Effect of conformal
cooling in Additive
Manufactured inserts
on properties of high
pressure die cast
aluminum component
PAPER WITHIN Materials and Manufacturing AUTHOR: Ruslan Sevastopolev
TUTOR:Anders Jarfors and Seshendra Karamchedu
Postadress: Besöksadress: Telefon:
Box 1026 Gjuterigatan 5 036-10 10 00 (vx)
year program.
The author takes full responsibility for opinions, conclusions and findings presented. Examiner: Roland Stolt
Supervisor: Anders Wollmar Jarfors Scope: 15 credits (first cycle)
Abstract
Additive manufacturing can bring several advantages in tooling applications especially hot working tooling as high pressure die casting. Printing of conformal cooling channels can lead to improved cooling and faster solidification, which, in turn, can possibly result in better quality of the cast part. However, few studies on advantages of additive manufactured tools in high pressure die casting are published.
The aim of this study was to investigate and quantify the effect of conformal cooling on microstructure and mechanical properties of high pressure die cast aluminum alloy. Two tools each consisting of two die inserts were produced with and without conformal channels using additive manufacturing. Both tools were used in die casting of aluminum alloy. Aluminum specimens were then characterized microstructurally in light optical microscope for secondary arm spacing measurements and subjected to tensile and hardness testing. Cooling behavior of different inserts was studied with a thermal camera and by monitoring the temperature change of cooling oil during casting. Surface roughness of die inserts was measured with profilometer before and after casting.
Thermal imaging of temperature as a function of time and temperature change of oil during casting cycle indicated that conformal insert had faster cooling and lower temperature compared to conventional insert. However, thermal imaging of temperature after each shot in a certain point of time showed higher maximum and minimum temperature on conformal die surface but no significant difference in normalized temperature gradient compared to the conventional insert.
The average secondary dendrite arm spacing values were fairly similar for samples from conventional and conformal inserts, while more specimens from conventional insert demonstrated coarser structure. Slower cooling in conventional insert could result in the coarser secondary dendrite arm spacing.
Tensile strength and hardness testing revealed no significant difference in mechanical properties of the specimens cast in conventional and conformal die inserts. However, reduced deviations in hardness was observed for samples cast with conformal insert. This is in agreement with secondary dendrite arm spacing measurements indicating improved cooling with conformal insert.
Surface roughness measurement showed small wear of the inserts. More castings are needed to observe a possible difference in wear between the conventional and conformal inserts.
Small observed differences in cooling rate and secondary arm spacing did not result in evident difference in mechanical properties of the aluminum alloy but the variation in properties were reduced for samples cast with conformal cooling. Future work may include more accurate measurement of cooling behavior with a thermocouple printed into the die insert, casting of thicker specimen for porosity evaluation and fatigue testing and longer casting series to evaluate the influence of conformal cooling on tool wear.
Sammanfattning
Additiv tillverkning kan ge flera fördelar i verktygstillämpningar speciellt varmbearbetningsverktyg som gjutning med högt tryck. Printing av konforma kylkanaler kan leda till förbättrad kylning och snabbare stelning, vilket i sin tur kan resultera i bättre kvalitet på den gjutna detaljen. Dock publiceras få studier om fördelarna med additivt tillverkade gjutnings verktyg.
Syftet med denna studie var att undersöka och kvantifiera effekten av konform kylning på mikrostruktur och mekaniska egenskaper hos gjuten aluminiumlegering. Två verktyg vardera bestående av två insatser tillverkades med och utan konforma kylkanaler med hjälp av additiv tillverkning. Båda verktygen användes för gjutning av aluminiumlegering. Aluminiumprover karakteriserades sedan i ljust optiskt mikroskop för sekundära dendrit armavståndsmätningar och utsattes för drag- och hårdhetsprovning. Kylbeteende hos olika insatser studerades med en termisk kamera och genom att övervaka temperaturförändringen på kylolja under gjutning. Ytjämnheten hos insatserna mättes med profilometer före och efter gjutning.
Termisk bild av temperaturen som en funktion av tid och temperaturförändring av olja under gjutningscykeln indikerade att konforma insatser hade snabbare kylning och lägre temperatur jämfört med konventionellt insats. Termisk avbildning av temperaturen efter varje skott under en viss tidpunkt visade högre maximala och lägsta temperatur på yta av den konforma insatsen, men ingen större skillnad i normaliserad temperaturgradient jämfört med den konventionella insatsen.
De genomsnittliga värdena för sekundära dendritarmavstånd var ganska lika för prover från konventionella och konforma insatser, medan fler prover från konventionella insatsen visade grövre struktur. Långsammare kylning i den konventionella insatsen kunde resultera i det grova sekundära dendritarmavståndet.
Draghållfasthet och hårdhetstest avslöjade ingen signifikant skillnad i mekaniska egenskaper hos prov gjutna i konventionella och konforma insatser. Emellertid observerades reducerade avvikelser i hårdhet för prover gjutna med konforma insatser. Detta är i överensstämmelse med sekundära dendrit armavståndsmätningar som indikerar förbättrad kylning med den konforma insatsen.
Ytjämnhetsmätning visade litet slitage på insatserna. Flera gjutningar behövs för att observera en eventuell skillnad i slitage mellan konventionella och konforma insatserna. Små observerade skillnader i kylningshastighet och sekundärt armavstånd resulterade inte i en tydlig skillnad i mekaniska egenskaper hos aluminiumlegeringen men variationen i egenskaper minskades för prover gjutna med konform kylning. Framtida arbete kan omfatta en mer exakt mätning av kylbeteende med ett termoelement inprintad i insatserna, gjutning av tjockare prov för porositetsbedömning och utmattningstestning och längre gjutningsserier för att utvärdera påverkan av konform kylning på verktygsslitage.
Keywords
High Pressure Die Casting (HPDC), Additive Manufacturing (AM), Aluminum (Al), Tensile test, Cooling channels, Conformal cooling, Laser Based Power Bed Fusion of Metals (PBF-LB/M), Microstructure, Secondary Dendrite Arm Spacing (SDAS).
Acknowledgement
I would like to express my gratitude to Uddeholms AB and Jönköping University for giving me the opportunity to be a part of this work.
I would also like to thank everyone who has been directly or indirectly involved in this project. Special thanks to my supervisors Anders Jarfors and Seshendra Karamchedu, this work would not be possible without your support and patience. My sincerely gratitude also goes to Qing Zhang and Jorge Santos from Jönköping University, Ronny Karlsson from IM Tools, Henrik Andersson and Panos Katsanos from Uddeholms AB for their practical support.
Contents
1.
Introduction ... 6
1.1BACKGROUND ... 6
1.2PURPOSE AND RESEARCH QUESTIONS ... 7
1.3DELIMITATIONS ... 7
1.4OUTLINE ... 7
2.
Theoretical background ... 9
2.1HIGH PRESSURE DIE CASTING ... 9
2.2ALUMINIUM ALLOYS ... 11
2.2.1 Types and designations ... 11
2.2.2 Heat treatment ... 13
2.2.3 Influence of cooling rate on microstructure during solidification ... 14
2.2.4 Influence of cooling rate on mechanical properties ... 16
2.3TOOL STEEL ... 17
2.3.2 Tool steel manufacturing ... 19
2.3.3 Tool steel powder manufacturing ... 21
2.4TOOL AND DIE MANUFACTURING ... 22
2.4.1 Additive manufacturing ... 23
3.
