Civilingenjörsprogrammet i teknisk fysik med materialvetenskap
Upps al a univ ersit ets l ogot yp
UPTEC Q 21005
Examensarbete 30 hp Juni 2021
Evaluation of new powder
grade for furnace control pieces in sintering process
Selma Halilovic
Civilingenj örspr ogrammet i t ek nisk fysik m ed materi alvet e nsk ap
Teknisk-naturvetenskapliga fakulteten Uppsala universitet, Utgivningsort Uppsala/Visby
Handledare: Sead Sabic Ämnesgranskare: Jannica Heinrichs Examinator: Åsa Kassman Rudolphi
Upps al a univ ersit ets l ogot yp
Evaluation of new powder grade for furnace control pieces in sintering process
Selma Halilovic
Abstract
To be able to supervise the quality of a sintering process, furnace control pieces are therefore used. The current furnace control piece is not sensitive enough small variations during insert production. The goal of the project was to find and evaluate a new suitable cemented carbide grade, which better captures temperature variations during sintering process, likewise, evaluate the alternative placing in the production furnace and methods to supervise and follow the status of the sintering process. The cemented carbide grade 592, which is a DQ-grade, captured larger temperature variations during the sintering compared to the current furnace control piece. The process charge for 592 that captured the largest variations also had the highest charge weight, which indicates that the new grade is also sensitive to the charge weight. The purpose of the project was fulfilled when a more suitable cemented carbide grade, 592, was evaluated for both sintering temperatures 1410 ℃ and 1450 ℃.
Tek nisk-nat urvetensk apliga f ak ulteten, Upps ala universit et . Utgiv nings ort U pps al a/Vis by . H andledare: Sead Sabic , Äm nesgranskar e: Jannica Hei nrichs , Exami nator: Ås a Kassman R udol phi
Utvärdering av möjlig hårdmetallsort för ugnskontroll under en sintringsprocess
Detta examensarbete har utförts på Sandvik Coromant i Gimo under våren 2021. Sandvik Coromant
grundades år 1942 med syftet att utveckla produkter inom hårdmetall och har sedan dess varit världsledande leverantör av verktygsskär. Dessa verktygsskär tillverkas genom pulvermetallurgi, där metallpulver med olika egenskaper blandas ihop. Metallpulvret pressas till önskad geometri i form av en så kallad grönkropp, som sedan värmebehandlas genom sintring. Sintring ger ett fast material som kombinerar egenskaperna seghet och hårdhet hos skäret. För att sedan säkerställa att sintringsprocessen har gått rätt till används en så kallad ugnskontroll vid varje sintringsprocess, för att fånga upp avvikelser i produktion. När
sintringsprocessen är klar så slipas, eggavrundas och beläggs skären innan de levereras till kund.
För att säkerställa kvaliteten mäts de magnetiska egenskaperna på den sintrade ugnskontrollen, vilka är koercivitet, Hc, och viktspecifik mättnadsmagnitisering, CoM.
Syftet med en ugnskontroll är att den ska fånga upp större variationer (stora temperaturförändringar, produktionsavbrott etc.) i processen. Med nya krav på produkter, till exempel, snävare toleranser på sorter, behövs en mer känslig ugnskontroll. Syftet med detta arbete är att ta fram en ny ugnskontroll.
Detta projekt delades upp i tre delar. Den första delen av projektet bestod av att utvärdera tio olika sorter hårdmetallskär, inklusive den sort som används i dagsläget som ugnskontroll, i labugnen på pulver- och ämneslaboratoriet. Elva nya sintringscykler programmerades i labugnen, inom temperaturintervallet 1400 – 1500 ℃. Dessa var baserade på en så kallad DQ-process, vilket är en sintringsprocess för en viss sort av pulversammansättning av hårdmetallskär, liknande de sorter som har använts här. Därefter utvärderades de tio olika sorterna, genom att Hc och CoM uppmättes efter varje sintringsprocess. Syftet med denna del var att komma fram till vilken av dessa sorter som har starkast linjäritet mellan Hc och CoM mot temperatur ökningen i labugnen. Resultaten visade att det var fyra sorter som var mest linjära mot temperaturökningen.
Därefter mättes mikrostrukturens gradienttillväxt för valda sorter för att se om det finns fler parametrar som är temperaturberoende och kan indikera om något i substratet är utanför gränserna. Resultaten visade att gradienttillväxten ökar med högre sintringstemperaturer, vilket överensstämmer bra med teorin.
Vid byte av ugnskontroll, vilket sker var sjätte månad för att undvika åldringseffekter av grönkropparna, tas gränsvärden för olika kvalitetsparametrarna fram för de olika sintringsprocesserna och temperaturerna.
Dessa gränsvärden skall indikera när en sintringsprocess är godkänd. I del två av projektet jämfördes de sintrade produktionsresultaten mot de uppmätta resultaten från labugnen i del ett av projektet, inom temperatur intervallet 1400 – 1500 ℃. Resultaten visade att de olika hårdmetallsorterna från
produktionsugnen upplever stor variation i temperatur, trots att de sintrades i samma produktionsugn med samma sintringstemperatur. Resultaten visade att en produktionsugnscykel är väldigt mycket beroende av flera parametrar (laddningsvikt, storlek på skär, omgivande atmosfär etc.) än en sintringscykel i labbugnen där det endast är temperaturen som varierar.
Med de resultaten inleddes del tre av projektet som bestod av att testa utvalda reperesentanter i produktionsugnar. Totalt genomfördes åtta sintringscykler för sintringstemperaturen 1410 ℃ och tio sintringscykler för 1450 ℃, för de skären som visade högst linjäritet från lab-ugnen. Sintringsprocessen som tillämpades var återigen DQ. Skären placerades på topp- och botten-positioner i ugnen, för att se spridningen hos kvalitetsparametrarna. Därefter utvärderades vilken position som spred mest, och resultaten visade att testbitarna i bottenpositionerna spred lika mycket som alla testade positioner i ugnen. Dessa resultat visade även att sorten 592 hade störst känslighet jämfört med de andra kandidaterna, varför den valdes för vidare utvärdering. Sorten 592 är ett gradientskär, vilket är fördelaktigt när man använder en sintringsprocess då även gradienten i sig kan ge indikatioer kring den upplevda temperaturen i substratet. I de fall Hc och CoM ligger utanför de tillåtna gränsvärdena kan även gradientmätning utföras för att bekräfta resultaten, eftersom den är temperaturberoende. Detta är inte möjligt för dagens ugnskontroll som inte är en gradientsort. Med andra ord var syftet med detta arbete uppfyllt, då sorten 592 var mer känslig mot små variationer under en sintringsprocess jämfört med dagens ugnskontroll.
