Sample quality effects of laser cutting

(1)

IN

DEGREE PROJECT TECHNOLOGY, FIRST CYCLE, 15 CREDITS

STOCKHOLM SWEDEN 2020,

Sample quality effects of laser cutting

An empirical study on the heat affected zone and the surface quality in laser cut samples

EMMA BEVIN

MATILDA BJÖRKLUND

KTH ROYAL INSTITUTE OF TECHNOLOGY

SCHOOL OF INDUSTRIAL ENGINEERING AND MANAGEMENT

(2)

Abstract

This study has been conducted with the aim to examine the extent of the heat affected zone in order to use laser cutting when making samples for tensile testing. When cutting with a laser the material absorbs heat energy from the laser beams which causes changes in the microstructure. The heat affected zone affects the properties of the materials, lowering the accuracy of the tensile test. Therefore, it is desired to know the extent of the heat affected zone in order to remove it before tensile testing.

In this study two materials were used, the high strength steel Docol 1000DP and the duplex stainless steel LDX 2101. The materials were cut in the shape of dog bones for tensile testing using two different laser powers, 2500 W and 3500 W. The samples were cut with different cutting speeds, starting at lower cutting speed, increasing until the laser was unable to cut through. Thereafter, the heat affected zone, and the surface quality was examined.

The results from this study showed that the heat affected zone decreases with increased cutting speed. When cutting with high cutting speeds in Docol 1000DP it is enough to turn away 0.30 mm in order to remove the heat affected zone with margin. Negligible difference in heat affected zone was observed between the samples cut with 2500 W and 3500 W. The heat affected zone in LDX 2101 was very small, in order of 50 μm, making it hard to measure. This resulted in no exact measurements being made. However, the heat affected zone was in the order of 50 μm for all samples, concluding that turning away 0.15 mm is sufficient to remove the heat affected zone with margin. No difference could be observed between the samples cut with a laser power of 2500 W or 3500 W. Common to both materials is that the amount of dross decreases with increased cutting speed.

Keywords

Laser cutting, Laser cutting quality, Heat affected zone, Surface quality, Surface angularity

(3)

Sammanfattning

Denna studie har utförts i syfte att undersöka utsträckningen av den värmepåverkade zonen för att kunna använda laserskärning för att skära prover till dragprovning. Ett problem med laserskärning är att metaller absorberar värmeenergin, vilket orsakar förändringar i mikrostrukturen. Denna värmepåverkade zon bör minimeras och avlägsnas från metallen, då det minskar kvalitén på proverna samt på dragprovningen.

I denna studie har två material undersökts, det höghållfasta stålet Docol 1000DP och det duplexa rostfria stålet LDX 2101. Stålen skars ut i form av hundben för dragprovning med två olika effekter på lasern 2500 W och 3500 W. Proverna skars ut med olika skärhastigheter, började med lägre hastigheter och ökade sedan tills lasern inte längre kunde skära igenom materialet. Därefter undersöktes både den värmepåverkade zonen och kvaliteten på skärytan.

Resultaten från denna studie visade att den värmepåverkade zonen minskar med ökad skärhastighet. Vid skärning i Docol 1000DP med höga skärhastigheter räcker det att avlägsna 0,30 mm för att ta bort den värmepåverkade zonen med marginal. Ingen skillnad i värmepåverkad zon observerades mellan proverna skurna med 2500 W och 3500 W vid skärning med högsta skärhastigheten för varje lasereffekt. Den värmepåverkade zonen i LDX 2101 var mycket liten, i storleksordningen 50 μm, vilket gjorde den svårt att mäta.

Slutsatsen gav att det är tillräckligt att avlägsna 0,15 mm för att avlägsna den värmepåverkade zonen med marginal. Ingen skillnad kunde observeras mellan proverna skurna med en lasereffekt på 2500 W eller 3500 W. Gemensamt för båda materialen var att gradbildningen minskade med ökad skärhastighet.

Nyckelord

Laserskärning, Laserskärnings-kvalité, Värmepåverkad zon, Ytkvalité, Ytvinkel

(4)

Table of content

1 INTRODUCTION ... 1

1.1BACKGROUND ... 1

1.2AIM OF THE PRESENT WORK ... 1

2 LASER CUTTING AND MATERIALS ... 2

2.1LASER CUTTING ... 2

2.2LASER CUTTING QUALITY ... 3

2.3MATERIALS ... 4

2.3.1 Docol 1000DP ... 4

2.3.2 LDX 2101 ... 5

3 EXPERIMENTS ... 7

3.1SAMPLE PREPARATION ... 7

3.2CHARACTERIZATION METHODS ... 9

4 RESULTS ... 11

4.1DOCOL 1000DP ... 11

4.1.1 Heat affected zone ... 11

4.1.2 Surface angularity ... 13

4.1.3 Surface quality ... 14

4.2LDX2101 ... 16

4.2.1. Heat affected zone ... 16

4.2.2 Surface angularity ... 18

4.2.3 Surface quality ... 19

5 DISCUSSION ... 21

5.1DOCOL 1000DP ... 21

5.2LDX2101 ... 22

5.3SOURCES OF ERROR ... 22

5.4SUSTAINABILITY AND ETHICAL ASPECTS ... 23

6 CONCLUSIONS ... 24

6.1DOCOL 1000DP ... 24

6.2LDX2101 ... 24

7 ACKNOWLEDGEMENTS ... 25

8 REFERENCES ... 26

APPENDIX A: DOCOL 1000DP ... 28

APPENDIX B: LDX 2101 ... 32

(5)

