Evaluation of ground anchor for temporary shelters

(1)

IN

DEGREE PROJECT TECHNOLOGY, FIRST CYCLE, 15 CREDITS

STOCKHOLM SWEDEN 2018,

Evaluation of ground anchor for temporary shelters

Performance, design and alternative materials

PATRIK TER VEHN VIKTOR OLSSON

KTH ROYAL INSTITUTE OF TECHNOLOGY SCHOOL OF ENGINEERING SCIENCES

(2)

INOM

EXAMENSARBETE TEKNIK, GRUNDNIVÅ, 15 HP

STOCKHOLM SVERIGE 2018,

Utvärdering av markankare för temporära

flyktingbostäder

Prestanda, design och alternativa material

PATRIK TER VEHN VIKTOR OLSSON

KTH

(3)

Public Version

Confidential information has been changed or removed

(4)

Abstract

Better Shelter is a company, developing and providing module housing units for refugees in crisis areas. These module housing units are kept in place against wind and harsh conditions by 10 ground anchors, that are buried and tied to the frame. This bachelor thesis study and evaluate the mechanical behavior of these anchors, investigate the possibility to replace the current material from an aluminum alloy to a glass fibre-reinforced plastic and present a new and improved design.

Evaluation of the ground anchor wire is not relevant in this report.

The results are based on four simulations executed in the FEM program, ANSYS Workbench, including provided and custom made CAD models of the anchor for both materials. From relevant loading cases, static and alternating, with an amplitude force of 2000 N , relevant information was collected and analyzed. Relevant data in this case include equivalent and maximum principal stress as well as the safety factor and cyclic fatigue life.

The simulations showed that minor plasticity and damage will occur to the current anchor in both materials. This is debatable since the computer simulation is based on models that are often an unreal representation of reality. As an example in this case, the stress levels in sharp corners of the CAD model highly exceeds the levels that would be considered realistic. However, great improvements could be observed with the new anchor design in both materials regarding strength and reduced volume. This concludes that a change of design and material is motivated mechanically and economically. Improvements include lowering cost to 4.44 SEK per unit which is a reduction by 54.8% as well as reducing the weight from 29 g to 11.8 g per unit.

(5)

Abstract in Swedish

Better Shelter är ett företag som utvecklar och tillgodoser lättbyggda modulhus för flyktingar i krisomr˚aden. Detta modulhus h˚alls p˚a plats mot vind och andra förh˚allanden av 10 stycken markankare som är nedgrävda och fastspända i husramen. Detta kandidatexamensarbete stud- erar och utvärderar de mekaniska egenskaperna för dessa ankare och undersöker möjligheterna för ett materialutbyte fr˚an aluminium till en glasfiberplast och presenterar en förbättrad design.

Utv¨ardering av ankarvajern ¨ar inte relevant i denna rapport.

Resultaten är baserade p˚a FEM-simuleringar i programmet, ANSYS Workbench, vilket inkluderar givna samt egendesignade CAD-modeller för ankaret i b˚ada materialen. Relevanta data samlades in och analyserades för olika lastfall, statiska och växlande med en kraftamplitud p˚a 2000 N . Relevanta data inkluderar effektivspänning och maximal huvudspänning samt säkerhetsfaktor och utmattningslivslängd.

Simuleringarna p˚avisade mindre plastisk deformation och skada för nuvarande ankaret i b˚ada materialen. Detta är diskutabelt eftersom datorsimuleringen är baserad p˚a modeller vilka inte är en helt korrekt representaion av verkligheten. Som ett exempel i detta fall s˚a uppn˚ar spänningarna i CAD-modellens vassa kanter ett högre värde än vad som skulle tyckas vara realistiskt. Dock kunde större förbättringar observeras hos den nya ankardesignen i form av h˚allfasthet och minskning av volym. Detta innebär att ett byte till den nya designen och det andra materialet är motiverat.