Experimental ... 27
3.1ALUMINIUM ALLOY FOR CASTING OF COMPONENT ... 27
3.2TOOL AND TOOL MANUFACTURING ... 27
3.3HIGH PRESSURE DIE CASTING ... 31
3.4HEAT TREATMENT ... 34
3.5MECHANICAL PROPERTIES ... 35
3.6MICROSTRUCTURE INVESTIGATION ... 35
3.7TOOL INVESTIGATION ... 37
4.
Results and discussion ... 39
4.1CHEMICAL COMPOSITION ... 39
4.2COOLING BEHAVIOUR ... 39
4.4INFLUENCE OF COOLING ON MECHANICAL PROPERTIES ... 45
4.5SURFACE ROUGHNESS OF TOOLING INSERTS ... 47
4.6METHODS DISCUSSION ... 47
5.
Conclusions and future work ... 49
6.
References ... 51
1. Introduction
This thesis work is a part of larger cooperation work between Uddeholms AB in Hagfors and Jönköping University and mandatory final project work of the one-year Foundry Master Program at Jönköping University.
Nowadays, additive manufacturing (AM) is the industrial trend and becoming more common as a manufacturing technology proposing advantages and challenges we need to understand more. With additive manufacturing components of complex design can be 3D-printed to final shapes and lead-time is significantly decreased [1]. Additional advantage is the possibility to reduce stock of spare parts and avoid capital binding in spare parts by printing them in house on demand [2].
The investment cost in a 3D- printer is comparable to the investment cost of the metal cutting machine [3]. However, the printing time is longer than machining time resulting in the higher cost of the final part in many cases [3].
Additive manufacturing can bring several advantages in tooling applications [3], especially hot working tooling as high pressure die casting. Printing of conformal cooling channels can lead to improved cooling and faster solidification, which, in turn, result in higher productivity and better quality of the cast part.
In order to investigate and quantify the effect of conformal cooling on the properties of high-pressure die cast aluminum two tools each consisting of two die inserts (fixed and ejector) were produced by additive manufacturing – one with traditionally manufactured cooling channels and another with conformal channels. Both tools were used in die casting of aluminum and the aluminum samples were then characterized microstructurally for solidification characteristics and subjected to mechanical testing. The results obtained were then interpreted to understand the impact of AM tooling on the aluminum alloy properties. Literature survey was performed to check coinciding of the obtained results with the similar work.
1.1 Background
Voestalpine AG with headquarter in Linz is a big technology group represented worldwide and consists of four divisions such as Steel, High Performance Metals, Metal Engineering and Metal Forming [4]. The company Uddeholms AB in Hagfors is a part of the group in the High Performance Metals Division and a producer and supplier of highly alloyed tool steels. Uddeholms AB provides products in four tooling application areas, such as tool steels for plastic moulding, special components, hot and cold work tool steels. Uddeholms AB has also knowledge and competence in many areas such as heat treatment, machining, forming and punching, additive manufacturing but also own research and development department making investigation of different properties in tool steels [5].
Since several years back Uddeholms AB has invested in production of steel powder for additive manufacturing and developed several powder products for tooling applications [6]. To perform proper recommendations and printing parameters for the products and to the customers R&D department has also invested in a laser based powder bed fusion of metals (PBF-LB/M) printer. Obtaining of the printer has led to new opportunities and cooperation in new business areas. For example, to evaluate properties of additive manufactured tools has let to collaboration with Jönköping University that has high knowledge and expertise in foundry technology. The Jönköping University has a high pressure die casting (HPDC) machine and PBF-LB/M printer where the experiments can be performed.
Usually, high pressure die casting dies have cooling channels drilled from both sides. As dies usually have a complex shape, the cooling channels get difference in distance to the
surface resulting in somehow uneven cooling of the part during casting. Big advantage of additive manufacturing for casting dies is the possibility to place cooling channels along the die surface ensuring even cooling and solidification of the part. Conformal cooling can probably lead to faster cooling and more rapid solidification. It is well known that alloy properties can sometimes be improved by faster solidification [7]. However, limited amount of publications are available describing advantages and limitations of additive manufacturing for hot working tools. To check and investigate if conformal cooling leads to improvement in cooling efficiency and component properties of high pressure die cast aluminum alloy, it was decided to investigate two types of tools each consisting of two die inserts that were produced by additive manufacturing method. The inserts were similar in design, material, manufacturing method, the only difference was that one pair of the inserts had conformal cooling channels. Aluminum specimens were cast in both tools with similar casting parameters and microstructure and mechanical properties of the specimens were compared.
1.2 Purpose and research questions
The purpose of this work was to investigate the effect of conformal cooling in additive manufactured inserts on properties of high-pressure die cast aluminum.
The research questions were:
How does conformal cooling influence microstructure of aluminum specimen cast in additive manufactured tool compared to the sample cast in the tool with traditionally produced cooling channels?
How does conformal cooling influence mechanical properties of aluminum specimen cast in additive manufactured tool compared to the sample cast in the tool with traditionally produced cooling channels?
1.3 Delimitations
This work did only cover investigation of aluminum component, which was in this case a tensile specimen, and did not cover a component of complex shape. Only one geometry for the aluminum specimen was chosen for tensile testing. The investigation of the aluminum specimen was limited to microstructure analysis and testing of mechanical properties and did not cover the investigation of physical or other properties.
1.4 Outline
The work described below is outlined in the following way:
Theoretical background contains a survey of current literature relevant to the research field and contains short description of HPDC process, data about the aluminum alloy used in the work, its microstructure and properties as well as some information about the tool steel and tool steel powder and additive manufacturing process. This section gives the basic information for better understanding of the following parts of the work.
Experimental section contains description of methods, experimental procedure and equipment used to perform the experimental work.
Results and discussion section contains the experimental results from the performed tests, the findings are analysed and evaluated together as well as discussed in the view of the objectives of the study.
Conclusions and future work section summarizes conclusions that can be drawn from the work including suggestions for future work to be able to answer the research questions more accurately or in full.
2. Theoretical background
2.1 High Pressure Die Casting
There are several methods to manufacture products from metals, the most common are casting, machining and sintering [8], Figure 1.
Figure 1. Most common manufacturing methods [9].
The casting methods include the following [10]: High pressure die casting – cold chamber High pressure die casting – hot chamber Low pressure die casting
Gravity casting Squeeze casting
Thixocasting/Rheocasting Sand casting
Tilt casting
High pressure die casting (HPDC) is a well-known process for manufacturing of advanced and lightweight components with good dimensional accuracy and excellent surface finish. The components are usually used in automotive and also other industries because of low-cost and high production efficiency. HPDC process is used for large series and has high productivity [11]. Two main types of HPDC processes exist: hot chamber and cold chamber process [12]. The hot chamber machines are usually used to cast brass, zinc and magnesium alloys. Aluminium alloys are usually cast in the cold chamber process, because of high melting temperature, application of this process is also increasing for magnesium alloys [10].
HPDC process is an injection casting technique i.e. molten metal is sprayed into the die cavity under high pressure to produce finished components requiring little machining, see Figure 2 [13]. The cold chamber HPDC machines consist of holder block and die cavities split into halves, piston/plunger and shot sleeve (a cylinder called a cold chamber), Figure 3. The moving die half has ejector pins that push the casting off the die cavity, Figure 3, [12]. The total cycle time can vary from 2 seconds to 1 minutes depending on machine type [12].