Table of contents
1 Introduction ... 1
1.1 Background ... 1
1.2 Purpose ... 1
1.3 Limitations ... 1
2 Theory ... 2
2.1 Introduction of cemented carbide... 2
2.2 Microstructure of cemented carbide... 2
2.3 Introduction to the powder grades ... 2
2.4 The manufacturing process ... 3
2.4.1 Cemented carbide production ... 3
2.4.2 Sintering process ... 4
2.4.3 Quality control of sintered cemented carbide ... 4
2.4.4 Sintering temperatures ... 6
2.5 Microstructure gradients ... 6
3 Method and equipment... 8
3.1 Equipment ... 8
3.1.1 Sintering furnace ... 8
3.1.2 Magnetic measurement ... 8
3.1.3 Characterization equipment ... 8
3.2 Method ... 8
3.2.1 Evaluation in laboratory furnace ... 9
3.2.2 Evaluation in production furnace ... 10
4 Results and discussion ... 13
4.1 From the laboratory furnace, part I ... 13
4.1.1 Evaluation of cemented carbide blanks ... 13
4.1.2 The empiric limits for the FCP from production... 15
4.1.3 The microstructural gradient for standard DQ-blanks ... 19
4.2 Laboratory furnace temperature results translated in production furnace results, part II ... 21
4.3 The experimental results from production furnace for DQ1410 and DQ1450, part III ... 22
4.3.1 Test of different candidates in the production furnaces ... 23
4.3.2 Investigation of top and bottom positions for DQ1410 & DQ1450 ... 27
5 Conclusion ... 29
6 Further investigations ... 30
7 Reference ... 31
8 Appendix ... 33
1
1 Introduction
1.1 Background
In order to evaluate the status and condition of the sintering furnace, furnace control pieces (FCP) are used.
The FCP’s are described as pressed inserts of cemented carbide, from the same powder batch and geometry.
Deviations in magnetic properties of the FCP indicate variations with temperature, insulation, gas flow, condition of the furnace etc. It also shows differences between the sintering furnaces. The powder grade that is used today at Sandvik Coromant, is a stable grade and it is not sensitive enough to small variations during production.
The current FCP have been used in the production since 1970. At that time, the FCP was considered as a grade that was sensitive to variation of the carbon and oxygen partial pressure in the sintering furnace. There was a large difference between the different production furnaces, DMKs, that was used then. For example, the furnaces had not as good insulation as the DMK-furnaces have today. The FCP was more sensitive to smaller air leakage in the furnace. The color on the surface of the FCP becomes bronze (nitration on the surface), due to a thin gamma phase enrichment on the surface. [1]
1.2 Purpose
The aim of the project is to evaluate possible new powder grades to be used as FCP, considering temperature variations and its influence on magnetic properties. The scope is to evaluate if there is a powder grade that is more sensitive to small variations during insert production.
The assignment is to evaluate different candidates to see which is most suitable to capture temperature variations. At the same time to get more rapidly indications if the furnace is starting to drift from expected results is needed. This specific grade used today (561) is currently a low volume grade and will increase in price since it is only used for some specific products. Likewise try to evaluate alternative methods to supervise and follow the status of the sintering process in the production furnace.
1.3 Limitations
Due to the project being time-limited, some limitations have been drawn. Four cemented carbide grades that indicated the most linear behavior towards the increase of temperature in the laboratory furnace, were
chosen out of ten after the first part of the project. The evaluation of the four grades were only performed for one sintering process, DQ process, in the production furnaces. Two sintering temperatures, 1410 ℃ and 1450 ℃ were compared.
2
2 Theory
2.1 Introduction of cemented carbide
The cemented carbide is a composite material. The main elements of cemented carbide are tungsten carbide cemented together with cobalt as a binder. In order to control and characterize the achieved material
properties of the cemented carbide, the magnetic properties of cobalt are therefore for great use. To optimize the desired properties of cemented carbide, some other elements are also included. During pressing of the green bodies, polyethylene glycol (PEG) is used to assist the pressing. PEG is a polymer that consists of chains with various lengths. These chains contain carbon-oxygen-hydrogen. However, PEG is degassed during the debindning step of sintering process. Sintering can briefly be described as a process used to set the mechanical properties and microstructure of the cemented carbide. [2] [3]
The mechanical properties of the cemented carbides, such as high hardness, good strength, etc., is due to the microstructure, which consists of two different phases. The carbide which is hard and brittle, and the
metallic binder which is soft and ductile. [4] [5] [6]
2.2 Microstructure of cemented carbide
The microstructure of cemented carbide can be described as tungsten carbide (WC) grains embedded in a cobalt-binder phase, Figure 1. Cobalt (Co) have excellent properties of the binding between WC and Co and is therefore used as the binder phase. The tungsten carbide phase is present as a hexagonal WC-phase as well as a mixed cubic carbide phase, the letter is also called the gamma-phase (γ). The microstructure of an etched cross section in one cemented carbide grade is illustrated in Figure 1. [4][7][8]
Figure 1. The microstructure of a cemented carbide from an etched cross section, characterized by using a light optical microscope (LOM).
The carbon content is an important parameter of the considered microstructure. Carbon is not only present in the WC it is partly also dissolved into the Co. Small variation of the carbon content can affect the total phase composition as well as the microstructure. This contributes to a modification of the properties. It is essential to control the carbon content to obtain excellent properties of the WC-Co based cemented carbides. It is possible to control the hardness of a powder grade, by controlling the carbon content in the specific material.
[7] [9] [2]
2.3 Introduction to the powder grades
Different cemented carbide grades are used at Sandvik Coromant. The grades are distinguished by their microstructure and properties. The difference between standard DA- and DQ-grades, which can be seen in Figure 2, is that the DQ-grades (right picture) besides WC and Co normally also contain gamma-phase raw materials, Titanium (Ti), tantalum (Ta) and niobium (Nb). The standard DA-grades (left picture, Figure 2) primarily consists of WC-Co with chromium (Cr) embedded. From Table 1 and 2, it is indicated that the different grades contains various amount of WC and Co. [3] [10]
3 Figure 2. The Figures are the microstructures from an etched cross section for a standard DA- respectively standard DQ-grade.
The left Figure is the microstructure of a DA-grade (477), as illustrated there is no γ-phase present. As for the right Figure, the microstructure of a standard DQ-grade (592) that contains γ-phase.