Förbättringarna inkluderar en reducerad kostnad till 4.44 SEK, vilket är en sänkning p˚a 54.8%

per enhet samt en minskning i vikt fr˚an 29 g till 11.8 g per enhet.

(6)

Acknowledgement

Firstly, we would like to thank our supervisor, Per-Lennart Osk Larsson. Not only for a well- hearted continuous support and sharing his valuable point of views, but also for giving us a chance to think for ourselves and discuss our results, which is important as future engineers. We would also like to thank our course coordinator, Jonas Neumeister, whom have contributed with a great learning experience during this course as well as great support. Lastly, we would like to thank Tim de Haas and Better Shelter for sharing valuable data, files and materials, which made our journey a lot easier and giving us the opportunity and joy of an interesting and meaningful bachelor degree assignment.

(7)

List of Figures

1 Cut through illustration of anchor and attachment of wire. . . . . 4

2 Cut through illustration of new design anchor and attachment of wire. . . . 5

3 Sub figure A shows how the anchor and wire is applied in the ground. Sub figure B showing how to activate the anchor in the ground. . . . 5

4 Side views of the long and short side of the shelter with dimensions. . . . 6

5 3D model of the anchor. . . . 8

6 3D model of the wire. . . . . 8

7 Ground anchor with the blue highlighted area being the fixed surface on the top of the anchor representing the ground to keep the anchor in place. . . . 9

8 Top of wire with applied force. . . . 9

9 Equivalent stress in the aluminum alloy ground anchor. . . . . 11

10 Safety factor for the aluminum anchor. . . . 11

11 Calculated life of the aluminum anchor. . . . 12

12 Principal stress in the plastic ground anchor. . . . 12

13 Safety factor the glass fiber reinforced plastics anchor. . . . 13

14 Life length for the glass fiber reinforced plastics anchor. . . . 13

15 Equivalent stress for the new designed aluminum anchor. . . . . 14

16 Safety factor the new designed aluminum anchor. . . . 14

17 Life length for the new designed aluminum anchor. . . . 15

18 Principal stress for the new designed glass fiber reinforced plastics anchor. . . . 15

19 Safety factor for the new designed glass fiber reinforced plastics anchor. . . . 16

20 Life length for the new designed glass fiber reinforced plastics anchor. . . . . 16

21 Technical data of alternative material sub figure 1. . . . 22

24 SN curve for aluminum alloy. . . . 25

25 SN curve for GFRP. . . . 26

(9)

List of Tables

1 Parameters introduced, with designation, value and unit . . . . 21

(10)

1 Introduction

In this section, the background of the project is presented, followed by a formulation of the assignment. The major conditions, assumptions and restrictions are also presented here.

1.1 Background

Better Shelter is a nonprofit organization that provides local, temporary modular housing for refugees or displaced people due to causes like war, persecution or natural disasters. These shelters are delivered in two flat packages, each weighing about 80 kg, and can easily be assembled by four people during four hours with no extra tools required. The expected lifespan is three years in moderate climates and each shelter can easily be dismantled, moved and reassembled at a different location. Each part can be replaced individually if damage would take place. In order to withstand vertical forces, primarily due to wind loads, the shelter is fixed by ten ground anchors, four on the long side and one on the short side, which are buried and strapped to the frame.

1.2 Assignment formulation

The aim of this assignment is to evaluate and review the mechanical performance of the existing ground anchor, investigate the possibility to replace the current material with an alternative, a glass fibre-reinforced plastic (GFRP as of table [1]) as suggested, as well as design optimization’s.

This will serve to give Better Shelter valuable data of the current situation and the utilization of a suitable replacement for the material will result in a decrease of cost and weight, which will not only benefit the company but also favor the transportation. A change in design, if possible, will make the anchor forcible to withstand greater loads which could motivate a transition to a lighter, and therefore weaker, material.

The assignment is summed up by the following tasks taken directly from the project description by Better Shelter.

• Evaluate the performance of the current ground anchor with static/dynamic loads (baseline).

