Mechanical Engineering - Applied Metallurgy

Completed notes of the course

Complete course

1 Chiara Moreschini NOTES OF APPLIED METALLURGY Prof. C. Mapelli – AA 202 1 /2 2 0. INTRODUCTION STEEL STRUCTURAL CONSTITUENTS à See chapter 2.1 of the book The diagram shows the equilibrium phases of steel , which are: a) Austenite ( ! ) b) Ferrite ( " ) c) Cementite (Fe 3 C) d) Perlite ( " + Fe 3 C) 2 The main temperature s to be remembered on the phase diagram are: - A 1 equilibrium temperature of alpha to gamma phas e; - A 3 alpha to gamma transformation temperature, when we reach this line the austenite starts to decompose to form ferrite ; - A CM equilibrium temperature between austenite and cementite . Non - equilibrium phases: bainite and martensite The diagram only shows the equilibrium phases of steel, but there are also non - equilibrium phases reached when the cooling rate of the steel is too high, such as: a) Bainite (lower and upper) b) Martensite à Bainite is formed when the CR (cooling rate) of austenite is too high to allow the formation of the lamellae which characterize the pearlite equilibrium phase, so we have the formation of a phase still composed of Fe 3 C and ferrite but this time we have Fe 3 C particules inside a matrix of ferrite instead of lamellae. In case of bainite we have still nucleation and growth of the grains because the CR is high but not critical (diffusion of carbon allowed even if small). à The formation of martensite is instead characterized by a critical CR which do not allow any diffusion of carbon inside the steel, so we have a diffusionless transformation in which any nucleation and growth happen but we have the instantaneous formation of the new phase. The formation of martensite is the only transformation with no diffusion, wh ile all the oters (also bainite formation) have at least a little C diffusion. N.B. • Carbide : compound composed of carbon and a metal 1 , for example iron - carbides, Mn - carbides, etc… 1 Carbide = carburo = molecola formata da carbonio ed uno o più elementi metallici 3 • E utectic system : is a heterogeneous mixture of substances that melts or solidifies at a single tem perature that is lower than the melting point of any of the constituents (before be coming solid the composition pass through the liquid phase) . This temperature is known as the eutectic temperature and is the lowest possible melting temperature over all of the m ixing ratios for the involved component species. • Eutectoid system : is an eutectoid system in which the transformation to the solid phase begins from the solid phase of one of the components instead of from the liquid phase. When the solution above the transformation point is solid, rather than liquid, an analogous eutectoid transformation can occur. For instance, in the iron - carbon system, the austenite phase can undergo a eutectoid transformation to produce ferrite and cementite , often in lamellar structures such as pearlite and ba inite . This eutectoid point occurs at 723 °C and about 0.8% carbon. • Eutectoid ic point: point in the phase diagram in which the solid phase of a certain composition transform into a mixture of 2 or more different solid phases (so the transformation from a solid phase lead to another solid phase, without becoming liquid lie in the eutectic case ). For steel the eutettoidic point is around the concentration of carbon of # = % . '' % ( perlite ) • Hipoe utectoid ic is a composition in which the concentration of the alloying element is composition: lower than the one of the eutettoidic point. In the case of steel we have hipoeutettoidic steels for concentrations of steel # < % . '' % ( ferrite + perlite ) • Hypere utectoid ic is a composition in which the concentration of the alloying element is composition: higher than the one of the eutettoidic point. In the case of steel we have h y p er eutettoidic steels for concentrations of steel # > % . '' % ( perlite + cementite ) • Pearlite : 4 THERMAL TREATMENTS à See chapter 2.2 of the book Classification of treatments Classification of treatments is made on the basis of the heating temperature . If we are over the critical thermal temperature we can have: a) Annealing treatments (= ricottura) b) Normalizing treatments (= normalizzazione) c) Quetching treatments (= tempra) If we are under the critical thermal temperature we can have: d) Tempering (= rinvenimento) All the types of treatments are characterized by the same 3 steps: heating , maintenance and cooling . The first 2 steps are similar in all the process, while depending of the cooling rate we obtain different kind of treatment s: - Annealing has the slower cooling rate and it is close to the equilibrium cooling; f irst cooling is made in the oven with a ,- = − 5 ÷ 10° , / ℎ and then in the air when the material has reached a temperature 6 < 300° , ; - Normalizing