When the sintered grain size is 0.8 – 2 µm it is said that the structure is fine-grained. If the grain size is larger, 2 – 5 µm it indicates that the grain size of the structure is medium grained. The grain size of WC of a sintered structure and the Co- and γ-phase for each of the blanks can be found in Table 1 and 2. The FCP (561) contains gamma phase, despite that it is a standard DA-grade. [3] [10]
Standard DA-grades WC-grain size of a sintered structure
Co – phase [Vol.%]
Information about the grade HV 561 (FCP) Medium approx. 5 [µm] 11 Standard FCP grade gamma
phase containing grade
1320 – 1460
465 Fine 0.8 [µm] 10 Low Cr content 1550 – 1650
477 Fine - Medium 2 [µm] 6 WC – Co grade, no Cr. 1530 – 1630
903 Fine 0.8 [µm] 13.5 High Cr content 1450 – 1550
Table 1. The Standard DA-grades and their properties of the microstructure.
Standard DQ- grades
WC-grain size of a sintered structure
Co – phase [Vol.%]
γ – phase [Vol. %] HV
564 Medium size 4 [µm] 7 4.99 1450 – 1550
591 Medium size 4[µm] 7.2 4.88 1450 – 1550
592 Medium size 4 [µm] 7.25 5.64 1450 – 1550
574 Medium size 3,5 [µm] 5 4.93 1530 – 1600
594 Medium size 3,5 [µm] 5.75 6.57 1575 – 1675
575 Medium size 3 [µm] 4 4.93 1590 – 1650
Table 2. The standard DQ-grades contain medium size WC and different amounts of Co and γ-phase.
2.4 The manufacturing process 2.4.1 Cemented carbide production
The production of cemented carbide products can be described as a complex powder metallurgical process.
The first step towards cemented carbide blanks is mixing and milling of raw powder materials. Mixing of raw powder materials is a crucial step for mechanical properties and likewise for the final microstructure. To produce a homogeneous material and check the grain size, which can be performed by measuring the
coercivity (Hc), milling is used during the process. Milling fluid, which consists of a mixture of water (H2O) and ethanol (C2H5OH), is added to keep the particles in the mixture separated and to dissolve the pressing agent. The pressing agent is usually polyethylene glycol (PEG). PEG is normally added to the mixture to support the structure of powder during manufacturing of cemented carbides. Consequently, spray drying is used to produce granulated, ready to press, powders with desirable flow properties, and then pressing the green bodies into desired shape by using a mold and pressure. [11] It is important to consider that all steps during the powder production are linked to each other. If there is any change in any of the manufacturing step, it will influence the subsequent process as well as the final product. [2]
4 2.4.2 Sintering process
The sintering is a crucial process to get a dense sintered insert with desired properties from a porous pressed green body. The final dimensions and a pore-free microstructure of the component is obtained during sintering. [2]
Four main steps can be used to describe the sintering process; debindning, vacuum sintering (which is as well called solid state sintering), liquid phase sintering and cooling. To gain the desired mechanical properties, such as toughness and hardness, the process conditions are optimized to fit the powder grade.
[3][12]
Debindning
PEG is used in the pressed body, and the purpose of the debindning step is to remove PEG and impurities.
The reason is to avoid unwanted residue to be left in the sintered cemented carbide. The optimal temperature during debindning depends on several factors, such as constituents and composition of the pressed body, heating rate and the atmosphere. [2] [3]
During debindning the chamber is filled with hydrogen gas (H2). The H2-gas will help the decomposition of PEG-molecules and this process starts around 250 ℃. PEG-molecules will decompose into carbon
monoxide (CO), water (H2O) and methane (CH4) which can be described as gaseous products. The final stage of the decomposition is complete at 325-350 ℃. However, there are several factors, such as the load of the process charge, access of H2 etc. that can affect both the time and the temperature of the decomposition.
[3][13]
Solid state sintering
When debindning is complete, the solid-state sintering starts. This step is of great use to chemically reduce the oxides that have been created during powder production and to densify the blank. The aim is to avoid unwanted structural defects, such as porosity. Carbon is used as a reduction agent, meaning that the carbon content in the blank will be affected. The magnetic saturation (CoM) decreases due to more oxides are present. Carbon monoxide is formed both during debindning and solid state sintering, a vacuum pump is used to remove the CO from the cemented carbide. [3] [13]
Liquid state sintering
High temperatures are applied during liquid state sintering, to obtain fine grain growth of the microstructure.
The mechanism behind liquid state sintering can be described as the small grains dissolve in the liquid, which facilitate that larger grains grow. [3] [14] [15] [16]
Cooling
The solidification of Co-binders starts, depending on the powder blank composition, in the temperature range of 1250-1350 ℃. During cooling and depending cooling rate, Co may be relocated to the surface of the sintered blank, described as surface-Co. Slow cooling rate causes the amount of surface-Co increase. The composition of the gas atmosphere can also influence the amount of surface-Co. [3] [13]
2.4.3 Quality control of sintered cemented carbide
The quality control of the microstructure of the sintered cemented carbide blanks is performed by applying a non-destructive magnetic method, Figure 3. By applying the magnetic measurements, it is possible to
measure the coercivity (Hc) and the magnetic saturation (CoM). The measured CoM value is affected by the presence of carbon (C) and tungsten (W) redeemed in Co. [6] [2] [3] [17]
5 Figure 3. Before the measurements of Hc, corresponding to the WC grain size, on a concerned cemented carbide sample is performed. The non-magnetic Co-phase and the magnetic domains are arranged randomly (step 1). During Step 2, a magnetic will fully magnetize the binder of Co to magnetic saturation. The magnetic domains are then arranged in the same direction. What happens during step 3, is that the binder will try to de-magnetize again, as the magnetic field is no longer present. This
contributes to that the magnetic domains will almost be arranged randomly again, as for the first step. During the last step 4, the magnetic field needed to fully de-magnetize the magnetic remanence, from the last two steps (3 and 4) is measured and
corresponds to Hc. At last, the magnetic domains are arranged randomly again. [17]
The Hc and CoM values depends on the microstructure of the cemented carbide, which is illustrated in figure 4. The two parameters Hc and CoM are connected since the composition of the binder (expressed by CoM) also influences the WC grain size (expressed by Hc). For example, if the CoM-value is low, it
indicates a higher Hc-value which is related to a fine-grained microstructure. And if the CoM-value is high it is related to a low Hc-value (meaning that the microstructure is coarse-grained). A high CoM-value,
compared to grade limits, indicates that graphite is formed in the microstructure of the sintered cemented carbide blank. As the amount of tungsten in solid solution in cobalt decreases, it contributes to that the CoM increases for the sintered cemented carbide and vice versa. [2] [18] [19] [15]
The CoM/Co ratio is considered when the magnetic properties of the sintered cemented carbide is analysed.