• Evaluate the alternative material, a glass fibre reinforced plastics, in regards to e.g cost, weight and performance.

• Propose optimization such as shapes, other materials etc.

1.3 Method

1.3.1 Literature study

Studies have been performed with focus on different mechanical properties of the materials, which in this case is an aluminum alloy and a GFRP, as well as the price. Fluid mechanical data for the wind load analysis and data from tensile experiments, including the ground anchorage, has been collected and used for this assignment. Chemical influence on the plastic material has been reviewed but is not included in the calculations.

1.3.2 Solid Edge

Parts for the ground anchoring were given by the employer and consisted of a part file for the ground anchor wire and a part file for the anchor itself, which were assembled in Solid Edge ST9 into an assembly file. The program was also used to design a new anchor with most of the original design intact except for smaller alterations based on the hypothesis of improvement.

(11)

1.3.3 FEM Analysis

Simulations in the FEM program ANSYS Workbench has been used to calculate the stresses, safety factors and fatigue life for the given anchor model as well as the new design for different materials.

1.4 Restrictions and assumptions

A number of restrictions and assumptions were needed to make this assignment feasible. The main restriction we relied on, given by the employer, was that the performance evaluation solely comprised of the anchor itself. According to the tensile experiments [11], two different kinds of outcome were observed. Either were the anchor pulled out of the ground or the wire snapped at the weakest point just outside the casted ball joint. This concludes that the wire is the weakest link in the ground anchorage and the resulting stress in the wire during the FEM simulations will be dismissed, even if the wire is used to distribute the load to the anchor. In the simulations, the wire was set as a program default steel material. Since the anchor is affected by the loads while buried in the ground, the loads will be considered static and no dynamic loads will arise. However, a fatigue analysis was done to simulate the effect of alternating wind loads.

The assignment treated the data and results only from the materials given by the employer. No other materials have been taken into account except for minor studies, see ”Discussion”. The materials were, for the sake of simplicity, assumed to be homogeneous, isotropic and linearly elastic. This means that the principal stresses align with the principal strains. Hooke’s law was also assumed [1]. Deformation processes like creep and cracking has not been taken into account, also for the sake of simplicity.

While calculating the wind loads it is assumed that the house experience drag load in a homogeneous, enclosing stream with the reference area consisting alongside and modeled as a rectangular cuboid. The resulting load was equally distributed over the four side anchors and, for simplicity, the contributed load on the short sided anchors were dismissed. Turbulence is not considered.

The conditioned tensile strength for the GFRP were given and was also used in the calculations.

”Conditioned” means, in this case, that the plastic has absorbed moisture from it’s surroundings which diminishes the tensile strength and stiffness but also increase flexibility and impact toughness [13]. Physical wear from the soil has not been considered as well as chemical wear.

Price of materials proved to be difficult to find because it is rarely suggested by the manufacturers directly on their website. The reason being that price is often adjusted during bargaining between them and the customer. It’s also dependent on the amount that is being sold. Therefore, the price is assumed according to the employer’s suggestion and correlate to material consumption.

2

(12)

2 Materials and design

The following chapter presents the material and design used in today’s anchor as well as the new alternative material and a new suggested design. The characteristics, pros and cons on each of these are highlighted.

2.1 Aluminum alloy

Today’s design is made out of an aluminum alloy. It consist mainly of aluminum, composed with a mixture of silicon, copper and lead, nickel, iron, tin, zinc, titanium, mangan and magnesium. The alloy is considered to be homogeneous, ideally plastic and isotropic elastic [1]. Due to it’s ductile abilities and hence not being brittle [2], when testing the aluminum anchor the equivalent stress (plasticity) is the dimensional factor and therefore results for this is what will be presented.

The benefits of using aluminum is the high strength in it, it’s ductile and can easily be shaped but is also easily dented. Compared to other metals it’s much lighter but still has the characteristic strength of a metal, and aluminum’s strength rise with cold temperatures [3] [4].