1

1 Introduction

This is a project is a bachelor thesis for Degree Project in Material and Process Design, at the Royal Institute of Technology. The background and the aim of this project is presented below.

1.1 Background

Laser cutting is an efficient method for cutting materials as it allows complex geometries to be cut at a high rate while maintaining high-quality cutting surfaces. The first laser cutting machine was produced in the 1960s and was only used for drilling holes. The modern laser cutting machine can be used to cut the following materials; metals, wood, multiple plastics, fabric, and paper. Materials such as glass and stone are difficult to cut using a laser. When using laser cutting, the area closest to the cutting surface will be affected by the heat generated by the laser beams which can cause changes in the microstructure. It is of great importance to study the extent of the affected area, also known as the heat affected zone (HAZ), as the microstructure can be crucial in order to maintain product quality. However, one can minimize the HAZ and improve the cutting surface quality by optimizing the laser parameters. The present work will concentrate on optimizing the laser parameters in order to minimize the HAZ and improve the cutting surface quality for metals.

1.2 Aim of the present work

Laser cutting’s efficiency and flexibility make it a preferred method when cutting specimens for tensile testing. However, it is vital to consider how it affects the material and its microstructure to make the testing reliable. The metals research institute SWERIM AB is equipped with a new laser cutting machine and are interested in using it to cut samples for tensile testing. The goal is to learn to what extent the laser affects the metal in order to guarantee the reliability of the tensile test. In this study the HAZ will be examined in two steels, Docol 1000DP and LDX 2101. The laser parameters that will be varied are the laser power and the cutting speed. The HAZ will be studied in three positions on the sample, covering three different geometries. Obtaining knowledge about the extent of the HAZ in different materials will enable SWERIM AB to cut samples for tensile testing and to remove the affected area with a lathe.

(6)

2

2 Laser cutting and materials

A theoretical study was performed to obtain the necessary knowledge for this project. The relevant information found is stated below.

2.1 Laser cutting

Laser cutting is a particularly efficient process as it has a very high cutting speed and high- quality cutting surfaces. The principle of laser cutting is that multiple laser beams are concentrated to one point, generating enough heat to melt the material in order to blow it away, creating the cutting surface. A laser source generates laser beams and sends them through a light diaphragm, which obstructs all light except the light passing through the aperture. The laser beams that passed through are reflected towards a lens through a reflector. The lens concentrates all the laser beams to one spot, the focal point, where heat is generated. The laser is aimed through a nozzle along with an assisting gas, which blows away the molten material, creating the cutting surface [1]. The laser setup is shown in Figure 2.1 [2].

Figure 2.1: A schematic figure showing the set-up of a laser cutting head.

(7)

3

The material absorbs energy from the laser beams which is converted into heat energy. The irradiated surface melts due to the rapid rise of temperature and vaporize. The thermal diffusion causes the surrounding area to melt. The evaporation expands, causing a micro-explosion and the molten material is quickly ejected to create an anti-shock wave. This creates a hole with a large upper side and a small lower side, this phenomenon is known as a key-hole formation. In order to remove the molten material a high-speed gas flow is used. The gas must be adapted to the material, commonly used gases are air, nitrogen, and oxygen. The chosen gas cools the heated surface, but it is important that the pressure and velocity of the gas are adapted to suit the material as rapid cooling can cause thermal strains and stresses [1].

The laser parameters that can be varied are cutting speed, gas flow, laser power, focal position, and the nozzle size [3].

The laser cutting machine used in this study is Precitec SolidCutter Lasermatic® Z. The laser source is an Nd-YAG fiber laser and has a maximum effect of 10 kW. However, the cutting tool's maximum effect is 4 kW and the focal point can be moved +4/-6 mm relative to the material's surface. The standoff distance, the distance from the tip of the nozzle to the surface of the material, is recommended to 1.0 mm, but distances between 0.3-10 mm are possible [4].

The cutting tool is placed on a robotic arm, which limits the possible cutting speed as it gets unstable at higher cutting speeds.

2.2 Laser cutting quality

The quality of the cutting surface is directly dependent on the cutting process. Parameters as laser power, cutting speed and focal position are examples of parameters that directly affect the quality. A high-quality cutting surface is smooth, without striations, and does not have any dross left on the cutting surface. Dross is residues of the molten material that cooled on the lower cutting surface creating an uneven edge. It is also important to consider the angularity of the cutting surface [5].