The aim of the ratio is to indicate the tungsten content in the Co-binder phase and the carbon content in the sintered blank. For example, a low CoM/Co-ratio indicates that the carbon content is low but the tungsten content in the binder is high. While a high CoM/Co ratio indicates that the microstructure consists of a high content of graphite or redeemed carbon. [2] [3]
If the CoM-value changes, it indicates that the Hc-value likewise change. For example, if the magnetic saturation indicates low values, that means higher coercivity which is relatable to a fine-grained
microstructure. The microstructure of the cemented carbide tends to be coarse when Hc decreases and the magnetic saturation value increases. In this case the binder has a low W content due to the presence of high carbon content. To make Hc independent of the Co content in the grade, corrected Hc (Hc_corr) can be calculated. Hc_corr is calculated with help of equation (1) for respectively cemented carbide blanks. The equation of Hc_corr considers the expected Co-content (CoMT) for respectively candidate also the measured Hc- and CoM-value (CoMrv). [2] [20]
𝐻𝑐_𝑐𝑜𝑟𝑟 = 𝐻𝑐 ∗𝐶𝑜𝑀𝑟𝑣
𝐶𝑜𝑀𝑇 Equation 1
6 The relationship of the increased or decreased value of Hc and CoM with respect to the microstructure of the cemented carbide blanks is schematic illustrated in Figure 4. The Figure illustrates for instance that an increased value of Hc indicates that the microstructure of the cemented carbide blanks is fine-grained.
Figure 4. A schematic illustration of the microstructure for a cemented carbide blank, which indicates what causes the two quality parameters, Hc and CoM to increase or decrease. When the CoM value increases, graphite is formed in the microstructure while low value of CoM indicates that the cemented carbide consists of eta phase, which can be described as WC deficient of carbon.
[2]
2.4.4 Sintering temperatures
To obtain a dense cemented carbide blank with the right shape, composition and preferred microstructure different sintering processes are used. The two main sintering process that are used are the DA- and the DQ- process. These processes are optimized for different powder compositions, as the powder grades are divided into the category’s standard DA and DQ-grades. The DQ-process is used when the aim is to accomplish a controlled gradient of the microstructure, since it is a process used for gradient sintering. The DA-process is the original process used for basic cemented carbides sintered in the production furnace. [3]
The sintering temperature for cemented carbide is determined by their composition. Cemented carbides that contains of WC-Co normally have a sintering temperature that is above 1350 ℃. For cemented carbide blanks that have a fine-grained microstructure and that also have a medium-high cobalt content, is it more accurate to use the sintering temperature 1410 ℃. Higher sintering temperatures are applied when more coarse blanks are sintered in order to achieve a coarse grained material in the microstructure of the blank. In this case, the sintering process DA is applied with sintering temperature 1520 ℃. [3] [21]
2.5 Microstructure gradients
The formation of the gradient in the microstructure is possible for cemented carbides that have a multiphase structure. The cemented carbide must consist of WC, Ti, Ta, Nb (which are gamma phase raw materials) and the metallic binder, Co. This will lead to a higher temperature strength for the cemented carbide and the material will be more brittle. The formation of the gradient in the microstructure is normally used to obtain a gamma phase free surface region, to prevent the propagation of cracks in the sintered cemented carbide blank. [22]
The formation of the gradient is due to the large driving force from the lower nitrogen activity in the sintering furnace atmosphere, compared to the internal nitrogen activity in the cemented carbide. The activity of the gradient between the cemented carbide surface in contact with the atmosphere in the furnace and the inner parts of the cemented carbide results in a diffusion of the nitrogen through the Co-binder- phase. The gradient growth is parabolic, meaning that the growth varies linearly. Several factors influence
7 the gradient growth, such as the growth rate increases linearly with volume fraction of the binder phase. The gradient growth is schematically illustrated, Figure 5. [23]
Figure 5. Schematical illustration of the gradient growth in a microstructure of the cemented carbide in a DQ process
8
3 Method and equipment
3.1 Equipment 3.1.1 Sintering furnace
Two sintering processes were used to perform the evaluation of the cemented carbide blanks. Both
considered sintering furnaces, laboratory and production, are using the technique of vacuum sintering. [12]
The powder laboratory at Sandvik Coromant in Gimo uses a sintering furnace (DDK), to perform quality controls of the cemented carbide blanks. The DDK-furnace is used for small scale tests, while the production furnace (DMK) is used for larger scale. The DDK-furnace is a crucible furnace while the DMK-furnace is a muffle furnace. The considered furnace types are used for sintering of the cemented carbide grades. The atmosphere in the furnace is, for different process steps, vacuum, inert gas, or active gas.
The support for the blanks, that are used in both furnaces, are graphite trays that are coated with yttria. A charge in DMK-furnace also contains graphite boxes, for furnace control pieces, normally placed at the bottom of the furnace.
When the liquid state sintering temperature in the furnace was stabilized, an optical pyrometer was used to estimate the temperature. This step was performed in both laboratory and production furnace.
3.1.2 Magnetic measurement
To measure the magnetic properties of each cemented carbide blank, Koerzimat 1,096 HCJ/J-H was used for the evaluation. This specific instrument is a PC controlled measuring system for magnetic properties. By using a built-in program, the Hc- and CoM-value was determined.
The sintered cemented carbide blanks were arranged on a sample tray. The blank is moved from the sample tray to a precision scale, with help of a magnet. This is performed for one blank at the time, and the
precision scale measures the weight of the blank. Thereafter, the blank is placed in a holder to measure the Hc and CoM.
The CoM is measured with help of a magnetic field of a permanent magnet, that magnetizes the cemented carbide blank. The blank is withdrawn from the magnet when the magnetic saturation, corresponding to CoM, is achieved for the considered blank. The CoM is measured in precents.
To measure the Hc values of the cemented carbide blanks, an open magnetization circuit is used. The blank becomes polarized and the measured polarization corresponds to Hc. The Hc data is measured in kA/m.
These specific values were determined after each run in the laboratory and production furnace on every evaluated cemented carbide blank.
3.1.3 Characterization equipment
The most common method for characterization of the gradient growth is to measure the gradient growth in a light optical microscope (LOM). The LOM that was used for this project was ZEISS (Epiplan – Neoflur).
This method is not particularly exhaustive, although it can perform quality controls of the sintered cemented carbide.
3.2 Method
The experimental part of the project was divided into three parts. The first part contained an evaluation of several cemented carbide blanks in the DDK-furnace, to evaluate the blanks that are the most linear to the temperature increase, considered for Hc, Hc_corr and CoM. The second part was to apply the DDK results in the DMK furnace to see which temperatures the blanks are experiencing. The sintered measured results from DMK-furnace for DQ1410 and DQ1450 were compared to the measured results from DDK- furnace, from the temperature interval 1400 – 1500 ℃, to evaluate the temperature distribution in DMK- furnace. Last part was to evaluate the real production behavior in the DMK of suitable grades, from part 1, in DQ (1410 ℃ and 1450 ℃) to find good candidate for replacement of FCP (561).