Cold chamber high pressure die casting process can be divided in 4 main stages: Manufacturing
methods Machinning
Casting
1. Clamping - the holder blocks are clamped together with a very high force.
2. Injection and pressurisation - piston injects the molten metal into the die through shot sleeve and pressurises the filled cavity at pressures from 6.9 MPa up to 137.9 MPa. 3. Cooling - rapid cooling and solidification of component due to cooling channels placed under the surface of the die cavity. The usual cooling media is preheated oil, but sometimes preheated water is also used.
4. Ejection - After full solidification the component is taken out from the die cavities with help of ejection pins. The die cavity is then sprayed with a water based lubricant that cools the die and helps extend die life. After that manufacturing cycle is repeated until some maintenance of the dies will be needed [12].
a b
Figure 2. a – die casting cold chamber machine overview [4], b – injection of molten metal
into die cavity [14].
Figure 3. Tooling for high pressure die casting [15].
1. Clamping plates 2. Holder plates 3. Die insert 4. Fixed insert 5. Cores 6. Sprue bushing 7. Sprue pin 8. Ejector pins
2.2 Aluminium alloys
Aluminium alloys exhibit many excellent properties such as light weight and easy to recycle which makes this material a good choice for future sustainable solutions [16]. It has good corrosion resistance, high thermal and electrical conductivity and high dimensional stability. Compared to other non-ferrous materials it is relatively easy to cast [7].
2.2.1 Types and designations
Aluminium alloys are usually classified into two groups: wrought alloys and cast alloys [17]. Wrought alloys are designed for products produced by rolling, forging, extruding or drawing such as sheet, plate wire, rod, bar, tube , pipe, forgings, angles, structural items, channels, and rolled and extruded shapes. Cast alloys are designed for casting where molten metal is filled in the form [18].
The most common alloying elements in aluminium alloys are silicon, iron, manganese and magnesium, Figure 4 [18].
Figure 4. Different types of aluminium alloys [19].
Silicon usually improves castability, fluidity, and feeding capability, but also decreases the possibility for appearing of hot tear. For the Al alloys with Si content with less than 12.6 % α-Al dendrites in the Al-Si eutectic will form during solidification. The solidification temperature of A356 used in this work is around 625ºC [11-12].
The aluminum-silicon alloys are divided into three groups depending on the Si content in the alloy: hypoeutectic with Si content less than <12wt%, eutectic with Si content of 12-13wt % and hypereutectic containing 14-25 wt% Si, Figure 5 [21]. Aluminum alloy A356 used in this work is hypoeutectic alloy containing 6.5-7.5 wt% Si.
a
b c d
Figure 5. a – Aluminium-silicon equilibrium phase diagram, b – hypoeutectic alloy,
microstructure, c – eutectic alloy microstructure, d – hypereutectic alloy microstructure [21].
Iron in aluminium alloys is mainly used to decrease the possibility of reaction between die surface and cast component, i.e. soldering and sticking effect [22]. Sometimes higher amounts of iron is used to reduce above mentioned failures, but at the same time the ductility and elongation will be decreased. The reason for that is ability of iron to build brittle intermetallic faces such as β-Al5FeSi [23]. To avoid detrimental effects of decreased elongation and ductility iron content should be kept as low as possible to avoid brittle Fe rich phases[19].
Manganese is very important for alloying of aluminium alloys because it can compensate for the detrimental effect of iron because the iron content decreases the elongation and ductility.. The ductility can be improved by adding Mn, and Mn content should be around half of the Fe content. This reduces reduce the risk of building brittle β-iron phases and promotes α-phase [24].
Aluminium alloys can also be alloyed with magnesium to get strength. Strength in aluminium alloy is achieved because of reaction between silicon and magnesium during heat treatment and by formation of precipitations such as Mg2Si. When adding Mg content
up to 0.5 wt.% the brittle β-Al5FeSi phase transforms to Al8Mg3FeSi6 resulting in improved strength, addition of Mg more than required will not result in any strength increase [25]. Casting alloy designations are built of four digit numerical designations used to identify aluminium alloys, Table 1 [26] .
The first digit shows the principal alloying element, which has been added to the aluminium alloy. The second two digits identify the aluminium alloy within the group, and the last number shows the product form. The digit following decimal point shows if alloy is a casting or an ingot. A capital letter prefix shows a modification to a specific alloy.
Table 1. Cast aluminium alloy designations [5], [8]
Alloy Series Principal Alloying Element
1xx.x Min 99.000% Aluminium
2xx.x Copper
3xx.x Silicon + Copper and/or Magnesium
4xx.x Silicon 5xx.x Magnesium 6xx.x Unused Series 7xx.x Zinc 8xx.x Tin 9xx.x Other Elements
The USA designation of the grade A356 used in this work stands for the following: A shows that alloy is modified, 3-shows that it is the silicon plus copper and/ or magnesium series, 56- identifies the alloy within three series [18].
The grade A356 has also European destinations such as EN AB-42000 and EN AC-42000. The alloys named above have similar chemical composition, the main alloying elements are silicon 6.5-7.5 wt%, magnesium 0.25 -0.65 wt% and titanium 0.05-0.25 wt% and a low iron content under 0.45 wt%, Table 2 [18].
Table 2. Chemical composition of cast aluminium alloy A356 in wt% (EN AB-42000 and
EN AC-42000) [10] Cu Fe Mg Mn Pb Si Sn Ti Zn A356 EN AB-42000 ≤0.15 ≤0.45 0.25-0.65 ≤0.35 ≤0.15 6.5-7.5 0- 0.05 0.05-0.20 ≤0.15 EN AC-42000 ≤0.20 ≤0.55 0.20-0.65 ≤0.35 ≤0.15 6.5-7.5 ≤0.05 0.05-0.25 ≤0.15 2.2.2 Heat treatment
After casting the aluminium component is usually post-processed to improve the mechanical properties. The basic heat treatment designations for the cast aluminium alloy can contain four different capital letters such as F, O, W, T. F means that the alloy is in as-fabricated condition, O stands for annealed condition, W stands for solution heat treated condition and T means that alloy has been thermally treated. The thermal treatment can be different, recognized by the digit following the letter T:
T1 - Naturally aged after cooling from an elevated temperature shaping process, such as extruding.
T2 - Cold worked after cooling from an elevated temperature shaping process and then naturally aged.
T3 - Solution heat treated, cold worked and naturally aged. T4 - Solution heat treated and naturally aged.
T5 - Artificially aged after cooling from an elevated temperature shaping process. T6 - Solution heat treated and artificially aged.
T7 - Solution heat treated and stabilized (overaged).
T8 - Solution heat treated, cold worked and artificially aged. T9 - Solution heat treated, artificially aged and cold worked.
T10 - Cold worked after cooling from an elevated temperature shaping process and then artificially aged [18], [19].
Solution treatment at temperatures close to the eutectic temperature is done to dissolve soluble phases, homogenize alloying elements and to spheroidize eutectic Si particles. Quenching consists of a rapid cooling and obtains a supersaturated solid solution of solute atoms and vacancies within the Al matrix. Age hardening at room temperature is called natural ageing and while ageing at elevated is called artificial ageing. Ageing results in precipitation from solid solution [25]. A T5 treatment is usually applied to HPDC components [25].
2.2.3 Influence of cooling rate on microstructure during solidification
Solidification is a phase transformation process at which molten metal transforms into solid state on cooling [10]. Pure metals solidify at a fixed temperature while the alloys solidify over a temperature range [10].