The cutting speed has a large impact on the cutting surface quality. When the cutting speed is too low the surface will be rough and as it increases the surface becomes smoother. However, using too low cutting speed will result in either dross or too high cutting speed will result in the laser being unable to cut through the material. Cutting with a lower cutting speed will allow the material to absorb more heat energy, increasing the width of the cut as well as the thermal influence on the material. A wider cut will increase material losses and increase the risk of an uneven cut. Thus, it is most often preferable to use a higher cutting speed. When cutting thicker materials, both the laser power and cutting speed should be high to be able to cut through the material. When cutting with higher speeds it is necessary to adapt the gas pressure in order to maintain a smooth cutting surface. When the heated surface is cooled, cracks can form, which decreases the surface quality [1]

(8)

4

Another parameter affecting the cutting quality is the extent of the heat affected zone, further called HAZ. The laser beam heats the area around the cutting surface, which can cause changes in the microstructure. If a higher laser power is used or lower cutting speed, more heat will be generated and absorbed by the material causing the HAZ to enlarge. Different materials will be affected differently depending on the thermal properties. The HAZ is decreased by adjusting the laser parameters and using short laser pulses [6].

The position of the focal point relative to the material’s surface is vital for obtaining high- quality cutting surface. The position of the focal point is dependent on the type of material and its thickness. When cutting stainless steel, the focal point should be located below the upper cutting surface. When cutting zero to six millimeters steel plates the focal point should be located on the upper surface, but as the material thickens, over 6 mm thick, the spot should be located either above or below the upper surface, depending on the assisting gas used [7]. If the focal point is measured and adapted to the material and its thickness a smooth and clean cutting surface can be achieved. Incorrect positioning of the focal point can result in striations and dross lowering the quality. Inaccurate positioning will cause the laser beams to spread resulting in a wider cut, which affects the angularity of the cutting surface [1].

2.3 Materials

Two steels were examined in this study and the relevant information regarding the chosen materials are presented below.

2.3.1 Docol 1000DP

Docol 1000DP is a high strength dual-phase steel that has been heat-treated in a continuous annealing line and therefore contains two structural phases, ferrite, and martensite. The ferritic phase gives good formability and is relatively soft. The martensitic phase is the strongest phase, giving the metal strength and structural integrity. While the martensitic phase is stronger than the ferrite, it is also more brittle [8]. The annealing makes the steel sensitive to heat treatment over 200 °C [9].

The Docol 1000DP that was used in this study had been cold-rolled and is without an electro- galvanized coating. It has a yield strength of 700-950 MPa and tensile strength of 1000- 1200 MPa, tested transverse to the rolling direction. The chemical composition is shown in Table 2.1 [10]. Physical properties such as thermal conductivity and heat capacity increase with the temperature, see Table 2.2 [11].

(9)

5

Table 2.1: The chemical composition of Docol 1000DP (Fe is balance).

Mass % C Si Mn P S Al Nb +

Ti

Cr + Mo

B Cu

Docol 1000DP

0.18 0.80 1.80 0.020 0.010 0.015- 1.00

0.10 1.40 0.005 0.20

Table 2.2: The thermal conductivity and heat capacity at different temperatures for Docol 1000DP.

Temperature 344 °C 464 °C

Thermal conductivity [W/m°C] 33.2 42.3

Heat capacity [J/kg°C] - 122

2.3.2 LDX 2101

LDX 2101 is duplex stainless steel with two structural phases, austenite and ferrite. The ferritic phase gives good formability and ductility. The austenitic phase is harder than the ferritic phase and contributes to structural integrity. LDX 2101 stainless steel is cold-rolled. It has high strength, good corrosion resistance, and mechanical properties. The chemical composition is shown in Table 2.3 [12].

Table 2.3: The chemical composition of LDX 2101 (Fe is balance).

Mass % C N Cr Ni Mo Others

LDX 2101 0.03 0.22 21 1.5 0.3 5 Mn

Exposure to high temperatures can cause changes in the microstructure of the stainless steel.

The sigma phase is formed at the temperatures in the range of 600-950 °C and when the temperature decreases to 350-525 °C formation of ferrite occurs. At 475 °C the steel is at risk for embrittlement due to a miscibility gap with two ferritic phases. The risk is low for welding and heat-treatment operations but increases when working on thicker materials with slow cooling. Heat treatments at 500-550 °C should be followed by rapid cooling to prevent embrittlement. The physical properties as thermal conductivity and thermal capacity increase with the temperature, see Table 2.4 [12].

(10)

6

Table 2.4: The linear expansion at room temperature to a given temperature and the thermal conductivity and heat capacity at different temperatures for LDX 2101.

