinder

3.2.4 Concepts for transferring the force to the container corner

The concept for transferring the force from the turnbuckle to the corner of the container is based on the current equipment. It has a heel that fits into the space of the corner and in the other end there is a threaded part that fits the turnbuckle.

(29)

Chapter 3. Results 19 3.2.5 Combined concepts

With all of the above described concepts, a matrix of possible solutions can be generate, seen in table3.2below. The amount of total solutions that can be generated with these sub-solutions are 24 and they are put into an elimination matrix, appendix B.1, for further narrowing down of the concepts.

Table 3.2: A matrix over the different sub-solutions

Sub-problem Sub-solution #1 Sub-solution #2 Withstand load Egg-shaped Rectangular Transmit load (turnbuckle to rod) All up Half up, half down Measure force Axis-cylinder Strain gauges Transmit load(turnbuckle to container) Threaded part

Sub-problem Sub-solution #3 Sub-solution #4 Withstand load Wide rectangle Elliptical shape Transmit load (turnbuckle to rod) Every other up

Measure force Transmit load(turnbuckle to container)

3.2.5.1 Elimination of combined concepts

The elimination matrix had six different criteria to evaluate which concepts that were thought to be possible solutions to the problem. The elimination matrix, and explana- tions of the concepts, can be seen in table 3.3below.

Every concept has been given a two letter abbreviation and they’re listed below:

• EG - Egg-shaped body

• RE - Rectangular body

• TR - Thin rectangular body

• EL - Elliptical body

• AU - All rod-connectors under

• HU - Half of the rod-connectors under

• EO - Every other rod-connector under

(30)

• AC - Axis-cylinder measuring

• SG - Strain gauges

• TH - Threaded part with heel

Table 3.3: Elimination matrix Version: 1.0

Concepts: Solvestheproblem Fulfillsalldemands Realizable Manufacturability Reliability Manageability Decision

1 EG + AU + AC + TH + + + + + + +

2 EG + HU + AC + TH + + + + + + +

3 EG + EO + AC + TH + - + + + - -

4 EG + AU + SG + TH + - + + - + -

5 EG + HU + SG + TH + - + + - + -

6 EG + EO + SG + TH + - + + - - -

7 RE + AU + AC + TH + + + + + + +

8 RE + HU + AC + TH + + + + + + +

9 RE + EO + AC + TH + - + + + - -

10 RE + AU + SG + TH + - + + - + -

11 RE + HU + SG + TH + - + + - + -

12 RE + EO + SG + TH + - + + - - -

13 TR + AU + AC + TH + + + + + + +

14 TR + HU + AC + TH + + + + + + +

15 TR + EO + AC + TH + - + + + - -

16 TR + AU + SG + TH + - + + - + -

17 TR + HU + SG + TH + - + + - + -

18 TR + EO + SG + TH + - + + - - -

19 EL + AU + AC + TH + + + - + - -

20 EL + HU + AC + TH + + + - + - -

21 EL + EO + AC + TH + + + - + - -

22 EL + AU + SG + TH + - + - - - -

23 EL + HU + SG + TH + - + - - - -

24 EL + EO + SG + TH + - + - - - -

After the general selection of the initial concepts, a more detailed concept evaluation was conducted which ended in concept 7 being chosen for further development. The evaluation of the six concepts can be seen in table 3.4below.

(31)

Table 3.4: Detailed concept evaluation

Concept → 1 2 7 (ref) 8 13 14

Demands ↓

Design - robust 0 0

2014-04-15

0 0 0

Design - ergonomic + + 0 - -

Operation 0 + + 0 +

Manufacturing 0 0 0 + +

Function + + 0 - -

Storing - - 0 0 0

Sum + 2 3 1 1 2

Sum 0 3 2 5 3 2

Sum - 1 1 0 2 2

Net value 1 2 0 1 -1 0

Rank 2 1 4 2 6 4

Further development No Yes No No No No

3.2.6 Additional concept

During the analysis of the chosen concept, another concept was developed. It was a combination of concepts 2 and 8 which makes use of the two strengths of each concept.

From concept 2 the egg-shaped part was taken and from concept 8 the rectangular body, with rectangular cross-sections was taken. They were combined into a single concept where the part that connects the rod to the turnbuckle was from concept 8 and the opposite end is from concept 2. A figure of this concept, from here on called concept A, can be seen below, figure3.7.

Figure 3.7: Concept A, the additional concept

(32)

3.3 Calculations

As mentioned in2.5.1, 231943 elements and 936235 nodes were used in the calculations.

The areas with highest stresses had a higher concentration of elements, see figure3.8a.

However in the contact area between the innermost knob and the turnbuckle some singularities arose. These elements are shown in figure3.8b.