Solidification starts when a nucleus is formed from a group of atoms. The crystal or grain is then growing when additional atoms attach building a crystal lattice.
Crystal is growing in three dimensions into branches called dendrites. Dendrites, in turn, grow forming the main branches and later secondary branches, Figure 6 [28]. Thicker secondary branches absorb thinner ones as they grow by diffusion of from surfaces of higher curvature to surfaces of lower curvatures [29]. The grain boundary is formed at a line where several dendrites meet each other. Finally, all the gaps between branches are filled with crystals and the whole volume is solidified.
a b
Figure 6. Secondary dendrite arm spacing a) Line intercept method [30] and b) three
dimensional representation [31].
The secondary dendrite arm spacing (SDAS), the distance between dendrite secondary arms, is the most useful parameter to evaluate the rate at which the material has solidified. SDAS is measured according to Figure 6a [32].
Line intercept method is widely used to determine the dendrite arm spacing. In this method, parallel dendrite arms in the microstructural image are selected and the total length from the first to the last arm is measured in millimetres. The length is than divided by number of dendrite arms and converted to µm taking into account the magnification of the image, Figure 6a [32].
Final microstructure is highly dependent on the solidification conditions such as chemical modification, inoculation and cooling rate [6]. The slower the cooling rate, the coarser microstructure will solidify. With higher cooling rates, increased undercooling happens and secondary dendrite arm spacing becomes smaller [33] .
SDAS is reported to follow the equation:
ds = a∙tn (Eq. 1)
where ds is the secondary dendrite arm spacing, t is the solidification time, the time at which liquid and solid have coexisted, a and n are the constants for specific material and solidification condition.
The secondary arm spacing in Al-6%Si material has been found to follow the equation: ds = 50 × 10−6(ⅆT
ⅆt) −0.33
(Eq. 2)
where T is the temperature in Kelvin and t is the time in seconds. The faster is the cooling rate, the smaller dendrite arm spacing is obtained. The typical dendrite secondary arm spacing is reported to be of 3 µm, 12 µm and 25 µm at the cooling rates of 4630 Ks-1, 72Ks -1 and 8 Ks-1, respectively [34].
Different defects may happen in die casting. The most common are: surface defects, laminations, gas porosity, blisters, flow porosity, shrinkage porosity, heat sinks, leakers, cracks and tears, inclusions, solder, carbon build up, die erosion, outgassing, edge porosity, bending, warping, stained castings, waves and lakes, drags, deformation from ejector pins, excessive flux [12].
Due to the improper cooling the following casting defects can occur, here described accordingly by the international classification system [35].
B221 Internal shrinkage appears due to lack of melt in hot spots during solidification and volume contraction during cooling. The molten alloys has lower density than the alloys in the solid state and due to that the shrink always present when the alloy changing from molten state to solid state.
Some porosity can be formed within the grain in the dendritic branches or at the grain boundaries in case the amount of molten metal is not enough. The former is called inter-dendritic porosity, the latter is the inter-crystalline porosity.
Higher cooling rate is reported to increase the percent share of porosity in casting microstructure, Figure 7 [36]. The high rate of the cooling promotes the dendrite growth, inside which the porosity may be formed. To eliminate the porosity, the solidification time can be prolonged. However, the reduced mechanical properties are reported to occur [36]. Entrainment porosity is dominant in HPDC formed by air trapped within the die and not fully vented out during the die filling. In case of aluminum alloys, hydrogen may precipitate
from the melt as its solubility is greater in liquid than the solid state creating hydrogen porosity [37].
Figure 7. Effect of casting cooling rate on the severity of microporosity defects in casting
[38].
To prevent shrinkage casting defects one need to design a running gate system with risers that ensures a continuous flow of molten metal. Another activity is to increase local heat dissipation by better cooling. The final step may be to reduce casting temperature to limit the total volume deficit.
C221 Hot tearing appears often close to hot spot, heavy sections or areas with changes in section size. Hot tearing has random occurrence and extent, alloy specific and can be caused by other defects like shrinkage pores formed during cooling.
E231 Early shakeout appears if casting was shaken from the mould before the solidification is completed or if cooling time is not optimized.
Porosity in aluminum alloys is caused by shrinkage or hydrogen porosity and usually by combination of these two mechanisms [37]. Shrinkage, hot tearing and early shakeout defects can be eliminated by more efficient cooling.
2.2.4 Influence of cooling rate on mechanical properties
Final microstructure and, as a result, properties of cast component is highly dependent on the solidification conditions such as chemical modification, inoculation and cooling rate [6]. Cooling rates of 50-500°C/s are reported to be practicable in HPDC of metals [39]. As mentioned above, SDAS decreases with increasing cooling rate. As a result, the microhardness and strength is reported to increase correspondingly for aluminium alloys and for A356 alloy in particular [29], [40], [41], Figure 8.
a b c
Figure 8. a- secondary dendrite arm spacing vs cooling rate for A356 alloy, b-
microhardness vs cooling rate for A356 alloy, c- ultimate tensile strength vs secondary dendrite arm spacing [40].
2.3 Tool steel
Steel types intended for tool making of various kinds are called tool steel. The chemical composition of the tool steel varies a lot, where the steel can be alloyed with different elements of different amount to obtain specific properties. In delivery condition, the tool steel is usually soft-annealed but can also be supplied as martensitic, hardened and tempered to the desired hardness [42], [43].
2.3.1 Types and designations
With regard to the international standardization of tool steel ISO 4957: 2018, tool steels can be grouped as follows:
Unalloyed cold working steels Alloyed cold working steels Alloyed hot working steels Speed steels [44].
In addition to the ISO model, the classification system from AISI (American Iron and Steel Institute) is the most widely used. AISI identifies and classifies the tool steel according to three different characteristics such as alloy composition, application and heat treatment. In addition, each identified area can itself include several types of steel. The classification of these main groups can be seen in Table 3, where each group has its own letter designation [43].
Table 3. The main tool groups and identification symbols according to AISI [43]
Water – hardened tool steel (W)
Shock – resistant tool steel (S)
Oil – hardened cold work tool steel (O)
Air – hardened cold work tool steel (A)
Cold work tool steel with high content of carbon and chromium (D)
Low alloyed tool steel (L)
Low-carbon mould steels (P)
Chromium hot-work tool steels (H)
Tungsten hot-work steels (H)
Molybdenum hot work tool steel (H)
Tungsten high sped steels (T)
Molybdenum high speed steel (M)
Extra hard high speed steels (M)
Hot work steels are primarily intended for work in a warm environment where high stresses occur in the tool. Hot work tool steels are most common in dies and inserts for HPDC.
These application areas make specific demands on the hot work steel, where the following properties are important [45]:
Toughness Creep strength Abrasion resistance
Ductility Hot strength Temper resistance
Hardness Heat conductivity Expansion coefficient
These properties are responsible for the tool's service life and quality. In order to meet the property requirements, the hot working steels are often alloyed with tungsten, chromium, vanadium and molybdenum and relatively low carbon containing where the average range of the carbon content is 0.3-0.5% by weight. These alloying elements provide strength at high temperatures and also give the steel high hardenability. The high temperature strength is obtained in the material due to the formation of secondary tempering carbides in the steel microstructure during high temperature tempering. Depending on the alloying elements, carbides can be formed of the following types, where M symbolizes carbide forming elements or their combination [43], [46]:
Vanadium: Matrix → M3C → MC
Molybdenum/Tungsten: Matrix → M3C → M2C → M6C
Chromium: Matrix → M3C → M7C3
The AISI classification system, see Table 4, divides the hot working steel into three subgroups, denoted H, according to the dominant alloy content [43], [47]:
Chromium alloyed hot working steels are widely used for die casting of aluminium and magnesium. The steel has good tempering resistance.