Temperature 20

°C

100

°C

200

°C

300

°C Linear expansion at (Room temperature to

temperature) [x10^-6/°C] - 13.0 13.5 14.0

Thermal conductivity [W/m°C] 15 16 17 18

Heat capacity [J/kg°C] 500 530 560 590

(11)

7

3 Experiments

The methods chosen for this study, when examining the heat affected zone and the surface quality in laser cut samples, are presented below.

3.1 Sample preparation

The geometry chosen for the samples was the shape of dog bones for tensile testing, see Figure 3.1. This is as the heat affected zone can differ with the geometry and the aim of this study is to guarantee the quality of tensile testing. The two materials were cut in the shape of dog bones, cut with different laser power and cutting speed. Fourteen samples of Docol 1000DP and ten samples of LDX 2101 were made. The materials had a thickness of 1,5 mm. The laser cutting nozzle had a diameter of 1,5 mm and the focal point was set on the upper surface of the materials. Two commonly used laser powers were chosen, 2500 W and 3500 W. The samples were cut with different cutting speeds, see Table 3.1 to 3.2 for the Docol 1000DP samples and Table 3.3 to 3.4 for the LDX 2101 samples. The cutting speed was chosen empirically, starting at a lower cutting speed around two to three m/min increasing until the laser was unable to cut through. This was done in order to find the highest possible cutting speed for each material and laser power.

Figure 3.1: The geometry of the samples cut with the laser cutting machine.

Table 3.1: The test matrix for the Docol 1000DP samples with the laser power 2500 W and varied cutting speed.

Sample name P6 P5 P7 P8 P9 P13 P14

Cutting speed [m/min] 2 2.5 3 3.5 4 6 7

Table 3.1 shows the cutting speeds used to cut the Docol 1000DP samples using a laser power of 2500 W. The lowest cutting speed was 2 m/min and the highest cutting speed was 7 m/min.

(12)

8

Table 3.2: The test matrix for the Docol 1000DP samples with the laser power 3500 W and varied cutting speed.

Sample name P3 P1 P2 P4 P10 P11 P12

Cutting speed [m/min] 2.5 3 3.5 4 5 7.5 10

Table 3.2 shows the cutting speeds used to cut the Docol 1000DP samples using a laser power of 3500 W. The lowest cutting speed was 2,5 m/min and the highest cutting speed was 10 m/min.

Table 3.3: The test matrix for the LDX 2101 samples with the laser power 2500 W and varied cutting speed.

Sample name A8 A6 A7 A9 A10

Cutting speed [m/min] 2 4 6 8 9

Table 3.3 shows the cutting speeds used to cut the LDX 2101 samples using a laser power of 2500 W. The lowest cutting speed was 2 m/min and the highest cutting speed was 9 m/min.

Table 3.4: The test matrix for the LDX 2101 samples with the laser power 3500 W and varied cutting speed.

Sample name A4 A1 A5 A2 A3

Cutting speed [m/min] 3 4 5 7.5 10

Table 3.4 shows the cutting speeds used to cut the LDX 2101 samples using a laser power of 3500 W. The lowest cutting speed was 3 m/min and the highest cutting speed was 10 m/min.

Three positions on the dog bone samples were analyzed. The first position, position A, was located in the upper right corner, the second, position B, was located in the lower right bend and position C in the middle of the sample, see Figure 3.2. The laser cutting machine starting point was at the lower left end of the sample and then cut in an upwards direction to the right.

The material absorbs heat from the laser beams and when cutting on the lower side of the sample, the material will have absorbed more energy. Thus, the three positions were located on the lower side of the samples as more energy had been absorbed, which could result in a larger heat affected zone.

(13)

9

Figure 3.2: The specimen’s geometry, showing the laser cutting machine’s starting point, the cutting direction, and the three analyzed positions.

The samples were cut through positions A, B, and C in order to analyze the HAZ in the different places in the geometry, creating three pieces of each sample. The three pieces were cast in thermoplastic, using TransOptic powder. The samples were ground, polished, and etched. The Docol 1000DP samples were etched with Nital 2 % for 15 seconds. The LDX 2101 samples were polished just before they were etched with Beraha for 15 seconds twice. The chemical composition of Beraha was 1 g Potassium Disulfate, 40 ml Hydrochloric acid 37 %, and 60 ml H2O. As Beraha only is active for a short period of time, new etchant was mixed after 4-5 samples had been etched.

3.2 Characterization methods

To analyze the samples microstructure Light Optical Microscopy (LOM) was used. LOM is microscopy that illustrates the sample through a series of convex lenses and creates a magnified image. The microscope is equipped with a camera and is connected to a computer. The magnified image facilitates the study of the specimen’s microstructure [13].

The LOM images were saved with a program called Kappa. The program made it possible to measure the distance between two selected points with a preset scale. The length of the HAZ was measured in four positions on each sample. Position 1 was located in the upper corner, position 2 was located a quarter from the upper corner, position 3 was located a quarter from the second and position 4 in the lower corner, see Figure 3.3. This in order to get an average as well as being able to study how the size of the HAZ differs over the cutting surface.