9 3.2.1 Evaluation in laboratory furnace
3.2.1.1 Sintering in laboratory furnace
Nine cemented carbide grades, see Table 1 and 2, were evaluated together with the FCP (grade 561) in the laboratory furnace (DDK). The DDK-furnace is illustrated in Figure 6. To create the test pieces, the weight of the powder was measured and thereafter 22 pieces of respective powder composition was pressed into green bodies. Two pieces of each considered powder composition was included in the evaluation in DDK- furnace, to accomplish an average value of the Hc, Hc_corr and CoM.
Figure 6. A schematic illustration of the laboratory furnace, DDK that is used at Sandvik Coromant in Gimo.
One process charge consists of several graphite trays. The top and bottom trays marked as blue in Figure 7 are present as covers for the five graphite trays, which are white marked. The five graphite trays are arranged, as illustrated in Figure 7, for the most optimal sintering result, and therefore is position three and four used for this evaluation. The FCP (561) was included on both trays. The standard DQ-blanks (564, 574, 575, 591, 592 and 594) were arranged on the third graphite tray, and the standard DA-blanks (465, 477 and 903) was arranged on the fourth graphite tray.
Figure 7. Schematic illustration of one charge in DDK-furnace. Each charge consists of top and bottom trays that works as covers for the other five graphite trays that are used for the cemented carbide blanks.
10 Eleven process charges were performed in the DDK-furnace, within the temperature range 1400 – 1500 ℃.
For each process charge, the sintering temperature was increased with 10 ℃, Table 3. Table 3 indicates the process charge and the sintering temperature for each charge. Thereafter was a quality control performed for respectively sintered cemented carbide blanks, with help of Hc and CoM measurements.
Process Charge Sintering temperature
(1) CH834 1400 ℃
(2) CH835 1410 ℃
(3) CH836 1420 ℃
(4) CH837 1430 ℃
(5) CH838 1440 ℃
(6) CH832 1450 ℃
(7) CH840 1460 ℃
(8) CH843 1470 ℃
(9) CH846 1480 ℃
(10) CH847 1490 ℃
(11) CH848 1500 ℃
Table 3. The sintering temperature for each process charge in DDK-furnace.
3.2.1.2 FCP cycle in standard production
Every six months, the batch of the FCP (561) changes, to avoid the aging effects of the green bodies. And new limit values are derived for the considered sintering processes and temperatures. These limit values shall be a measure for the sintering process, to indicate when the sintering process is approved. The limit values have not been specially calculated for this work; it is data that Sandvik Coromant uses and is an average value from production. The average values (Hc and CoM) are created by evaluating the new batch of FCP (561) together with the outgoing batch of FCP (561) for several production runs before the old batch of FCP (561) is replaced. The limit values are produced for all processes in production.
The empiric limit values for Hc and CoM for both DQ1410 and DQ1450, see Table 6 and 7, were compared to the experimental data from DDK-measurements, that was measured for the temperature range which are presented in Table 3.
3.2.1.3 Evaluation of the microstructure gradient
After the sintering process was performed in DDK-furnace, in the temperature range 1400 – 1500 ℃, the following standard DQ-blanks was selected for evaluation of the microstructure gradient; 574, 575, 591 and 592. The gradient of the microstructure was evaluated for the following sintering temperatures 1400 ℃, 1410 ℃, 1430 ℃, 1450 ℃, 1470 ℃, 1490 ℃ and 1500 ℃.
The sintered cemented carbide blanks were mounted into bakelite. The following step was grinding (approximately 500 µm) of the mounts to avoid influence of surface defects, next step was polishing in several steps. The total preparation time for a mount is 10-15 minutes. Samples were then etched in 20 % Murakami so the microstructure can be seen. The gradient of each blank was characterized with help of a light optical microscope (LOM).
3.2.2 Evaluation in production furnace
3.2.2.1 Sintering temperature in production furnace
The ten cemented carbide grades (465, 477, 903, 561, 564, 574, 575, 591, 592 and 594) including the FCP (561) were sintered in the DMK-furnace. Two process charges, one for DQ1410 (DMK10 – 585) and one for DQ1450 (DMK10 – 586), was sintered in DMK10, the same DMK-furnace was used for both
temperatures. The graphite trays were arranged at the top of stack 1 (T1), middle of stack 3 (M3) and bottom of stack 6 (B6), Figure 8. When the sintering process was completed, a quality control was performed on each sintered cemented carbide blank, with help of Hc and CoM measurements. The Hc_corr was also calculated with help of equation 1. The measured sintering results (Hc, Hc_corr and CoM) from DMK- furnace was translated to the measured results from DDK-furnace, in the temperature range 1400 – 1500 ℃.
The aim was to estimate the experienced sintering temperature of the sintered cemented carbide blanks.
11 Figure 8. The schematic illustration of the two process charges in DMK-furnace, seen from the side.
3.2.2.2 Evaluation of suitable candidates
The four chosen cemented carbide blanks that demonstrated the most linear behavior of Hc and CoM with increasing temperature were further evaluated in DMK-furnace for two sintering temperatures, 1410 ℃ and 1450 ℃. For each process charge, three graphite trays and boxes were prepared. Each process charge, both DQ1410 and DQ1450, included other production orders of different cemented carbide grades from the production.
The arrangement of the graphite trays and boxes were the same as it is in the reality for the FCP (561), Figure 9, and it was the same arrangement for each process charge, both DQ1410 and DQ1450. The graphite trays were positioned at the top of stack 1,3 and 5 while the graphite boxes were arranged on the bottom of stack 2,4 and 6. The schematic illustration of each process charge, seen from the side and above, that was performed in DMK-furnace can be seen in Figure 9.
Figure 9. The schematic illustration of each process charge in DMK-furnace. The left picture is an illustration from the side, while the right picture illustrates the process charge from above.
The number of blanks was the same for each process charge. A total of six samples of the FCP (561) was used, one blank each on the trays that were arranged on T1, T3 and T5 and one blank of the FCP (561) in each graphite box. In the graphite box there was only room for one sample of each grade and four samples in total, Figure 9.
12 For the evaluation in DMK-furnace, in total 18 production runs were performed. Due to that the DQ1450- process is planed more frequently in the production compared with DQ1410, ten of the production runs were performed in DQ1450. While eight runs were performed in DQ1410-process. The charge number for each process charge performed in DQ1410 respectively DQ1450 can be found in the appendix.