(a) Mesh concentration in the high-stress area

(b) Some of the singular elements in the contact area between the knob and the turnbuckle Figure 3.8: The mesh of the simulated turnbuckle

During the calculations, concept 7 did not meet the requirements listed by GL [16]

without modifications. This contributed to an increased weight to the point where it did not meet the weight requirement. However, the combined concept, concept A showed good potential both for withstanding the required forces and measuring the deformation.

Therefore, this concept was further developed and an arc was added connecting the two sides where the bending force was at its largest. A picture of the stresses in concept A with focus on the bending stresses can be seen in figure3.9.

Figure 3.9: Stresses in concept A

(33)

From figure3.9it can be clearly seen that the stresses in the first body are above the yield strength of the chosen material (see chapter 3.4). To solve this problem two different solutions were tested and analysed. The first involved an arc that was added on both sides of the body to decrease the bending stresses, causing the problems. The results of this solution can be seen in figure3.10a. The second geometry had a thicker wall where the bending stresses were at their highest. The result from the changed geometry can be seen in figure3.10b.

(a) Stresses of concept A with the arc

(b) Stresses of concept A with thicker wall

Figure 3.10: Stresses in the two different versions of concept A at breaking load

3.3.1 Force measurement

To measure the deformation, the axis-cylinder concept was placed where the deformation was at its highest, see figure 3.11. The deformation of the part in the Y-direction was simulated to 0,73mm at breaking load and 0,55mm at proof load.

Figure 3.11: Deformation along the Y-axis at breaking load

3.4 Material selection and manufacturing

The demands of the material in the turnbuckle, as described by the Germanischer Lloyd classification [16] states that material must be made out of steel and fulfill the re- quiremets listed below.

(34)

• The steels shall be killed, i.e. completely deoxidized, and fine grain treated

• All products shall be heat treated, that means normalised or quenched and tem- pered.

• The steels shall fulfil the requirements for impact strength mentioned in the Stan- dards and approved specifications respectively, at least fulfil the requirements mentioned in Table3.5.

• Unalloyed steels intended for welding shall not have a higher carbon content than 0.22 % (ladle analysis)

• If the type of product requires it, additional non-destructive test can be required.

The table mentioned in the requirements (table 3.5) is regarding the impact strength of the material at certain temperatures. Since the turnbuckle is to be used above deck, the lower working temperature to be expected is −20^◦C [16]. Furthermore the material

Table 3.5: Required impact strengths

Product from

Impact energy KV 1

[J]

min

longitudinal transverse Rolled products R_emin ≥ 235N/mm² 27 (19) 20 (14) Rolled products R_emin ≥ 355N/mm² 34 (24) 24 (17)

Forged steels 27 (19)

Cast steels 27 (19)

should be able to withstand corrosion, as specified in the requirement specification (table 3.1). There are several different ways of ensuring corrosion resistance of the final product, but for this project two different concepts were evaluated. The first concept was having a corrosion resistant bulk material (e.g. stainless steel) and the second was having a non-corrosion resistant material in the bulk and then applying a surface treatment in order to protect the material. However, since no stainless steels are recommended in GL’s rules for their classification of lashing devices [16], a cast steel was favoured. The chosen alloy was EN 10083 (42CrMo4) as it is a high-strength steel with good castability.

The final design of the turnbuckle was done with casting as manufacturing method, in

(35)

mind. This was to comply with the requirement that the product should be suitable for mass-production.

3.5 Chosen concept and its functionality

From the FEA-simulations it was concluded that concept A with the additional walls fulfilled the requirements specified in table3.1. The highest stress in the part when the breaking load of 490 kN was applied was 630 MPa, which is below the ultimate yield strength of the material, see table 3.6. The highest stresses at SWL and PL were 316 MPa and 472 MPa respectively, this can be seen in figures 3.12aand 3.12b.

Table 3.6: Mechanical characteristics of untreated 42CrMo4 steel.[19]

Tensile characteristics Hardness

Yield strength RP0.2 (MPa) Tensile strength Elongation Young’s modulus HV0.1

RP0.2 (MPa) Rm (MPa) At (%) E (GPa)

978 1050 16.5 201 356

(a) Stresses at safe working load

(b) Stresses proof load

Figure 3.12: Stresses of the final design at SWL & BL

The final weight of the turnbuckle, including the axis and cylinder was 12.5 kg which is within AMV’s recommendations for lifting by persons in unnatural positions. The final design is shown in figures3.13 and 3.14.

(36)

Figure 3.13: A 3D printed plastic prototype of the final design

Figure 3.14: A rendered picture of the final design with all its function

The deformation measuring concept of an axis going into a cylinder can be seen in figure 3.14, however this was not included in the first prototype as it complicated the

(37)

manufacturing process. The idea behind this concept is explained in chapter 3.2.3, but in short the idea is to push a colored cylinder with the axis. When the cylinder is pushed a certain, pre-determined, length, the user will be able to see a green marking in the small window of the bigger cylinder. If the cylinder is pushed too far, i.e. the turnbuckle has deformed more than it’s allowed to, the window will instead show red. This allows for easy recognition of over-strained turnbuckles which are not safe to use.