(14)

10

Figure 3.3: The length of the HAZ measured in position 1-4. Sample P3 position A.

The angularity of the cutting surface of the samples was measured with the program Fiji: ImageJ. The program is an open-source program for image processing, allowing the angle between two selected points in an image to be measured with high accuracy. Figure 3.4 shows how the angle was measured.

Figure 3.4: Sample P3 position A, illustrating the measurement of the surface angularity.

(15)

11

4 Results

The results for each material respectively are shown below. The HAZ, the surface angularity, and the surface quality was examined.

4.1 Docol 1000DP

4.1.1 Heat affected zone

The heat affected zone in the Docol 1000DP samples are presented in Table 4.1 to 4.2 below, showing the average length of the zone.

Table 4.1: The average length of the HAZ in position A-C. Samples cut with a laser power of 2500 W.

Sample P6 P5 P7 P8 P9 P13 P14

Position A 0.274 0.300 0.246 0.223 0.193 0.207 0.159

Position B 0.256 0.230 0.174 0.160 0.133 0.162 0.112

Position C 0.234 0.195 0.178 0.167 0.184 0.141 0.150

Table 4.1 shows the average length of the HAZ with increasing cutting speed. The smallest HAZ in position A was found in sample P14 as well as for position B. The smallest HAZ in position C was found in sample P13 and the next smallest in P14. The average length of the HAZ in each position was calculated to the following; position A - 0.229, position B - 0.175, and position C - 0.178 mm. The average length of the HAZ decreases with increasing cutting speed, see Figure 4.1. The length has decreased for all positions.

(16)

12

Figure 4.1: The relation between the average length of the HAZ and the cutting speed.

Table 4.2: The average length of the HAZ in position A-C. Samples cut with a laser power of 3500 W.

Sample P3 P1 P2 P4 P10 P11 P12

Position A [mm] 0.288 0.188 0.263 0.288 0.233 0.143 0.132 Position B [mm] 0.244 0.229 0.208 0.224 0.142 0.160 0.200 Position C [mm] 0.234 0.160 0.156 0.200 0.240 0.188 0.140

Table 4.2 shows the average length of the HAZ with increasing cutting speed. The smallest HAZ for position A as well as for position C was found in sample P12. The smallest HAZ for position B was found in sample P10 and the next smallest in P11. The average length of the HAZ in each position was calculated to the following; position A - 0.219 mm, position B - 0.172 mm, and position C - 0.188 mm. The average length of the HAZ decreases overall with increased cutting speed, see Figure 4.2.

0 0,05 0,1 0,15 0,2 0,25 0,3 0,35

2 2.5 3 3.5 4 6 7

Length [mm]

Cutting speed [m/min]

HAZ DOCOL 1000DP 2500 W

A B C

(17)

13

Figure 4.2: The relation between the average length of the HAZ and the cutting speed.

4.1.2 Surface angularity

The angularity of the cutting surface is presented below in Table 4.3 to 4.4, showing the measured angle of the cutting surface.

Table 4.3: The cutting surface angle in position A-C. Samples cut with a laser power of 2500 W.

Sample name P6 P5 P7 P8 P9 P13 P14

Position A [°] 4.3 7.3 9.5 5.5 4.8 7.5 5.1

Position B [°] 3.2 5.5 5.2 4.1 3.0 6.8 6.5

Position C [°] 6.6 5.3 4.9 5.9 4.3 5.4 6.7

Table 4.3 shows the angle of the cutting surface with increasing cutting speed for the samples cut with 2500 W. The smallest angle for position A was sample in P6, for position B and C the smallest angle was found in sample P9. The average on each position was calculated to the following; position A - 6.3, position B - 4.9, and position C - 5.6 degrees.

0 0,05 0,1 0,15 0,2 0,25 0,3 0,35

2.5 3 3.5 4 5 7.5 10

Length [mm]

Cutting speed [m/min]

HAZ DOCOL 1000DP 3500 W

A B C

(18)

14

Table 4.4: The cutting surface angle in position A-C. Samples cut with a laser power of 3500 W.

Sample name P3 P1 P2 P4 P10 P11 P12

Position A [°] 7.7 2.6 4.9 3.1 6.9 2.9 4.0

Position B [°] 8.6 2.6 3.8 6.7 7.0 5.7 6.4

Position C [°] 1.4 2.8 3.0 6.5 9.7 5.5 5.5

Table 4.4 shows the angle of the cutting surface with increasing cutting speed for the samples cut with 3500 W. The smallest angle for position A and B was in sample P1, for position C the smallest angle was found in sample P3. The average on each position was calculated to the following; position A - 4.6, position B - 5.8, and position C - 4.9 degrees.

4.1.3 Surface quality

Cracks were identified on the cutting surface on the samples P5 and P10, see Figure 4.3 to 4.4.