For DQ1410 process, the eight process charges were performed in different furnaces: DMK03, DMK05, DMK06 and DMK10. Most of the process charges for DQ1410 were performed in DMK10, which can be seen in the appendix. For DQ1450, the ten process charges were performed in DMK05, DMK10, DMK11 and DMK13, additional data of each process charge can be found in the appendix.
13
4 Results and discussion
4.1 From the laboratory furnace, part I 4.1.1 Evaluation of cemented carbide blanks
The Hc_corr value decreases meaning that the WC grains grows with increased sintering temperature for all cemented carbide blanks, Figure 10. To follow the temperature in a DDK-furnace by calculating Hc_corr, equation 1 section 2.4.3, it is beneficial if the calculated value have a high linearity towards the temperature.
Figure 10. The experimental results of Hc_corr indicates a linear behavior to the increase of temperature.
The DDK-results demonstrates a linear behaviour of the quality parameters Hc_corr, i.e. grain size, and CoM, i.e. carbon content in the binder phase, to the increase of sintering temperatures. When selecting the candidates for further evaluation in DMK-furnace, the R-value of the three quality parameters were considered. When the R-value for respectively parameter, being almost or close to 1, it indicates that the cemented carbide blanks are strongly linear to the increase of temperature. The R-value close to 1
(approximately 0.99x) indicates a perfect positive linear relationship – as one variable increases in its value, the other variable increases in its values through an exact linear rule.
The R-value of Hc_corr, which can be seen in Table 4 and Figure 11, indicates that the most linear candidates are 903, 592, 564, 591 and 465. Meaning that the grain size grows linearly for these cemented carbides. While the candidates that approaches the least linear behaviour are 574, 594, 477 and 575. It also indicates that there is no significant difference between the positions in DDK-furnace, which is seen by the FCP (561) that was positioned on both tray 3 and 4.
The data from the R-value of CoM, which shows the linearity of carbon content in binder phase with increased sintering temperature, shows that the FCP (561) independent of the position in the furnace, is the most linear candidate followed by 592, Table 4. While some blanks are not linear to the temperature (465, 477 and 903). The R-value of Hc_corr measurements indicates that the FCP (561) is not as linear to the increase of temperature as other cemented carbide blanks, 465, 591, 564, 592 and 903, Figure 11. The R- value of Hc_corr for the FCP is 0,968 while the blank 903 has the highest R-value 0,992 which is closer to the value 1, Table 4 and Figure 11. The higher the linearity of Hc and Hc_corr meaning that the grain size of WC grows equally linearly within the entire temperature range 1400 – 1500 ℃ in the DDK-furnace.
8 11 14 17 20
1380 1400 1420 1440 1460 1480 1500 1520
Hc_corr [kA/m]
Temperature [℃]
Evaluation of the Hc_corr with sintering temperature
561 - T3 465 477 903
561 - T4 564 574 575
591 592 594
14 Blank R-value
(Hc)
Blank R-value (Hc_corr)
Blank R-value (CoM)
903 0.994 903 0.992 561 (4) 0.981
592 0.991 592 0.992 561 (3) 0.956
564 0.989 564 0.988 592 0.944
465 0.988 591 0.988 594 0.941
591 0.987 465 0.987 575 0.935
561 (3) 0.976 561 (4) 0.968 564 0.916 561 (4) 0.971 561 (3) 0.959 574 0.915
575 0.966 575 0.954 591 0.865
477 0.955 477 0.954 477 0.337
594 0.943 594 0.921 903 0.267
574 0.939 574 0.912 465 0.164
Table 4. The cemented carbides are arranged after the decrease of the R-value for respectively quality parameters.
Figure 11. The R-value for the calculated Hc_corr, i.e. corrected grainsize for the different cemented carbide blanks.
The behaviour of CoM, i.e. carbon content in the binder phase, towards the increase of temperature, does not vary a lot for the considered cemented carbide blanks, Figure 12. Meaning that it is hard to see that CoM has a linear temperature dependence in the DDK-furnace. The carbon content in the binder of each cemented carbide is probably more affected by other chemically parameters from e.g. debindning step than it is of the temperature, which can be seen in Figure 12. If the R-value of CoM is considered, Table 4, it shows that the cemented carbide that is almost as linear as the FCP (561) is the blank 592.
Figure 12. The experimental results of the CoM-value, i.e. carbon content in binder phase, with increasing temperature.
0,9 0,95 1
R-value from Hc_corr
Cemented carbide blanks
The linear behaviour of R-value from Hc_corr- values for the different cemented carbide blanks
3,5 4,5 5,5 6,5 7,5 8,5 9,5 10,5
1380 1400 1420 1440 1460 1480 1500 1520
CoM [%]
Temperature [℃]
Evaluation of CoM with increasing temperature
561 - T3 465 477 903
561 - T4 564 574 575
591 592 594
15 The results of CoM between the two different sintering temperatures 1400 ℃ and 1500 ℃, Table 5, shows that the standard DA-blanks has the smallest difference in variation compared to the other DQ-blanks and the FCP (561). From the same results is it exhibited that the FCP (561) indicates the largest difference in variation between the two sintering temperatures, compared to the other sintered cemented carbide blanks.
For the standard DA-blanks, 465, 477 and 903, the CoM value decreases with higher sintering temperature.
The FCP (561) is also a DA-blank, although the CoM value increases with increased temperature, independent of the position of the two trays in DDK-furnace, Table 5.
Cemented carbide blank 1400 ℃ CoM [%] 1500 ℃ CoM [%] Range
561 – 3 9.17 9.455 + 0.285
561 – 4 9.2 9.485 + 0.285
594 4.78 4.93 + 0.15
574 4.345 4.475 + 0.13
575 4.15 4.27 + 0.12
564 6.595 6.71 + 0.115
591 6.36 6.46 + 0.1
592 6.455 6.55 + 0.095
903 10.22 10.21 - 0.01
465 8.175 8.16 - 0.015
477 4.925 4.905 - 0.02
Table 5. The experimental results of the variation of CoM, i.e. carbon content in binder phase.
Due to the argument of R-value close to 1 indicates strongly linear behavior of the microstructure to the increase of temperature, the candidates that were further evaluated in DMK-furnace was 465, 591, 592 and 903. The cemented carbide 564 also indicated a strongly linear relationship to the temperature increase, although the blanks 591 and 592 are newer variants of the blank 564. Therefore, is the blank 564 excluded from the evaluation in DMK-furnace.
The carbon content in binder phase, CoM, for the different cemented carbides does not vary much with increased sintering temperature, Figure 11. This result indicates that the CoM, i.e. carbon content in binder phase, has a very small temperature dependence, compared to the grain size (Hc). The carbon content in the binder is probably more affected of the different chemical process steps during a sintering process, although this was not further evaluated. The R-value of CoM showed that the FCP (561) followed by the cemented carbide 592 was the most linear candidate with increased temperature, meaning that the FCP (561) is the candidate that is most equally affected throughout the temperature range in DDK-furnace.