Tungsten alloyed hot working steels can be used for tools where high wear resistance and high thermal strength are required, for example, forming of low alloy steel, bronze, nickel and other alloys.
Molybdenum alloyed hot working steels are cheaper than tungsten alloy hot work steels and are usually applied to applications where higher tool performance and hot strength are required.
Maraging steels is a family of low carbon, high-alloy steels typically containing 12%–18% nickel, 3%–5% molybdenum, 0%–12% cobalt, 0.2%–1.6% titanium, and 0.1%–0.3% aluminium (one cobalt-free grade also contains 5% chromium). The steels are characterized by high strength up to approx. 450ºC and toughness, exceptional corrosion resistance, simple heat treatment and good weldability. The main limitations of the maraging steels are their low heat conductivity, sensitivity to hydrogen embrittlement and rapid loss in strength, known as overaging, at long term temperature exposure above the aging temperature [48].
Uddeholm Maraging, developed both for making it suitable for AM and optimized for high performance in HPDC applications, is used in the current work [49]. More data about the new tool steel in the powder form will be presented in the experimental part of this work. 2.3.2 Tool steel manufacturing
Tool steel can be manufactured in several ways. Uddeholms AB producing high alloyed steel types by conventional and powder metallurgical process [5].
The conventional manufacturing process is based on raw materials which can be both internal and external scrap material. The raw material must be carefully selected to avoid contamination levels in the steel from the beginning. The steel manufacturing process consists of sex general steps where electroslag remelting (ESR) or homogenization is used in special cases:
a – Electric arc furnace b – Deslaging
c – Ladle furnace d – Vacuum station e – Uphill casting
f and g – ESR or homogenizing h and i – Forging or rolling, Figure 9 Heat treatment
In the electric arc furnace the scrap material is preheated to avoid water in the melt. Then the preheated scrap material is loaded into the arc furnace for melting. The melting takes place by transferring large amounts of electrical energy to the steel mixture by means of arcs between three graphite electrodes placed in the oven lid. As the steel is melted, slag formers are added to refine the melt. Carbon and phosphorus levels can be regulated in the arc furnace if necessary and the refining process takes about two hours, Figure 9a. After refining when the temperature is ~ 1670 ° C, the melt is dropped into a ladle where a secondary metallurgical treatment begins with the separation of the oxide-rich slag from the melt, Figure 9b.
To maintain the temperature of the melt during subsequent processes, it is heated in the same way as in the arc furnace with three graphite electrodes placed on the ladle lid, see Figure 9c. After the slag separation, aluminum wire is added to deoxidize the remaining oxygen.
After deoxidation, alloying as well as the addition of new synthetic slag occurs during inductive stirring. The stirring and heating of the melt to 1580 - 1650 ° C is done to maintain chemical reaction rates, which gives a homogeneity of the alloying elements in
the melt. During the process step, the composition of the melt is analysed by sampling. When the composition is obtained, the melt is passed on to a vacuum station, Figure 9d.
a - Electric arc furnace b - Deslaging c - Ladle furnace
d - Vacuum station e - Uphill casting f - Electro Slag Refining
g - Homogenization h- Hot rolling i – Hot forging
Figure 9. Illustration of the main stations in conventional process, courtesy of
Uddeholms AB.
In vacuum station, a vacuum lid is placed on the ladle to create partial vacuum, Figure 9d. During the vacuum degassing, hydrogen, nitrogen and oxygen are separated by continuous stirring with argon and induction. The strong highly alkaline slag together with continuous stirring of the melt results in favourable conditions for sulphur purification as well as the growth and separation of slag inclusions to the top slag. After the vacuum treatment the ladle is sent for casting.
Uphill casting carries a small risk of turbulence and re-oxidation of the melt, see Figure 9e. During casting the melting temperature, viscosity and flow rate of the melt are optimized for the respective steel to obtain a good filling of the molds. During cooling, the ingots solidify inhomogeneously due to segregation occurring in the melt. If you want to eliminate the segregations, the ingot will be sent for homogenization or ESR.
The goal of Electro Slag Refining (ESR), Figure 9f, is to improve the homogeneity and structure of the ingots while separating non-metallic inclusions. The ESR also provides uniform mechanical properties in the longitudinal and transverse direction after forging or hot rolling. In order to achieve this, the ingot is welded to the electrode to obtain a good electrical conductivity to the ingot. The ingot is lowered into an ESR oven with slag. Then
a voltage is applied between the slag and the electrode, causing the slag to heat up and melt the ingot. Molten steel gradually drops through the slag, solidify and form a solid body while non-metallic inclusions refine and become stuck in the slag.
Homogenization is the heat treatment of ingots is another. To homogenize the ingot, it is heated in an oven, Figure 9g, to 1300 ° C and maintained at this temperature for 12 hours, resulting in a levelling of alloying elements in the steel.
Forging and rolling are processes after homogenization or ESR, which leads to the material getting its final dimensions. During the process, the material is preheated and then subjected to plastic deformation. Since the growth of the grain of the steel can occur during homogenization, the strong deformations result in a continuous recrystallization of the material. Forging, Figure 9i, and rolling, Figure 9h, are processes that also increase the homogeneity of the steel and also provide an even distribution of any inclusions in the steel.
A soft and homogeneous structure in the material is preferred for the delivery condition which is obtained with the soft annealing. Soft annealing is a heat treatment process that makes the steel softer. The annealing of the steel is usually done to facilitate machining and to smooth the microstructure. During soft annealing, an even distribution of carbides and alloying elements as well as reduction of residual stresses in the material is obtained. The material is heated above the Ac1 temperature and cooled slowly to obtain a ferritic structure with spheroidized carbides [42], [45]. Ac1 is the temperature at which ferrite begins to convert to austenite upon heating of the steel and at Ac3 all ferrite has been converted to austenite. For hot working steels, the Ac1 temperature is at 800-850ºC and the Ac3 temperature is at 900-950ºC, depending on the alloy composition.
2.3.3 Tool steel powder manufacturing
The powder metallurgical process (PM) differs from the conventional steel manufacturing process. The PM process consists of several steps, where processes in the arc furnace and in the ladle furnace are the same as in conventional steel making. But instead of sending the steel melt to the casting, it is sent to a PM plant for atomisation.
At the gas atomization station nitrogen gas is blown on a jet of the melt, from which small and round shaped powdered steel grains are formed. On the way down to the bottom of the container, the powder grains solidify and accumulate in a capsule, Figure 10. Once the powder has collected, it is sieved to the sizes suitable for AM and packed [5].
For the case of tool steel for additive manufacturing, atomization is performed in a dedicated equipment. A close coupled nozzle is used for atomization. This provides possibility to achieve high quality powder with size and morphology needed for AM. The melting is done in this case through vacuum induction. After atomization, the powder is sieved to the desired size and is handled carefully and packed with bottles.
Figure 10. Powder metallurgical process, courtesy of Uddeholms AB [5].