Figure 4.3: Crack on the cutting surface. Sample P5 position A.

(19)

15

Figure 4.4: Crack on the cutting surface. Sample P10 position A.

Figures 4.3 and 4.4 show cracks in the cutting surface of sample P5 position A and sample P10 position A.

Figure 4.5: Sample P6 (left) cut with a cutting speed of 2 m/min and a laser power of 2500 W.

Sample P14 (right) cut with a cutting speed of 7 m/min and a laser power of 2500 W.

(20)

16

Figure 4.6: Sample P3 (left) cut with a cutting speed of 2,5 m/min and a laser power of 3500 W.

Sample P12 (right) cut with a cutting speed of 10 m/min and a laser power of 3500 W.

Figure 4.5 and 4.6 shows the amount of dross on the samples cut with the lowest cutting speed and the highest cutting speed for each laser power respectively. When increasing the cutting speed, a deviation in the geometry was detected. The corners were softened, and the straight parts became wavy. The geometry deviated more with increasing cutting speed, this is shown in Figures 4.5 and 4.6.

4.2 LDX 2101

4.2.1. Heat affected zone

The HAZ for the LDX 2101 samples consists largely of ferrite and can be identified as the darker area at the cutting surface when etched with Beraha, see example in Figure 4.7. The size of the heat affected zone was very small in all samples, in the order of 50 μm. The size differed very little between the three positions, see Figures 4.7 to 4.9.

(21)

17

Figure 4.7: The dark area on the cutting surface shows the length of the HAZ. Sample A6 position A.

Figure 4.8: The dark area on the cutting surface shows the length of the HAZ. Sample A6 position B.

(22)

18

Figure 4.9: The dark area on the cutting surface shows the length of the HAZ. Sample A6 position C.

4.2.2 Surface angularity

The angularity of the cutting surface is presented below in Table 4.5 to 4.6, showing the measured angle of the cutting surface.

Table 4.5: The cutting surface angle in position A-C. Samples are cut with 2500 W laser power.

Sample name A8 A6 A7 A9 A10

Position A [°] 8.5 3.0 4.9 5.1 3.3

Position B [°] 17.7 5.9 5.1 7.2 3.5

Position C [°] 6.8 4.9 3.7 2.3 2.6

Table 4.5 shows the angle of the cutting surface with increasing cutting speed for the samples cut with 2500 W. The smallest angle for position A was in sample A6, for position B the smallest angle was found in sample A10 and for position C sample A9 had the smallest angle.

The average on each position was calculated to the following; position A - 5.0, position B - 7.9, and position C - 4.1 degrees.

(23)

19

Table 4.6: The cutting surface angle in position A-C. Samples are cut with 3500 W laser power.

Sample name A4 A1 A5 A2 A3

Position A [°] 4.3 10.2 15.6 5.3 3.5

Position B [°] 9.8 6.9 10.3 5.7 4.8

Position C [°] 12.0 6.4 4.5 6.1 2.2

Table 4.6 shows the angle of the cutting surface with increasing cutting speed for the samples cut with 3500 W. The smallest angle was found in sample A3 for all three positions. The average angle on each position was calculated to the following; position A - 7.8, position B - 7.5, and position C - 6.2 degrees.

4.2.3 Surface quality

No cracks were found on the LDX 2101 samples.

Figure 4.10: Sample A8 (left) cut with a cutting speed of 2 m/min and a laser power of 2500 W.

Sample A10 (right) cut with a cutting speed of 9 m/min and a laser power of 2500 W.

(24)

20

Figure 4.11: Sample A4 (left) cut with a cutting speed of 3 m/min and a laser power of 3500 W.

Sample A3 (right) cut with a cutting speed of 10 m/min and a laser power of 3500 W.

Figures 4.10 and 4.11 show the amount of dross on the sample cut with the lowest cutting speed and the sample cut with the highest cutting speed for each laser power respectively. When increasing the cutting speed, the deviation in the geometry was detected. The corners were softened, and the straight parts became wavy. The geometry deviated more with increasing cutting speed.

(25)

21

5 Discussion

5.1 Docol 1000DP

The results show that the heat affected zone decreased with increased cutting speed when cutting with a laser power of 2500 W, see Figure 4.1. However, the relation was not as evident when cutting with 3500 W, see Figure 4.2. The curves are not as consistent, but there’s an overall decrease in length of the HAZ. The smallest HAZ was for positions A and C found in sample P12 with the highest cutting speed, strengthening the theory. When studying each sample individually it was found that there’s a lot of dross on sample P12 in position B, which could increase the HAZ. Overall, the largest HAZ was measured to 0.30 mm, as the surface can be crooked, removing around 0.50 mm with a lathe would be sufficient to remove the HAZ with a margin when using lower cutting speed. When increasing to maximum cutting speed, removing around 0.30 mm would be sufficient to remove the HAZ with margin.