is faster than the annealing about 1 order of magnitude in the cooling rate; it is characterized by ,- = − 50 ÷ 60° , / ℎ and only air cooling to make it faster , without any first cooling in the oven like annealing; - Quenchin g has the fastest cooling and allows to obtain martensite ; the c ritical cooling rate achi e ved by using a medium (t y pically water or oil, usually we use oil because water let the materi a l cool to fast and there is the risk of failure or cracking of the mater ial) . CCT (Continuous Cooling Transformation) curve Depending on the CR after the heating of the material we obtain different thermal treatments: the cooling rate can be represented as the slope of the curve in the temperature - time diagram, which is called CCT curve . A à F: austenite to ferrite A à P: austenite to pearlite A à B: austenite to bainite A à M: austenite to martensite 5 Quenching The external region of the material begins to transform to martensite as the cooling begins, while the internal part is still austenite , but the two phases have different geometry of the elementary cells which occupy different volumes (martensite occupies an higher volume than austenite due to its elongated elementary cell BCT) , so some stress starts to appear in the material, which can cause failure. In particular the inte rnal part is subjected to compression while the surface is put in tension. To avoid this we can use a lower cooling which still allow to have martensite or we can use 2 consecutive cooling steps : first we use a high cooling rate and th e n interrupt the cooling when martensite begins to form (at T=M s ) until the temperature equalizes in all the material , so that the internal part can reach a higher temperature, and then make a second cooling (until T=M f ) . Tempering Treatment under the critical temperature used to reduce the brittleness of the material, so it is used after the quenching to increase ductility and toughness of the steel. In CMn alloys (Carbon - Manganese alloys) tempering can lead to the Krupp disease , which is an undesired brittleness due to the formation of MN carbides on the grain boundaries. Quenching to obtain the martensite and then tempering to reduce brittleness [ See chapter 2. 1 and 2.2 of the book + slides for the other part of the lecture ] 6 TENSILE TEST The tensile test is a test in which a force is applied on the specimen in order to discover its stress - strain relationship. During the test the machine measures the deformation and the displacement of the bean mounted of the machine: - Extensiometer on the specimen measuring the deformation during the elastic phase of the material ; - Loading cell measuring the force applied to the specimen i n stant by i n stant ; First we enter in the elastic behavior of the specimen and i n this zone we use the extensiometer to detect the deformation ; b efore entering the plastic behavior the extensiometer is removed and when we reach the p lastic behavior the stress - strain relation is characterized by large displacements and stric t ion (or ne cking ), in which the deformation is concentrated in the center of the specimen . The higher the elongation up to fraction, the higher the ductility of the material. During a tensile stress we can measure: c) Elastic limit : maximum stress we can apply to the material remaining in the elastic zone, which often coincides with the yield strength 2 ; 9 ! = : ! ; " d) Yield strength : stress reached in the point where the plastic behavior begins 9 $ = : $ ; " e) Ultimate tensile strength : maximum stress achieved before failure 9 % = : % ; " f) Ultimate deformation : difference between ultimate length and initial length divided by initi al length (actually it would be the logarithm of Lu/L0 but for macroscopic deformation we can approximate to the ratio) < = = & − = " = " 100 = ∆ = = " 100 g) Necking : difference between ultimate area of the specimen and initial area divided by initial area ? = ; " − ; & ; " h) Hardening coefficients : @ i) Elastic modulus : A 2 Sforzo di snervamento 7 From the tensile test we can also deduce the c onstitutive relation s of materials , which are very important because they describe the relationship between the stresses in the material and the deformation when a force is applied to a material. In the elastic period we have a linear relationship between the stresses and the deformations (E: elas tic modulus) while in the plastic period it is an exponential relationship (n: hardening coefficient) : ABCDEFG HI@J : 9 = A L MBCDEFG HI@J : 9 = N L ' § The elastic modulus in the elastic period of a m aterial CANNOT be modified by any treatment performed on the material because it depends on the microscopical structure of the material ; some values of elastic modulus of important materials are: A ()!!* = 210 PMC A +**&%,',&% = 70 PMC A (-,. ,/0' ( 2 . 