2.4 Tool and die manufacturing
The following steps are usually done for manufacturing a die: Rough machining
Hardening and Tempering Fine machining
Surface treatment [5]
Machining is a process of chip removal from the material to obtain a certain shape and size. The material is machined in the soft-annealed delivery condition.
After machining, the die needs to be heat treated to obtain the right properties such as hardness, het resistance, toughness, ductility, etc. The heat treatment process includes hardening and tempering of a die, Figure 11a.
During hardening the material is heated above the Ac3 temperature and cooled to convert austenite to martensite. During austenitization, carbon and other alloying elements are dissolved in the austenite. The carbon and alloying elements are unable to diffuse from the matrix and form carbides at rapid cooling. Instead, they are locked into a space-centered tetragonal structure called martensite. The phase conversion occurs instantaneously without diffusion [42], [45].
The cooling of tool steel must be done quickly enough to avoid the formation of pearlite and bainite. The coolant can be water, salt, oil and air, where the cooling rate is fastest in water and slowest in air. The cooling rate of the tool steel depends on the alloy composition, since high alloy steels can be cooled more slowly than lower alloy steels to avoid perlite and bainite formation. The hot working steel is usually cooled in air.
The tempering of the material is done to increase the toughness and ductility of the martensite which, after hardening, is brittle and hard. The hardness changes that take place in the hot working steel of H13 type during a two hour tempering are shown in Figure 11b. During annealing, the alloying elements and carbon diffuse from the martensite to form carbides also called secondary carbides [45].
a b
Figure 11. a- hardening and tempering graph of tool steel [43], b- tempering graph for
H13 type of hot work tool steel [5] courtesy of Uddeholms AB.
A certain change in shape of a die usually occur after the hardening and tempering procedure. Fine machining or grinding is usually performed to adjust the die to correct dimensions.
Surface treatment can be performed on a die or certain parts, such as ejectors, cores, core pins, etc. to increase the erosion and soldering resistance. The common processes are oxidizing, nitriding and PVD coating.
Cooling system of the tool consists of several complex loops of cooling channels going through the tool and die inserts. It is the vital part in HPDC technology responsible for the total cooling behaviour and, as a result, for solidification, microstructure and properties of the component as well as the die life time [50].
Cooling channels are usually produced by drilling straight holes from both sides of the die [51] Cooling channels are reported to be not effective when positioned away from the die cavity. They need to be positioned close to the surface of the die to extract the heat and cool the component effectively [52].
2.4.1 Additive manufacturing
According to the ASTM standard F2792-10, additive manufacturing is the process of building the parts layer by layer. Metal additive manufacturing using powder is represented by the two main technologies – blown powder technology and powder bed technology while the latter is divided into laser beam melting, electron beam melting and inkjet printing [53], Figure 12.
Figure 12. Metal additive manufacturing processes [3].
Two main powder particle distributions are used for metal additive manufacturing. Particles in the range between 10-20 and 50 microns are used for laser beam melting, Figure 13, while powders in the range between 50 and 100-150 microns are used for electron beam melting and blown powder technology [3].
Figure 13. Example of powder size distribution curve for laser beam melting technology
[3].
In the present thesis, Laser Powder Bed Fusion (PBF-LB/M) technology is used to produce HPDC inserts. Thin layer of metal powder is distributed using re-coater and selectively melted by a high-power laser beam moving in X-Y directions, Figure 14. Once the required 2D slice of the component is welded, the building platform moves down in vertical Z axis, new powder layer is distributed and the process is repeated until the component is complete. The process takes place in a chamber with protective gas such as argon or nitrogen to ensure minimal contamination of component with oxygen [3].
a b
Figure 14. a- laser beam melting process, b- the powder bed manufacturing cycle, courtesy
of Fraunhofer [3].
Additive manufacturing is growing rapidly overcoming different challenges. It is still new technology which require research and development work to be able to take its full advantage [54]. Additive manufacturing has achieved high maturity level in medical and tooling industries as well as becoming a disruptive technology in aerospace and automotive, Figure 15 [3].
Figure 15. Manufacturing readiness level in various industry sectors, courtesy of Roland
Berger [3].
Initially, design freedom unachievable by other methods has been the driving force for additive applications. The component can be printed directly from the CAD model reducing the design and prototyping lead time. Moreover, assembly can be simplified as more components are manufactured in one. Stock of spare parts and cost related to that can be reduced as the component can be manufactured on demand [3].
Additive manufacturing has brought several advantages, additional to the mentioned above, to the tooling applications, especially to plastic moulding. Printing of conformal channels could result in faster and more even cooling leading to higher productivity and improved component quality [6], [55], [56]. It is expected that similar advantages of additive
manufacturing can be found for hot working application such as HPDC. However, it is difficult to find literature on the subject of conformal cooling and its effect on HPDC process and the properties of the cast materials. The aim of this study was to investigate and quantify the effect of conformal cooling on microstructure and mechanical properties of high pressure die cast aluminum alloy.
3. Experimental
3.1 Aluminium alloy for casting of component
The material used in the experimental casting had the following designations:
1. In Europe it is known as EN AC-42000 alloy 2. In USA as A 356.x
3. In UNS database it is known as A03560
4. In the specification database it known as CEN EN 1706(98). The standard composition for this alloy can be seen in Table 4 [57].
Table 4. Standard aluminium alloy compositions with following content in wt% [57] Al Cu Fe Mg Mn Pb Si Ti Zn
Balance ≤0.20 ≤0.55 0.20-0.65 ≤0.35 ≤0.15 6.5-7.5 0.05-0.25 ≤0.15 Chemical composition of ingot and melt was measured by Optical Emission
Spectroscopy using a Spectromax LMX06 according to the method Al-20-M. Before the measurement equipment was calibrated according to the spectrochemical standard B319.1 AF – C005.
3.2 Tool and tool manufacturing
Two different inserts for HPDC were manufactured with PBF-LB/M to perform experimental casting of components that had a shape of a flat tensile specimen.
Two inserts (fixed and ejector die inserts) with conformal cooling channel (referred to as conformal).
Two inserts (fixed and ejector die inserts) without cooling channels (referred to as conventional).
Tool steel powder composition is shown in Table 5. This material was specially developed for hot work applications and used for printing of the inserts.
Chemical composition analysis was measured by Optical Emission Spectroscopy using a Spectromaxx LMX06.
Table 5. Chemical composition of steel powder used for insert printing in wt%
Material C Si Mn Cr Mo Co Cu Ni O
Maraging 0.03 0.35 0.40 5.0 8.0 12.0 2.0 2.0 ˂250 ppm The powder was analysed by the equipment which is shown in the Figure 16 at Uddeholms R&D powder lab. The results are shown in Table 6.
a b c
Figure 16. a- Camsizer XT Retsch X-dry for size distribution measurement, b-
Pycnometer Ultrapyc 1200e for aspect ratio and c- AutoTap for density measurement.
Table 6. Physical powder characteristics of the tool steel powder
Material D10 [µm]
D50 [µm]
D90
[µm] Sphericity Aspect Ratio
True density [g/cm3] Tap density [g/cm3] Uddeholm Maraging 23 34 47 0.93 0.90 8.0 5
An EOS M290 PBF-LB/M system was used to produce inserts for this thesis work. Below is a representative picture of the equipment, Figure 17a, along with the capabilities, se figure 17b.
a
b
Figure 17. a- EOS M290 printer used to manufacture the die inserts, b- technical
The parameters for printing the tool steel powder were developed and optimized at Uddeholms AB and for the current thesis work, the following parameter set was used:
Layer Height = 40 µm Laser Power = 295,6 W Scan speed = 790 mm/s Hatch Distance = 0,126 mm Stripe width = 9,75 mm Overlap = 0,12 mm
The complete die assembly that was utilized for HPDC experiments in the thesis work is visualized in the image below, Figure 18. The weight of the total die assembly is
approximately 1 ton.
a b
Figure 18. a - complete die assembly [58] developed by Olof Granath at Jönköping
University and b – die mounted in the HPDC machine.