When cutting with a laser power of 3500 W the heat affected zone is larger in comparison with samples cut with a laser power of 2500 W using the same cutting speed. This is the result of more energy being absorbed when cutting with 3500 W. When studying the samples cut with the maximum cutting speed of each laser power, sample P14, and P12, the HAZ was marginally larger in samples cut with 2500 W for position A and C. However, in position B the HAZ was much larger for 3500 W. This is likely due to the previously mentioned dross on sample P12 in position B.

The different positions on the sample, position A, B, and C show that there is a geometrical parameter to consider when using laser cutting. When cutting with a laser power of 2500 W the heat affected zone was smallest in position B, next to smallest in position C and largest in position A. The samples cut with a laser power of 3500 W showed that the HAZ had its smallest value in position A, followed by position C and then position B. However, when taking an average on each position it showed that the HAZ often is larger in position A, and almost equally small in position B and C. This shows that the extent of the HAZ increases in tight corners, but also that the curvature in position B is not enough to cause noticeable changes.

This seems to be the case regardless of using a laser power of 2500 W or 3500 W.

The surface angularity varied a lot between samples without showing a pattern. The smallest angles were generally found in samples with lower cutting speeds, but the results are not consistent enough to draw a conclusion. However, position A was shown to be the position with the most vertical cut. Increasing the cutting speed resulted in unstable movements and deviation in the sample geometry. It can also be argued this affects the surface angularity and its lack of consistency.

Cracks were seen on the surface of the Docol 1000DP samples P5 and P10 in position A, see Figures 4.3 and 4.4, reducing the surface quality. The cracks can act as initiators for fractures and can cause stresses in the material. The crack on the surface of sample P5 can be the result

(26)

22

of too rapid cooling and stresses in the material. The crack on the surface of sample P10 could be a pocket formed when the molten material was cooled on the cutting surface. The cutting surface near the crack is uneven, which supports this argument. Cracks and residual stresses can be avoided by optimizing the gas parameters.

Figure 4.5 and 4.6 shows that the amount of residual dross left on the cutting surface decreases with increased cutting speed. A higher cutting speed creates a narrower cut, resulting in a cleaner surface. This is the case for both laser powers.

5.2 LDX 2101

The results show that the LDX 2101 stainless steel was affected to a much smaller extent. The size of the HAZ was in the order of 50 μm, see Figures 4.7 to 4.9, which made it difficult to measure. This is likely due to the steel's low thermal conductivity preventing heat transfer. The small zone in combination with indistinct borders hindered measurements. Thus, no exact measurements were made for each sample as it would not be accurate. All samples were studied in LOM and no samples deviated from the size order of 50 μm. There was negligible difference between the samples cut with a laser power of 2500 W or 3500 W. A laser power of 3500 W allows higher cutting speeds and therefore higher efficiency. Based on the results, removing 0.15 mm would be sufficient to remove the HAZ with margin.

The surface angularity varied a lot in positions A-C in each sample. The samples cut with 2500 W laser power were generally more vertical than the samples cut with 3500 W. The surface angularity was the most vertical for the samples cut with high speed. This is in agreement with known theory, specimens cut with higher cutting speed will have a narrower cut resulting in a more vertical surface. Position C had the most vertical average, and this follows the theory in the introduction stating that a straight line is the most beneficial geometry to cut for the cutting surface quality.

The results show that the amount of dross left on the cutting surface decreases with an increased cutting speed for both laser powers, see Figure 4.10 and 4.11. This is in agreement with the known theory stating that higher cutting speed results in less dross.

5.3 Sources of error

A source of error in this study is the number of samples. This study only analyzes one sample for every combination of parameters. By doing multiple samples with the same combination of parameters, the accuracy of the results would increase.

Another source of error is the cutting geometry. The geometry was affected as the cutting speed increased, causing the corners to be softer as well as the bend in position B being straightened.

This contributes to uncertainties in the results. Analyzing multiple geometries with different

(27)

23

radii would increase the accuracy and would implicate how much this deviation in geometry affects the size of the HAZ.

The accuracy of the measurements made with the following measuring programs; Kappa and Fiji: ImageJ, can be questioned. The Kappa program can provide images with weak contrasts which makes it difficult to identify the length of the heat affected zone and can lead to inaccurate results. This together with the fact that the extent of the HAZ and where it ends often is hard to distinguish, contributes to uncertainty in the results. The program Fiji: ImageJ can enlarge the source of human error when the measuring points are not set precisely in the corners of the sample, this could contribute to inaccurate results. The program gives four value figures and two were chosen to reduce the error. No margin of error could be calculated or estimated.

5.4 Sustainability and ethical aspects

Laser cutting is an efficient method for cutting specimens for tensile testing, which is vital for examining a material’s mechanical properties. Laser cutting is both time and cost efficient in comparison with turning or electrical discharge machining that’s commonly used. Through improving laser cutting for tensile testing specifically and examining how it affects the material, more data and knowledge will be obtained in a shorter time and with less material losses.