3 ,(+ ) = 165 PMC § Instead in the plastic period the elastic modulus defined as the slope of the cure changes depending on the load applied on the material. The Gauge length The Gauge length is the length marked on the specimen by the operator (for example two lines marked on the specimen with a pe n) and in is the useful to measure the deformation of the specimen during the test . The G auge length occupies the central part of the specimen and does not contain the extremes where the specimen is constrained to the machines. The r elationship between t he G auge length of the specimen and its area can be in general expressed as = " = R S ; " but the specific relation depends on the type of specimen we use during the test: a) For round specimen : = " = 5 T " b) For prismatic specimen : = " = 5 , 65 S C " V " Round specimen Prysmatic specimen 8 If the fracture occurs out of the G auge length the result s obtained by the test are NOT significant, because it means the specimen is affected by the force applied by the mechanical constraints of the machines on the extremes of the specimen. To consider t he tensile test results valid the fracture of the specimen must happen inside the Gauge length . Elastic period T he elas tic period is characterized by deformation which disappear s when the load is removed from the material, instead p lastic behavior is characterized by a permanent deformation . The material il considered in its elastic period as long as the difference between its stress - strain curve and the initial slope of the curve (perfect elastic behavior) is les s than the elastic limit e = 0.002 . The elastic modulus E is the slope measured on the curve between stresses and strain and it points out how m u ch a material is rigid . If the elastic modulus increases , the material is affected by a lower deformation because it becomes more rigid . The higher the elastic modulus, the higher is the attitude of the material to transmit vibrations because it is more rigid and it is not able to absorb vibrations, which instead are dampe d by less rigid materials, with lower E. The area under the curve represents the specific energy absorbed by the material per unit of volume. In the elastic period the area is a triangle and is the energy absorbed by the material which can then be given back as the load disappears. Yield strength The elastic and the plastic period are separated by a point called yield strength , in which the behavior of the material changes from linear to exponential. In order to determine the yield point we take the initial slope of the curve and translate it by 0.2% (or 0.5%) according to the regulations. The point of intersection between the strass - strain curve and the new line is the yield point and its stress is named yield stress ( R y,0.2% ) : therefore the yield point is defined as the point in which the defor mation left inside the material after the load is removed is equal to 0.2% (or 0.5%). 9 The transition from elastic to plastic behavior could be different depending on the material and could show continuity , discontinuity , oscillations , etc. If the first elastic part of the curve shows a discontinuity wrt the second plastic part there could appear problems such as strain aging , which appears as oscillations of the curve in the point of discontinuity. The strain aging is a phenomenon of instabi lity which causes heterogeneous deformation of the grains of the material, so only some portion of the grains are deformed at the beginning of the plastic period and it causes localized failure in the alloys, so it is strongly unwanted. Plastic period When the stress in the material reach es the yield strength , the elastic period finishes and the material enter s the plastic period . The relationship between stress and strain changes and becomes exponential with a s exponent the hardening coefficient n: 0.1 < n < 0.25 The hardening coefficient n depends on the orientation of crystals inside the material (microscopic scale). When n is high it means that all the crystal are deforming and the deformation is homogeneous among all the crystals, while if n is low just a little portion of crystals is participating to the deformation . The ductility of a material increases as the hardening coefficient increases because of the wide cooperation of the crystals (effort to resist to the deformation distributed on a larger nu m ber of crystals) . Strain aging 10 Real and engineering curve The maximum point of the curv e is taken as the resistance of material at fracture, but this isn’t the real stress that causes the fracture, because the fracture happen s for lower stress and higher strain wrt this point. In fact around the point of maximum the striction begins to hap pen, but the machine cannot detect the necking, so the curve measured by the material does not take into account the