2.2 Glass fibre-reinforced plastic

Plastics has it’s benefits that it’s relatively light compared to metals and is cheaper to manufacture.

This comes mainly at the cost of decreased strength which can worsen in the long term due to environmental influence, like temperature fluctuations and humidity. The reason to reinforce the plastic with glass fibres is to increase it’s strength and make it more solid and durable. The area of use for plastic composites are limited to fields with low temperatures. In temperatures higher than 200^◦C most plastics tend to float.

A suggested material is a Polyamide 6 (Nylon 6) material filled with a fibre percentage of 15%.

When processing the material, chopped glass fibres are mixed in the plastic by injection molding.

The glass fibres orient themselves in random order throughout the material and the composite can be considered homogeneous, ideally plastic and isotropic elastic[1] [2]. Since the GFRP is brittle, the largest principle stress will be the dimensional factor and therefore results for this is what will be presented [3] [5].

2.3 Current design

It today’s design, the wire and anchor are connected by passing the wire through the hole on the anchor and stop by the ball joint at the end of the wire. Figure 1 below is an cut through illustration of the anchor, the wire with it’s ball joint at the end and how they are attached together. The area of the contact between the parts is marked out. Using tools in Solid Edge, the volume was calculated to be

V = 10.3 cm³ (1)

This corresponds to the masses

malu= ρalu· vga,t= 29 g (2)

mGF RP = ρGF RP · vga,t= 13 g (3)

(13)

The price, given by the employer for the aluminum anchor is 2.7 SEK and for the anchor in GFRP about half of that, 1.35 SEK [??].

Figure 1: Cut through illustration of anchor and attachment of wire.

2.4 Alternative design

With today’s design, the contact between the anchor and the ball joint consists of a thin line along the edge. This edge transfer the load from the wire to the ground which leads to an enormous pressure on a small area. A new, alternative design was made where the section of the contact area was changed to a hollow, spherical shape to increase the contact area between the anchor and the ball joint and therefore equalize the pressure. This would reduce stress in the material. Figure 2 illustrates a cut through view on the new design with it’s attachment between the ground anchor and the ball joint. Using tools in Solid Edge, the volume was calculated to be

V = 9.4 cm³ (4)

This corresponds to the masses

malu= ρalu· vga,n= 26 g (5)

mGF RP = ρGF RP · vga,n= 11.8 g (6)

4

(14)

Figure 2: Cut through illustration of new design anchor and attachment of wire.

3 Load cases

With help of a supporting steel beam being forced down in the ground the anchor will be set in place. When being exposed to a force pulling it up, it will act on the same principle as a fish hook and turn orthogonal to the wire causing it to stop any movement upwards. Below in figure 3 it is showed how the anchor is assembled and attached in the ground with the wire.

(a) Anchor put in place.

(b) Anchor being fixed.

Figure 3: Sub figure A shows how the anchor and wire is applied in the ground. Sub figure B showing how to activate the anchor in the ground.

(15)

When the shelter is exposed to vertical forces like wind, the anchors will keep it in place. The force on then anchors in these cases works the same way as the force exposing the anchor in figure 3b.

3.1 Static load

The static load on the anchor consist of a continuously applied force of 2000 N . This is based on previous tensile experiments where the anchor wire broke at the minimum force of 1960 N [??]

and can be simulated as the shelter being affected by a constant wind flow. However, this is a conservative load case and the maximum allowed air velocity would not cause such load, see ”Wind load”.

3.2 Alternating load and fatigue

In the alternating load case, the anchor was exposed to a pulsating force with a constant amplitude load. In other words, a force with a certain power was applied in periods so the overall load on the anchor periodically went between zero and the maximum power from the force. This to simulate a changing wind on the shelter which will test the fatigue of the anchor. Due to lack of data, some material parameters were estimated from studies of similar materials. This would cause an uncertainty factor in this load case. The same force of 2000 N as in the static load test was applied due to the results of previous test [??].