There’s an ethical aspect as companies need to be able to guarantee the quality and accuracy of the results obtained.

(28)

24

6 Conclusions

After analyzing the results, the following can be concluded:

6.1 Docol 1000DP

-

The heat affected zone decreases with increased cutting speed.

-

Negligible differences in heat affected zone were observed between the samples cut with 2500 W and 3500 W. However, cutting with 3500 W is preferable from a production point of view as it allows higher cutting speed and thus, higher efficiency.

-

When cutting with the maximum cutting speed, turning away 0.30 mm is sufficient to remove the heat affected zone with margin.

-

The amount of dross decreases with increased cutting speed for both laser powers.

6.2 LDX 2101

-

When cutting with the maximum cutting speed, turning away 0.15 mm is sufficient to remove the heat affected zone with margin.

-

The amount of dross decreases with an increased cutting speed for both laser powers.

(29)

25

7 Acknowledgements

We would like to pay tribute to the research institute SWERIM AB that funded this project by providing us with material and equipment. We would like to thank our two supervisors David Franklin, M.Sc., IWE, and EPW from the department of joining at SWERIM AB and Elias Repper, IWE from the department of joining at SWERIM AB for their support and guidance through this project. We would also like to thank our supervisor Joakim Odqvist, from the Department of Materials Science and Engineering at the Royal Institute of Technology for his support and guidance through this project.

(30)

26

8 References

[1] Machinemfg, (2020). The Ultimate Guide to Fiber Laser Cutting. [online]

URL: https://www.machinemfg.com/fiber-laser-cutting/#1_Good_cutting_quality [Accessed 5 March. 2020].

[2] Ion, John C. (2005). Laser processing of engineering materials principles, procedure, and industrial application. Oxford: Butterworth-Heinemann, pp. 354.

[3] Machinemfg, (2020). Basics of Laser Cutting (Knowledge You Must Know). [online]

URL: https://www.machinemfg.com/laser-cutting-basics/ [Accessed 5 March. 2020].

[4] Precitec KG, (2011). SolidCutter Lasermatic® Z, Compact cutting head for Nd: YAG robot applications. [pdf] The Federal Republic of Germany. URL:

https://www.precitec.de [Accessed 5 March. 2020].

[5] Machinemfg, (2020). Laser Cutting Quality Control (With Proved Solutions). [online]

URL: https://www.machinemfg.com/laser-cutting-quality-control/

[Accessed 5 March. 2020].

[6] Machinemfg, (2020). 9 Standard to Check Laser Cutting Quality. [online]

URL: https://www.machinemfg.com/check-laser-cutting-quality/

[Accessed 5 March. 2020].

[7] The Fabricator, (2010). The importance of focal positions in laser cutting. [online]

URL: https://www.thefabricator.com/thefabricator/article/lasercutting/the-importance- of-focal-positions-in-laser-cutting [Accessed 8 May. 2020].

[8] SSAB, (2020). Docol 1000DP. [online]

URL: https://www.ssab.se/produkter/varumarken/docol/products/docol-1000dp [Accessed 5 March. 2020].

[9] SSAB, (2020). Docol, Product offer. [online]

URL: https://www.ssab.com/products/brands/docol/docol-product-overview [Accessed 5 March. 2020].

[10] SSAB, (2017). Data sheet 2106 Docol 1000DP. [pdf]. URL:

www.ssab.com [Accessed 5 March. 2020].

[11] BBN Steel Stores, (2019). Q/B Docol 1000 DP. [online]

URL: https://www.steelestores.com/grade/qb-docol-1000-dp.html [Accessed 5 March. 2020].

(31)

27

[12] Outokumpu, (2007). Downloads, Duplex Series. [pdf] Avesta: Centrum Tryck AB. URL: http://www.outokumpu-armetal.com/ [Accessed 5 March. 2020].

[13] Holgate, J.H., and Webb, J. (2003). Microscopy, Light Microscopy, and Histochemical Methods. Encyclopedia of Food Sciences and Nutrition, 2:nd ed.

UK: Elsevier Science Ltd, pp. 3917-3922.

(32)

28

Appendix A: Docol 1000DP

2500 W

The pictures are placed starting with the lowest cutting speed for each laser power respectively.

P6

Figure 1. Sample P6 position A (left), position B (middle) and position C (right).

P5

P7

(33)

29 P8

P9

P13

P14

(34)

30 3500 W

P3

P1

P2

(35)

31 P4

P10

P11

P12

(36)

32

Appendix B: LDX 2101

The pictures are placed starting with the lowest cutting speed for each laser power respectively.

2500 W A8

Figure 1. Sample A8 position A (left), position B (middle) and position C (right).

A6

A7

(37)

33 A9

A10

3500 W A4

(38)

34 A1

A5

A2

A3

(39)

TRITA ITM-EX 2020:305

www.kth.se