The results also showed, Table 5, that the standard DA-blanks had a lower variation between the sintering temperatures 1400 ℃ and 1500 ℃ compared to the DQ-blanks. It is not further evaluated in this study why the CoM value increases for some sintered cemented carbide blanks (DQ). One can assume that these grades are more suitable for this sintering process DQ.
To summarize, the evaluated cemented carbide blanks that follows the linearity with increased sintering temperature in DDK-furnace, could be used as the future FCP (561).
4.1.2 The empiric limits for the FCP from production
The values, Table 6 and 7, for FCP (561) batch 646 are empirically established from several runs in
production furnaces, DMK, same limits are used in the lab furnace, DDK. The empiric limits are considered for both DQ1410 and DQ1450.
The FCP (561) has a production target value for both Hc and CoM. The target value determines the
maximum and minimum limits for the FCP in which the production sintering process is approved. For e.g.
DQ1410, Table 6, the production target value of Hc is 12.4 kA/m and for CoM 9.4% meaning that the maximum and minimum value of Hc appear 0.7kA/m respectively for CoM 0.5% higher and lower than the target value. Similar results are obtained for DQ1450, see Table 7.
16 Empiric limits for FCP for DQ1410 in DMK furnaces
Min Lower alarm Target Upper alarm Max
Hc [kA/m] 11.7 11.9 12.4 12.9 13.1
CoM [%] 8.9 9.1 9.4 9.7 9.9
Table 6. The empiric limit values for Hc and CoM for DQ1410 in DMK-furnaces.
Empiric limits for FCP for DQ1450 in DMK furnaces
Min Lower alarm Target Upper alarm Max
Hc [kA/m] 10.8 11 11.5 12 12.2
CoM [%] 9 9.2 9.5 9.8 10
Table 7. The empiric limit values for Hc and CoM for DQ1450 in DMK-furnaces.
From the results in 4.1.1, Figure 10, all the tested blanks showed linear temperature dependence of Hc_corr.
There are several factors that affects the cemented carbide, e.g. the microstructure and the chemistry (as the manufacturing process) although in this study it is assumed that the only parameter that affects the grain growth (Hc) and carbon content (CoM) is the temperature during a sintering process charge in DDK- furnace.
The empiric limit values, Table 6 and 7, for the considered batch of FCP are compared to the Hc, i.e. grain size, and CoM, i.e. carbon content in binder phase, results from the DDK-furnace, this is completed for both DQ1410, Table 8, and DQ1450, Table 9. To fit the empiric limits, the DDK temperature range 1400 – 1500 ℃ was extrapolated. There is no measured sintering data from DDK-furnace lower than 1400 ℃ and higher than 1500 ℃ although it is assumed that the empiric limits of Hc, i.e. grain growth, and CoM, i.e.
carbon content in binder phase, for respectively DQ1410 and DQ1450, occurs outside of the temperature interval. To confirm the results in Table 8 and 9, sintering processes should therefore be obtained for lower temperatures than 1400 ℃ and higher than 1500 ℃.
4.1.2.1 The empiric limit values DQ1410 and DQ1450
When comparing the production limit values with the results from DDK-furnace for the same batch of FCP (561), the results of Hc, grain growth, indicates that the production target value is reached when the sintering temperature is at 1410 ℃. Which also is the expected sintering temperature for the DQ-process in DMK- furnace. The maximum and minimum limits of Hc occurs at 1440 ℃ respectively 1360 ℃. The empiric limit values from production for Hc occur within the temperature range 1360 – 1440 ℃, Table 8.
Comparison of the production limit values, with the results from DDK-furnace for the same batch of FCP, the results of CoM indicates that the production target value is reached when the sintering temperature is at 1490 ℃. The extrapolated data of maximum and minimum value of CoM occurs 1570 ℃ respectively 1390 ℃. The empiric limit values from production for CoM occur within the temperature range, 1390 – 1570 ℃, Table 8. Similar results can be seen for DQ1450, for both the grain size (Hc) and carbon content (CoM) in Table 9.
17 From DDK –
results
Empiric limits for DMK – furnaces
From DDK – results
Empiric limits for DMK – furnaces
Temperature [℃]
Hc [kA/m] Hc [kA/m] CoM [%] CoM [%]
1360 13.1 Max
1370 12.9 Upper alarm
1380
1390 8.9 Min
1400 12.57 9.17 9.1 Lower alarm
1410 12.345 12.4 Target 9.21
1420 12.2 9.225
1430 11.97 11.9 Lower alarm 9.26
1440 11.73 11.7 Min 9.3
1450 11.39 9.33
1460 11 9.335
1470 10.48 9.365
1480 9.915 9.365
1490 9.545 9.415 9.4 Target
1500 9.315 9.455
1510 1520
1530 9.7 Upper alarm
1540 1550 1560
1570 9.9 Max
Table 8. The measured Hc and CoM values from DDK-furnace in the temperature range 1400 – 1500 ℃, compared to the empiric limit values for Hc and CoM from DMK- furnaces for the process DQ1410.
18 From DDK –
results
Empiric limits for DMK – furnaces
From DDK – results
Empiric limits for DMK – furnaces
Temperature [℃]
Hc [kA/m] Hc [kA/m] CoM [%] CoM [%]
1390 9 Min
1400 12.57 9.17
1410 12.345 9.21 9.2 Lower alarm
1420 12.2 12.2 Max 9.225
1430 11.97 12 Upper alarm 9.26
1440 11.73 11.5 Target 9.3
1450 11.39 9.33
1460 11 11 Lower alarm 9.335
1470 10.48 10.8 Min 9.365
1480 9.915 9.365
1490 9.545 9.415
1500 9.315 9.455
1510
1520 9.5 Target
1530 1540
1550 9.8 Upper alarm
1560 1570
1580 10 Max
Table 9. The measured Hc and CoM values from DDK-furnace in the temperature range 1400 – 1500 ℃, compared to the empiric limit values for Hc and CoM from DMK- furnaces for the process DQ1450.
With this FCP results in the DDK and the evaluation of tested results in DQ-process (1400 – 1500 ℃) new limits for the lab furnace can be implemented, Table 10. For DQ1410 and DQ1450, the limit values need to be adapted to fit the temperature range, which is more accurate compared to the experimental data. The aim of having a range in 40 ℃ is to have more narrow limits in DDK-furnace, ensuring the grain size (Hc) to be close to the target. It is easier to apply these narrow limits for Hc since the grain size is more temperature dependent compared to the carbon content (CoM) which have been showed as more dependent of the chemistry during a sintering process.