3.3 Wind load

The purpose of this section is to illustrate the anchor loading caused by the maximum allowed wind velocity. The results are not used in the calculations but serves as a comparison to the forces used in the FEM simulations. The weight of the shelter and friction is dismissed as well as the widespread support of the floor, assuming that the forces arising from the wind loads are absorbed only by the ground anchor connections, and the reactionary torque in the frame attachments is dismissed. To summarize, the worst case scenario for the ground anchors is assumed. Figure 4 shows the dimensions of the shelter.

Figure 4: Side views of the long and short side of the shelter with dimensions.

6

(16)

The wind force F_v [10] is calculated by the equation

F_v= 1

2C_dρ_∞V_∞²S (7)

From the wind force, roughly simplified force equilibrium calculations was made showing that the anchor load is much smaller than the 2000 N used in the simulations.

(17)

4 Solid Edge

The CAD models of the existing ground anchorage, consisting of the anchor and wire, were distributed by the company as STEP-files and imported into the CAD program Solid Edge ST9. The files were then assembled into an assembly file and re-converted to a STEP-file. Figure 5 and 6 showing the two separate CAD models of the anchor and wire.

Figure 5: 3D model of the anchor.

Figure 6: 3D model of the wire.

8

(18)

5 ANSYS Workbench

With the ground anchor and wire assembled from Solid Edge, the complete model was imported to ANSYS Workbench. In ANSYS the boundary conditions was set to make a representative model of how the structure work in the field. Fixed support was set as illustrated in Figure 7 to stimulate how the ground keeps the anchor in place. Depending on mesh size, the differences between results can be enormous. The meshing size was set to 1 mm and the contact area between the ball joint and anchor was refined in order to obtain accurate results.

Figure 7: Ground anchor with the blue highlighted area being the fixed surface on the top of the anchor representing the ground to keep the anchor in place.

A force vector was applied at the top of the wire acting as a force pulling the anchor upwards, see figure 8 below. The force is transmitted from the ball joint onto the anchor and causes normal stress and shear stress. The equivalent stress in the aluminum case is calculated according to the von Mises suggestion. S-N curves [C] [D] and Goodman mean stress correction was used for both materials while calculating the stress life of the anchor.

Figure 8: Top of wire with applied force.

(19)

The material data for the aluminum alloy, given by the employer, was insufficient for a complete analysis in ANSYS. Therefor, other sources were used to gather data, like MatWeb [6], for tensile strengths and density and [7] for Poisson’s ratio. Nearly all relevant parameters for the GFRP were collected from the given property sheet by the employer. The only parameter missing was the Poisson’s ratio, which were assumed to be 3,5 based on another type of GFRP [8].

10

(20)

6 Results

In this section the results of the anchor evaluations are presented in the form of saved images of simulations in ANSYS Workbench. Every case of evaluation is represented by an image and the results are presented by a spectrum of colors which corresponds to the colors of the model. Each color representing an interval of calculated values. The equivalent stress is based on the von Mises suggestion and safety factor presents the relation between the corresponding maximum stress and yield strength. Life represent the amount of loading cycles the anchor can endure until failure occurs.

6.1 Aluminum alloy, current design

The equivalent stress in the anchor is presented in Figure 9. The yield stress is exceeded and plastic deformations will occur for a limited area near the edge.

Figure 9: Equivalent stress in the aluminum alloy ground anchor.

The safety factor is presented in Figure 10.

Figure 10: Safety factor for the aluminum anchor.

(21)

The cyclic life length of the aluminum anchor is presented in Figure 11.

Figure 11: Calculated life of the aluminum anchor.

6.2 Glass fibre-reinforced plastic, current design

The maximum principal stress in the anchor is presented in Figure 12. The tensile strength is exceeded, which implies damage will occur for a limited area near the edge.

Figure 12: Principal stress in the plastic ground anchor.

12

(22)

The safety factor is presented in Figure 13.

Figure 13: Safety factor the glass fiber reinforced plastics anchor.