The fixed and ejector die inserts with conformal cooling are shown in Figure 19. Design of the inserts was done during the previous work performed at Jönköping University, Figure 20 [58]. The die inserts without cooling channel had the similar design as in Figure 19 but without internal conformal cooling channel.
a b
Figure 19. a- fixed die insert, b- ejector die insert. Drawings from Solid Works show the
Figure 20. Drawing of the ejector insert with conformal cooling channel courtesy of JU.
The inserts in the as-printed condition are shown in Figure 21. To relieve stresses introduced by the printing process and to get the desired microstructure and properties of the inserts, the heat treatment was performed in a vacuum furnace after the printing, Figure 22a. The plate with the printed inserts were heated up to the solution treatment/stress reliving temperature of 950°C, hold at the temperature for 1 h and the cooled down to the room temperature, cooling time between 800°C and 500°C was 100 s. The heat treatment graph, Figure 22b, is extracted from the Uddeholms R&D Vacuum furnace control system.
a b
Figure 22. a- vaccum furnace used for heat treatment of die inserts, b- heat treatment
graph showing heat treatment procedure – time, temperature and pressures.
Machining was performed at Intermekano Tools AB to obtain the final design, Figure 23. The inserts were then integrated into the total die set to perform HPDC experiments.
Figure 23. Die inserts after the heat treatment and after final machining.
3.3 High Pressure Die Casting
The HPDC experiments were performed at Jönköping University Casting facility which is equipped with a vertical configuration machine, Figure 24. The first set of castings were performed on the conventional inserts (without any cooling channels) and totally 13
Before machining
After machining, but before casting
Conventional inserts
samples were produced of which 8 samples were used for further evaluation and 5 samples was not cast completely. The next set of trials were performed the week later with the inserts with conformal cooling channels and 16 castings produced where 8 samples were used for further investigation and 8 samples were not properly cast.
Figure 27. Vertical high pressure die casting machine.
The aluminium alloy in the form of ingots was melted in an induction furnace at 650°C, Figure 25.
Figure 25. Induction furnace used for melting of the aluminium alloy.
The temperature of the melt was verified with a K-type thermocouple. After a proper temperature was reached a dosing cup was used to feed the molten aluminium into the shot sleeve, Figure 26a. The casting was performed at a pressure 175 bar and shot velocity was 0.1 m/s, Figure 26b. Injection and solidification time was about 18 s, total cycle time was about 45 s with manual extraction of the sample. Typical temperature of the pressurized oil cooling was 178°C. During the casting experiments, temperature measurements were performed on the inserts surface using a thermal imaging camera and the temperature and flow of the oil (coolant) was also recorded, Figure 27.
a b
Figure 26. a- shot sleeve for alloy pouring in the casting machine, b- control parameters
of the casting machine.
a b
Figure 27. a– oil pump and heater, b- oil temperature and flow meter.
Figure 28. Thermal camera used for temperature measurements on the die insert
Thermal imaging camera FLIR T640, Figure 28, was used to read off temperature on the die insert during casting using the following parameters:
Temperature range -40 to 2000ºC, IR-Resolution 640 x 480 pixel Accuracy ± 2 ºC or 2% Heat sensitivity 35 mK
After casting the cast samples, Figure 29, consisted of biscuit, tensile specimen and overflow, were cooled down to room temperature by immersion to the water bath. Later, biscuit and overflow were cut away from the tensile specimens.
Figure 29. Cast sample consisting of tensile specimen, biscuit and overflow.
3.4 Heat treatment
Although, not the main objective of the thesis, it was decided that heat treatment response of the samples after casting would be studied to help further research on this topic. Some pieces of tensile specimens cast with and without conformal cooling were heat treated in the Nabertherm TR-120 furnace, Figure 30, according to T5 heat treatment cycle. The pieces were heated to 175°C, hold for 20, 30, 40, 70, 90, 180, 240, 270 and 320 min and cooled to room temperature in air.
Figure 30. Nabertherm furnace used for aging of aluminium alloy pieces.
Hardness of the specimens before and after heat treatment was measured according to the standard EN ISO 6506-4:2005 (E) in Brinell HBW5 with ball indenter of 5 mm. Five measurements were made on each specimen. The average value and standard deviation of the series of three measurements were calculated when the highest and lowest measured values were removed from the series.
Biscuit Tensile specimen
3.5 Mechanical properties
Cast flat tensile specimens with a thickness of 3 mm, a width of 30 mm and a reduced section length of 50 mm were tested according to standard ASTM B557M-E8M. Tensile testing was performed using a Zwick/Roell Z100 testing machine with a 100 kN load cell at room temperature, using a test speed of 1/s. Strain measurements were made using a Zwick/Roell laser extensometer with an initial distance of 26 mm between the measuring points, Figures 31-32. The specimens were mounted with in-gate side facing upwards.
Figure 31. Schematic illustration of the experimental setup of the tensile specimen [59].
a b
Figure 32. a- Zwick/Roell Z100 testing machine, b- laser extensometer.
3.6 Microstructure investigation
The aluminium specimens cast with and without conformal cooling both in as-cast and heat treated condition were cut and prepared for metallographical investigation. Specimens were mounted in resin with Struers resin machine, Figure 33. The mounting temperature was 180°C, mounting time is 6 min. The specimens were exposed to that temperature for around 2 min. The properties of the as-cast specimen could have been minor effected by the mounting procedure and the effect is neglected in the study.
Figure 33. Struers resin mounting machine.
Rough grinding of the specimens were done with abrasive stone Al2O3 and applied force of 50 Newton at ATM Rubin 520, Figure 34a. Then, semi-automatic polishing at ATM machine, Figure 34b, was performed in three steps. First polishing step was grinding with 9 µm diamond suspension at applied force of 50 N. Second polishing step was 3 µm diamond suspension at applied force of 50 N. Final polishing step involved suspension with 1 µm particles and applied force of 20N. To succeed with specimen polishing, it is important to wash/rinse properly away the residues of abrasive particles from the previous polishing step avoiding to transfer them into the next step. After polishing with 1 µm diamond suspension, the specimens were carefully wiped with cotton dipped in liquid soup and ethanol in order to get rid of residues of lubricant and abrasive diamonds. Complete procedure of grinding and polishing step is shown in Table 7.
a b
Figure 34. a- ATM Rubin 520 used for rough grinding of the specimens, b-
semi-automatic ATM polishing machine.
All the specimens were etched with 5 % NaOH etching reagent and the holding time were 15 seconds.
Secondary dendrite arm spacing (SDAS) was measured using light optical microscopy (LOM) in a Zeiss Axiophot microscope, Figure 35, at 200x magnification using line intercept method [54].
Table 7. Grinding and polishing steps of aluminium specimens during the
metallographic preparation.