DQ1410 Narrow limit values for DDK DQ1450 Narrow limit values for DDK
Hc [kA/m] Temp. [℃] Hc [kA/m] Temp. [℃]
Min 11.9 1430 Min 10.8 1470
Lower limit 12.2 1420 Lower limit 11 1460
Target 12.3 1410 Target 11.4 1450
Upper limit 12.5 1400 Upper limit 11.5 1440
Max 12.6 1390 Max 12 1430
CoM Temp. [℃] CoM Temp. [℃]
Min 8.9 1390 Min 9.26 1430
Lower limit 9.1 1400 Lower limit 9.3 1440
Target 9.2 1410 Target 9.33 1450
Upper limit 9.26 1420 Upper limit 9.34 1460
Max 9.3 1430 Max 9.37 1470
Table 10. The new suggested values for Hc and CoM for both DQ1410 and DQ1450.
The current FCP (561) limits to be closer to the target temperature, can be narrowed for the DDK-furnace.
This would give the possibility to capture smaller variations using Hc, Hc_corr and CoM. Especially the grain size, represented by Hc and Hc_corr, have a high potential to become more at target by the narrower limits.
19 4.1.3 The microstructural gradient for standard DQ-blanks
The results are showing that the gradient, surface zone with a changed microstructure, of each standard DQ- blank grows with increasing sintering temperature. The gradient growth of each blank is dependent of the temperature, the gradient increases approximately 5-6µm/100 ℃. If the gradient growth is considered for a smaller temperature range, it is seen that the gradient increases with a maximum of approximately
1µm/10 ℃, Figure 13.
Figure 13. The experimental results of the microstructure gradient growth.
4.1.3.1 Structure evaluation for the blanks 591 and 592
The gradient was evaluated for the DQ-blanks 591 and 592 for the sintering temperatures 1400 ℃, 1450 ℃ and 1490 ℃ from DDK-furnace, Figure 14 and 15, these temperatures were evaluated to avoid unwanted structural defects as large grains in the microstructure. Since all the tested grades showed similar linearity only 591 and 592 were chosen to present in this report. The results, Figure 14 and 15, show the gradient growth for the blank 591 and 592 for increased sintering temperature. The microstructure shows that gradient, i.e. the zone with a slightly different chemical composition, grows with increased sintering temperature.
When the sintering temperature increases from 1400 ℃ to 1450 ℃, the gradient of the blank 591 increases with 0,9µm/10 ℃. The increase of gradient from 1450 ℃ to 1490 ℃ is approximately 0,4µm/10 ℃. This means that the gradient growth is two times larger before the sintering temperature exceeds 1450 ℃, Figure 14. Similar results were obtained for the cemented carbide blank 592, Figure 15. Although for the sintered blank 592 it is showed that the growth rate increases approximately 3 times when the sintering temperature 1470 ℃ is exceeded, Figure 13. It is known that the gradient growth increases linearly with volume fraction of the binder phase, Table 2 shows that the two considered cemented carbides consist of almost the same fraction of the Co-binder phase. Although the volume fraction of γ-phase differs between the two cemented carbides, 591 and 592. Which could be a reason for the increase of gradient growth for 592 when the temperature 1450 ℃ was exceeded, since the blank 592 contains of a higher volume fraction of γ-phase, although it was not further evaluated.
15 19 23 27 31 35
1380 1400 1420 1440 1460 1480 1500 1520
Gradient [µm]
Temperature [℃]
The growth of the microstructuregradient
591 592 594 574
20 Figure 14. The pictures are obtained for the sintered cemented carbide blank 591.
Figure 15.The pictures are obtained for the sintered cemented carbide blank 592.
4.1.3.2 Gamma phase porosity for grades 591 and 592
Since the gradient thickness, the surface zone with a changed microstructure, increases with increased sintering temperature, it was also interesting to see the influence on porosity. The gamma phase porosity were evaluated for the two sintering temperatures 1400 ℃ and 1490 ℃ for the two sintered cemented blanks 591 and 592, Figure 16 and 17.
1400 ℃ 1450 ℃
1490 ℃
1490 ℃
1450 ℃ 1400 ℃
21 Figure 16. The left picture is the microstructure for the sintered cemented carbide blank 591 and the right for 592.
Figure 17. The left picture is the microstructure for the sintered cemented carbide blank 591 and the right for 592.
The gamma phase porosity, Figure 16 and 17, decreased or even eliminated with increased sintering
temperature. For the sintered cemented carbide blanks 591 and 592, it was noticeable that the gamma phase porosity was much more visible at lower sintering temperature 1400 ℃. It is exhibited that the gamma phase porosity is the same for the two blanks. It is also exhibited that the gamma phase porosity decreases for all the tested DQ-grades towards the increase of sintering temperature. All the tested grades had similar
increase of gradient as the temperature increases. It could be seen that gradient growth increased more after 1450 – 1500 ℃, to confirm this more investigation needs to be done.
4.2 Laboratory furnace temperature results translated in production furnace results, part II
Blanks sintered in to two production process charges were evaluated for Hc and Hc_corr, and the results were translated to the temperatures from the lab furnace. The cemented carbide blanks were positioned on graphite trays at the top of stack 1 (T1), middle of stack 3 (M3) and bottom of stack 6 (B6). The sintered results from DMK-furnace from process charge DMK10 – 585 (DQ1410) respectively DMK10 – 586 (DQ1450) were plotted to the temperature measurements in DDK-furnace, under the assumption that the only parameter that affects the sintered results of the microstructure is temperature. For the two sintering temperatures 1410 ℃ and 1450 ℃, the same DMK-furnace was used, DMK10. The respectively three dots in Figure 18 and 19 represent the three sintered positions in DMK-furnace.
The translated grain size (Hc and Hc_corr) results from DQ1410 showed that the experienced temperature range, for all cemented carbides, is 1360 – 1435 ℃ respectively 1360 – 1430 ℃, Figure 18. It is detected that the sintered cemented carbide blanks experiences different temperatures, despite that they have the same placing in the furnace, i.e. have had the same temperature. Similar results were obtained for DQ1450, Figure 19. It is exhibited that some of the sintered cemented carbide blanks experiences lower temperatures than 1400 ℃, Figure 18, although there is no measured data in DDK-furnace lower than 1400 ℃. It is therefore assumed that some sintered cemented carbide blanks experience lower temperatures than 1400 ℃. To confirm that the sintered cemented carbide blanks, 561, 592, 591, 903, and 477 experiences lower
1400 ℃
1490 ℃ 1490 ℃
1400 ℃