The cyclic life length of the GFRP-anchor is shown in Figure 14.

Figure 14: Life length for the glass fiber reinforced plastics anchor.

(23)

6.3 Aluminum alloy, alternative design

The equivalent stress for the anchor is presented in Figure 15. The yield stress is exceeded for a minimal area, implying plasticity will occur.

Figure 15: Equivalent stress for the new designed aluminum anchor.

The safety factor is shown in Figure 16.

Figure 16: Safety factor the new designed aluminum anchor.

14

(24)

The cyclic life length is presented in Figure 17 .

Figure 17: Life length for the new designed aluminum anchor.

6.4 Glass fibre-reinforced plastic, alternative design

The maximum principal stress on the anchor is presented in Figure 18 . The tensile strength is exceeded for a limited area, implying that damage will occur there.

Figure 18: Principal stress for the new designed glass fiber reinforced plastics anchor.

(25)

The safety factor is shown in Figure 19 .

Figure 19: Safety factor for the new designed glass fiber reinforced plastics anchor.

The cyclic life length is presented in Figure 20 .

Figure 20: Life length for the new designed glass fiber reinforced plastics anchor.

16

(26)

7 Discussion

The goal of this assignments has been to evaluate the ground anchors, test these and see if they can be replaced with alternative materials and possibly new designs as well. In this section the rea- sonableness and the accuracy of the results obtained are discussed, along with possible suggestions for further research.

7.1 Accuracy disclaimer

Discussions regarding assumptions, accuracy and results are divided and presented in subsections below.

7.1.1 Isotropy

Both of the evaluated materials were assumed to be homogeneous, isotropic and linearly elastic.

This is to make the calculations mathematically feasible and is not an accurate representation of reality. The material structure on a microscopic level differs depending on porosity and grain size due to small casting deformations and thermal shrinkage. Therefore every unit is unique and quality is also affected by which casting or molding technology that is being used.

The GFRP is a laminate and is normally considered as an anisotropic material, which implies that different properties, like tensile strength, are directionally dependent. As mentioned in ”Materials and design” the anchors are produced by chopped glass fibres mixed with plastic by injection molding. Therefore the glass fibres will orient themselves in random directions depending on how the plastic flows in the mold. As a consequence the tensile strength will vary throughout the material which could affect the results.

7.1.2 Wind load

The anchor force calculated in the wind load case is incorrect for many reasons. In nature wind is seldom, if not ever, homogeneous, laminar and widespread as illustrated in Figure 5 and if it was, the velocity profile would decrease in magnitude closer to the ground resulting in an uneven pressure distribution on the wall. Due to the shape of the roof, the wind load would contribute with a pressure distribution on the roof as well. This is dismissed as the shape of the shelter is modeled as a cuboid, making the drag coefficient in use inaccurate. The suction pressure that would arise on the other side of the shelter due to the velocity retardation is dismissed. Lastly, the wind load is defined only to affect the anchors, which is untrue because reaction forces and friction from the shelter structure and soil is not taken into account. The forces applied to the anchors will therefore be assumed to be smaller in magnitude. However, these approximations are considered sufficient since the purpose is to illustrate the comparison between the roughly calculated force and simulated force.

7.1.3 Material data

Material data were gathered from different sources since a complete document with all required parameters could not be found. This can be problematic since data can vary from different sources, e.g measured tensile strength, which affects the results. For ANSYS to calculate cyclic fatigue life S-N curves were needed, which proved difficult to find and fatigue data were therefore gathered from similar materials. S-N data for aluminum were taken from Wikipedia [9] and ultimate tensile strength for the aluminium alloy were put as a start value. S-N data for the GFRP were gathered from [5] where the material has 30 percent glass fibre instead of 15 percent.This is not accurate due to variations of tensile strength depending on glass fibre content.