(38)

Chapter 4 Discussion

4.1 General discussion

4.1.1 Initial goal

The initial goal of the project (see chapter1.2.3) was to design a device that would fit trucks as well as trains and ships. However, during the project this goal was revised to only include large cargo/container ships. This was done in order to narrow down the problem, making it possible to finish the project within the allotted time. It also enabled time to be taken to do a thorough pre-study and to produce a single prototype in plastic, which wouldn’t have been possible if the device was to be compatible with all three vehicles. It was also decided, early on, that the turnbuckle should be compatible with the current equipment. This was to ease the introducing of this product on the market. It was thought that if the users of the turnbuckle had to buy completely new rods and container connections to be able to use the new turnbuckle, the market would be reduced. Therefore, compatibility with current lashing equipment was deemed necessary in order to make sure that there’s a market for the product when it was finished.

4.1.2 Problem

The next step was to identify the main problem with the current equipment and one of the main problems, since there are several reasons for loss of cargo, was the identification of faulty / over-strained equipment[6]. With the short times the ships are allowed to

28

(39)

Chapter 4. Discussion 29

spend in port, since longer turn-around times in ports means less money earned, the equipment might not always be inspected thoroughly enough. This can lead to inappro- priate equipment being used for another trip to save some time while in port.

When the main-problem had been identified and defined (see chapter 3.1) formulating the sub-problems were the next step of the project. Five different sub-problems were identified (3.1.1) and these eased the task of solving the main problem. This approach is solely based on the procedure described by Johannesson et. al.(2004) [15], which is thought to be a good way to avoid the problem of jumping to conclusions before identifying the main problem.

The first sub-problem was identified as A body designed to withstand tensional loads and was written this way to emphasize that the body shouldn’t have to experience compressional loads if compatible with today’s lashing equipment. If using a design of the turnbuckle and the lashing rod that is capable of resisting compressional loads it would be possible to relieve the turnbuckle on the other side of the container of some stress.

However, it might induce a problem with bending in the rod, since they’re quite thin compared to their length. This would introduce several new sub-problems to take into account while designing the turnbuckle and would at least require a change in the connection between the turnbuckle and the rod.

The second sub-problem concerns the transfer of the load from the lashing rod, which is connected to the container corner, to the turnbuckle. Three different concepts were developed for solving this problem, where the one with two semi-circles up and one down was favoured. This was to ease the task of connecting the rod with the turnbuckle as having all of the semi-circles on the same side of the turnbuckle would require the user to hold both the rod and the turnbuckle until some tensioning had been reached. The concept where every other semi-circle is up and every other is down was not deemed applicable to the design since it would’ve required a loose fitting between the rod and the turnbuckle. Without this loose fitting, it would be impossible to attach the rod to the turnbuckle. A loose fitting between the two parts would also decrease the contact area between the rod and the turnbuckle as the semi-circles would become smaller. With the decrease in contact area, the contact pressure would increase. Without reinforcement of the contact areas, or a material with a relatively, even for steels, high compressional strength, the contact pressure would lead to large deformations.

(40)

Chapter 4. Discussion 30

4.2 Discussion about the concepts

Early on in the thesis it was decided that focus was to be put upon developing a new body with the ability to measure forces acting on the fastening device. The idea of con- structing a additional part to be put upon the current equipment was discarded since it was believed to be hard to implement. The decision to design a new body also enabled the body to be adapted to a mechanical strain measurement, removing the eletronics such a strain gauges etc. The urge to remove the electronics existed because there oper- ational environment above deck on a ship crossing a sea are harsh in terms of electronics.

Since the turnbuckle only has to account for tension, considering the design of the current lashing bars, a body consisting of long, thin rods were the first idea to come to mind. Both concept 2 and 3 were inspired by this idea and the dimensions of them could’ve been adapted to the required loads. However, since there’s only tensional loads in them, the force measurement have to be along the axis of the force. This was thought to complicate the mechanical measurement of the deformation, therefore focus was put upon developing a body where the largest deformations would be somewhere easier to measure. This is what originated the ideas for concept 1 and 4, a curved body where the strain could be measured between the arms.

4.2.1 Pros and cons of body concepts

4.2.1.1 Concept 1

The strength of the first concept comes in its curved outer body and the large cross- sectional area required to withstand the loads. The curved arms enables the use of a mechanical strain measurer, as earlier mentioned. Its soft edges, with some additional rounding, would also make it suitable for e.g. casting.

However, to keep the radii of the outer curves at a value where the deformation is enough to measure, the body would have to either be wide and long or thin and short. A wide body would be bulky for the operators to handle and it would increase the weight of the turnbuckle. The thin and short body would introduce a problem with attaching the rod to the turnbuckle. It would only be possible to use the first two knobs on the rod, restricting the usable length of the rod. The reason for this that the rod and the

Design of a cargo fastening device