Step Type of disc Suspension Time Applied
force (N) Rinsing Grinding
stone Al2O3, 120 grid Water 1 min 50 Water & Ethanol Polishing
cloth MD – Allegro DiaP 9µm 1 min 50 Water & Ethanol Polishing
cloth
MD – Plus DiaP 3µm 1 min 50 Water &
Ethanol Polishing
cloth MD – Nap DiaP 1 µm 1 min 20 Water & Ethanol
Figure 35. Light optical microscope Zeiss Axiophot used for SDAS measurements.
3.7 Tool investigation
Surface roughness of the fixed die insert was recorded before and after casting trials. The surface roughness of die inserts can effect the heat transfer between the insert and the specimen and therefore were measured before the casting trials for the conformal and conventional inserts for comparison. Surface roughness of the die inserts after castings were measured to study the tool wear. The hypothesis was that the de insert with higher surface temperature may wear more than the die insert with lower surface temperature. Surface roughness was recorded with a portable profilometer Taylor Hobson Ametek, Figure 36. The surface roughness according to ISO 4287:1997 was measured at 9 positions on the inserts, Figure 37, with cut-off length 2.5 µm and profile curve length 16 mm. The average value was calculated for positions 1-3, 4-6 and 7-9.
a b
Figure 36. Portable profilometer Taylor Hobson Ametek used for surface roughness
measurements on die inserts.
Figure 37. Positions on the die inserts where the surface roughness were measured
before and after casting trials. Positions are shown on the specimen for easier visualization.
4. Results and discussion
4.1 Chemical composition
Chemical composition measured on the ingots before casting (Ingot g 1, 2 and 3) and after casting (Melt g 1, 2 and 3) are shown in Table 8 below.
The results show that the ingot and the castings were according to the standard for the alloy [57]. There were minor differences observed, for example in the Mg and Si content. The Mg content was higher and Si was lower after melting than in ingots. The observed results are not expected since a lower Mg content is usually obtained due to Mg loss during melting procedure, and on the other hand increases percentage of Si content in the alloy. The origin for this differences could be due to limited amount of analysis and sampling for the measurements or due to “contamination” during melting.
Table 8. Chemical composition of ingots and melts in wt%
Source Cu Fe Mg Mn Pb Si Ti Zn [57] ≤0.20 ≤0.55 0.20-0.65 ≤0.35 ≤0.15 6.5-7.5 0.05-0.25 ≤0.15 Ingot g 1 0.052 0.31 0.34 0.16 0.008 6.59 0.033 0.084 Ingot g 2 0.053 0.33 0.34 0.17 0.008 6.70 0.031 0.087 Ingot g 3 0.055 0.34 0.35 0.17 0.009 6.86 0.030 0.089 Melt g 1 0.052 0.34 0.53 0.20 0.010 6.66 0.034 0.084 Melt g 2 0.050 0.32 0.51 0.19 0.009 6.27 0.036 0.080 Melt g 3 0.052 0.34 0.55 0.20 0.012 6.75 0.036 0.11
4.2 Cooling behaviour
From the thermal imaging data, Figure 38, one section of the insert was chosen and temperature data was extracted from a linear profile/section directly after opening the die and removing the specimen. The values were then compared between different tool inserts as a function of time after casting. The results are plotted in Figure 39.
Figure 38. Typical thermal image of the casting obtained by thermal camera.
According to this data, conformal insert had lower temperature after casting compared to conventional in one of two measurements, Figure 39a. Additionally, conformal insert had faster cooling and reached its final temperature faster compared to conventional in both
measurements, Figure 39b. Power function regression curves declined faster for the conformal insert than for the conventional. The power was -0.629/-0.852 and -0.511/-0.593 for the conformal and conventional inserts, respectively, Figure 39b. All observations suggest more efficient cooling from the conformal cooling channels.
Another set of data was obtained from the sequence of thermal images after every shot of casting directly after the die opening and the specimen removal. The Figures 40a-c show that the maximum, minimum and average temperature across the section was lower for conventional inserts than conformal inserts in the most shots.
The conformal inserts were expected to show lower temperatures than conventional due to cooling channels. Figure 39 and Figure 40 show opposite trends and it is difficult to provide a good explanation for these observations. Although, one reason could be that Figure 40 shows data at one point in time while Figure 39 shows over a span of time. However, difference between the maximum and minimum temperature normalised to the average temperature on the insert surface did not show any significant difference between the inserts, Figure 40d.
a
b
Figure 39. Temperature vs time during HPDC cycle for conventional and conformal cooling inserts a – for the time interval from 2 to 14 seconds, b – for the time interval from 2 to 2.5 seconds with power function regression curves.
100 120 140 160 180 200 220 2 3 4 5 6 7 8 9 10 11 12 13 14 Tempe ratur e, de grees C Time, s
Conventional Conformal Conventional Conformal
y = 264,52x-0,511 y = 338,76x-0,852 y = 279,94x-0,593 y = 226,6x-0,629 120 140 160 180 200 2 2,5 Tempe ratur e, de grees C Time, s
a b
c d
Figure 40. Temperature vs shot number during HPDC for insert with conformal cooling
and without conformal cooling (conventional), a - maximum temperature, b – minimum temperature, c - average temperature, d - difference between maximum and minimum temperature normalised to the average temperature on the insert surface.
Another way of evaluation of the cooling behaviour was to analyse the temperature changes of the cooling oil during the casting cycle. Based on the thermal equilibrium during the HPDC, the following derivations could be made.
The heat/energy change of the process consists of the heat change of the die (D) related to the part solidification and heat change in the rest of the system/external loop (L) related to outer die surface, hoses and so on, Equation (3).
ⅆ𝑄 ⅆ𝑡 = ( ⅆ𝑄 ⅆ𝑡)𝐷+ ( ⅆ𝑄 ⅆ𝑡)𝐿 (Eq. 3)
The heat change can be calculated according to the Equation (4) [10]: (ⅆ𝑄
ⅆ𝑡) = 𝑚 𝐶𝑝 ⅆ𝑇
ⅆ𝑡 (Eq. 4)
where m is a mass, 𝐶𝑝 is a specific heat capacity andⅆ𝑇
ⅆ𝑡 is the temperature change. The change in the die is:
(ⅆ𝑄
ⅆ𝑡)𝐷 = 𝑚𝑀𝐶𝑝 𝑀 ⅆ𝑇𝑀
ⅆ𝑡 (Eq. 5)
The change in the loop: (ⅆ𝑄 ⅆ𝑡)𝐿 = 𝑚𝑂𝐶𝑝 𝑂 ⅆ𝑇𝑂 ⅆ𝑡 ≈ 𝑣𝜌𝑂𝐶𝑝 𝑂(𝑇 𝑂𝑜𝑢𝑡−𝑇𝑂𝑖𝑛) [10] (Eq. 6)
where 𝑚𝑂 is the mass of oil, 𝐶𝑝𝑂is the oil specific heat capacity, 𝑣 is the oil volume, 𝜌𝑂is the oil density, 𝑇𝑂𝑜𝑢𝑡−𝑇𝑂𝑖𝑛 is the temperature change in oil coming into the system (in) and going out from the system (out).
0 100 200 300 400 0 1 2 3 4 5 6 7 8 9 101112131415 Conventional Conformal 0 100 200 300 400 0 1 2 3 4 5 6 7 8 9 101112131415 Conventional Conformal 0 100 200 300 400 0 1 2 3 4 5 6 7 8 9 101112131415 Conventional Conformal 0 0,3 0,6 0,9 1,2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Conventional Conformal