(27)

7.1.4 Results

The results from the simulations are based on mathematical models. In ANSYS, the 3D model is divided into an finite amount of elements which is affected differently depending on mesh size and structure, in turn affecting the results. More elements mean more accurate results but at a higher cost of computer working time.

In the 3D model of the current anchor, the connection between the ball joint and the anchor consists of a thin line, as mentioned in ”Alternative design”. This leads to enormous concentrations of stress near and on the edge, which is illustrated by the simulations. In reality, this is not the case because the edge is more rounded which reduces the stress concentrations. Therefore, the stresses should be smaller than presented in ”Results”. It is also emphasized that these stress concentrations are limited to a small area, according to the simulations.

7.2 Agenda of further research

As mentioned in ”Restrictions and assumptions”, even if the wire is experimentally proven to be the weakest link, it has always been dismissed. Therefore, a suggestion is to make the wire stronger.

Maybe by making it wider or tieing the wire end into a knot before the ball joint casting.

A great difficulty during the assignment was the lack of relevant mechanical data for the specific materials, which forced usage of data from other, similar materials. Suggestively extensive testing should be performed in order to determine different material parameters. The GFRP react nega- tively to different chemicals, including common acids [12]. Therefore an investigation on how the plastics react to e.g acid rain due to causes like volcanic eruptions etc might prove useful.

In regards to change of material, as mentioned in ”Assumptions and restrictions”, no other materials were investigated in detail. However, there seemed to be a correlation between tensile strength and glass fibre percentage. If an increase of material stiffness is desired for the anchor, alternative GFRP:s with higher glass fibre percentage can be found at plasticsfinder.com [14].

Even if the wire snaps at 2000 N , the wind force that is required for that to happen far surpasses the maximum wind velocity that is allowed. This implies that both the anchor and the wire strength at the current state is overdimensioned in comparison to the shelter and therefore a suggestion for further research is to design the shelter to withstand stronger winds.

18

(28)

8 Conclusion

Considering the results given by this report, it is concluded that a switch from the current ground anchor model in aluminium to the new, proposed design in the GFRP is well motivated. Even if it is observed that the current anchor in the GFRP would resist failure, only changing the material of the current design would drastically reduce the strength properties of the anchor. The change of design would reduce the stress and make the anchor more durable, which thereby would make it possible to change the anchor material and maintain higher security levels. This would also result in a decrease of cost and weight compared to the anchor in use today due to the reduction of volume.

To summarize in regards to the assignment formulation, the performance evaluation of the current ground anchor show that the simulated loads would cause minor plasticizing in a small area, which is not enough to cause failure. The alternative material would result in a decrease of cost and weight but also a decrease in performance due to it’s lower tensile strength. A new ground anchor design was proposed in which the contact between the wire and the anchor was redesigned to reduce the arising stresses and, as a consequence, reduce the material volume. Otherwise the original design was kept more or less intact.

(29)

References

[1] Bo Alfredsson. Handbok och formelsamling i H˚allfasthetsl¨ara.

Instant book AB, Stockholm 2014.

[2] Hans Lundh. Grundl¨aggande h˚allfasthetsl¨ara.

Instant book AB, Stockholm 2013.

[3] J. W. D. Callister and D. G. Retwisch. Materials science and engineering: An introduction.

[online] https : //abmpk.f iles.wordpress.com/2014/02/book maretial − science − callister.pdf John Wiley Sons, inc .

[4] E. J. Lavernia. Materials science and engineering A.

[online] https : //www.sciencedirect.com/science/article/pii/S0921509312002912

[5] J. J. Horst and J. L. Spoormaker. Mechanisms of Fatigue in Short Glass Fiber Reinforced Polyamide 6.

[online] https : //onlinelibrary.wiley.com/doi/pdf /10.1002/pen.10671

[6] MatWeb, material property data. MatWeb, Your Source For Materials Information.

[online] http : //www.matweb.com/index.aspx

[7] N Kumor, S Soni and R.S. Rana. Mechanical properties enhancement of AL-SI(ADC12) alloy by heat treatment.

[online] http : //www.iraj.in/journal/journal f ile/journal pdf /2 − 338 − 148903626547 − 51.pdf

[8] Nylon 6/6 glass fiber reinforced (PA66-GF). MaterialUniverse Database.

[online] https : //sustainabilityworkshop.autodesk.com/sites/def ault/f iles/gallery − f iles/aircreation_materialdata_pa66 − gf.pdf

[9] Wikipedia. Materialutmattning.

[online] https : //sv.wikipedia.org/wiki/M aterialutmattning [10] Arne Karlsson, Mekanik, KTH.

Formelsamling i Str¨omningsmekanik.

29th of august 2007, Edition 2.11.

[11] Tim de Haas and Nicol`o Barlera, Anchor testing.

Refugee Housing Unit

RHU in house testing - Anchor testing.

14th of may, 2014.

[12] Property Data, Akulon.

Akulon K224-G3, PA6-GF15.

[online] https : //plasticsf inder.com/api/document/tech/Akulon%C2%AE%20K224 − G3/a3CyuGAwO/en/AkulonK224G3en.pdf

25th of may 2018.

[13] Prospector.

Dry vs. Conditioned Polyamide Nylon Explaindes.

[online] https : //knowledge.ulprospector.com/1489/pe − dry − vs − conditioned − polyamide − nylon/

29th of january, 2010.

[14] Plastics finder.

DSM Engineering plastics

[online] https : //plasticsf inder.com/

20

(30)

A Table of parameters introduced

Table 1: Parameters introduced, with designation, value and unit

Parameter Designation Value Unit

Glass Fibre Reinforced Plastics GFRP - -

Price per kilogram, aluminum alloy palu 93.1 SEK/kg

Price per kilogram, GFRP p_{GF RP} 103.8 SEK/kg

Density, aluminum alloy ρ_alu 2816.9 kg/m³

Density, GFRP ρ_{GF RP} 1262.7 kg/m³

Volume anchor, today’s design v_ga,t 10.295 cm³

Volume anchor, new design v_ga,n 9.353 cm³

Drag coefficient, cuboid Cd 2 -

Density, air (20^◦C) ρair 1.2 kg/m³

Homogeneous stream velocity V_∞ 18 m/s

Tensile yield strenght, aluminum σs,alu 165 M P a Ultimate tensile strenght, aluminum σb,alu 331 M P a Tensile yield strenght, GFRP σs,GF RP 70 M P a Ultimate tensile strenght, GFRP σb,GF RP 70 M P a

Poisson’s ratio, aluminum valu 0.33 -

Poisson’s ratio, GFRP vGF RP 0.35 -

Young’s modulus, aluminum Ealu 68.9 GP a

Young’s modulus, GFRP E_{GF RP} 3.5 GP a

(31)

B Appendix alternative material characteristics and tech- nical data

Figure 21: Technical data of alternative material sub figure 1.

22

(32)

(33)

24

(34)

C Appendix SN curve for aluminum alloy

Figure 24: SN curve for aluminum alloy.

(35)

D Appendix SN curve for glass fibre reinforced plastics

Figure 25: SN curve for GFRP.

26

(36)

Evaluation of ground anchor for temporary shelters

Evaluation of ground anchor for temporary shelters

Performance, design and alternative materials

PATRIK TER VEHN VIKTOR OLSSON

Utvärdering av markankare för temporära

flyktingbostäder

Prestanda, design och alternativa material

PATRIK TER VEHN VIKTOR OLSSON

Public Version

Abstract

Abstract in Swedish

Acknowledgement

Contents

List of Figures

List of Tables

1 Introduction

2 Materials and design

3 Load cases

4 Solid Edge

5 ANSYS Workbench

6 Results

7 Discussion

8 Conclusion

References

A Table of parameters introduced

B Appendix alternative material characteristics and tech- nical data

C Appendix SN curve for aluminum alloy

D Appendix SN curve for glass fibre reinforced plastics