decrease of area caused by the striction but still computes the stress as a ratio between the force and the original area S 0 . Therefore there is a d ifference between ideal stress F/S 0 and the actual stress F/S : the machine d raw a decreasing curve due to the fact that it still considers the area to be S 0 even if the real curve would be increasing . E ven if the curve drawn by the machine is not accurated it is in favor of security so it is still useful. : ; > : ; " VJGCWDJ ; " < ; On the curve drawn by the machine we cannot find with certainity the exact point of failure of the specimen . Quality index of metal alloy The quality index X 5 is the i nverse relationship between ductility (A) and resi stance ( R = strength of the material ) : Y + = - ∙ < = GIDE This inverse relation is due to the relation of ductility and resistance with the phenomenon of dislocation of the cells: t o obtain higher resistance we have to avoid the movement of dislocations while to improve ductility we have to promote dislocations. Another quality index ( X 6 ) we can use is the relationship between the strength (R) of the material and the necking (Z) : Y 7 = - 100 100 − ? 11 PROPERTIES OF THE MATERIALS: Toughness, resistance, resilience and ductility We can define many properties of the materials: • Ductility : ability of the material to deform up to fracture and represented by its Elongation ; this property of the material describe how much the material is able to absorb energy and converting it into deformation before failure . The higher the hardness coefficient n, the higher the ductility, because the defor mation is supported by a larger number of crystals . • Resistance : is t he maximum stress a material can stand and it is represented by the maximum point of the curve ; it does not coincide always with the point of fracture, which instead can happen also in t he decreasing part of the curve. • Toughness : is energy absorbed and given back by the material up to the fracture during a tensile test , so a test in which the increase of force on the material is quasi - static . In the elastic period all the energy absorbe d is then given back and this is w h y the material can recover its initial shape ; i n the plastic period instead the material dissipates the work imposed by the machines so part of the energy cannot be given back and remains stored in the material or dissipa ted (mostly as heat ) . We want the materials to be tough for safety reasons, to avoid failure under load. In general ductile materials are also tough while fragile materials aren’t tough, but we can also obtain both fragile and tough material through some treatment, like grain refinements 3 . • Resilience : is the energy the material can absorb during an impact . The test for resilience is affected by many factors , for example resilience depends on the shape of the material , so it is NOT an intrinsic property of the material. The tensile test does not allow to measure the resilience , which can be only measured by impact tests . So resilience and toughness are both measurements of how much energy the material can absorb but the main difference is that the toughness is referred to tensile tests while resilience to impact tests ! Toughness can be defined trough resilience and is an intrinsic property of the material, while the resilience depends on the shape of the material so it isn’t an intrinsic property like toughness. International standards fix at 27 Joule the limit to consider a material resilient /tough , but optimal values are around 40 - 60 Joule. 3 Grain refinement is a mechanism which allow the material to dissipate energy before failure so the material becomes more tough even if it is fragile. 12 There are some limits in the measurement of the resilience of the materials, because the results of the impact tests depend on the shape of the specimen tested, because very tough material tend to become thin in the center of the specimen so the change of shape induce a different state of stress If the specimen changes its shape there is a three - axial state of deformation, while we have a plane (biaxial) state of strain if the material doesn’t change shape; in the second case the material can dissipate energy only in 2 directions instead of dissipating it in 3 d irections, so it is less tough. IMPACT TEST Impact tests are based on the m easurement of the energy absorbed by a specimen during an impact and is an index used to characterize the resilience of a material , but also the toughness since this property can be defined through resilience . The test uses a gravitational hammer which impacts on the specimen. The test is based on the energy conservation principle : the hammer which is like a pendulum transform gravitational energy in kinetic energy , which is then dissipated on the specimen during the impact. The difference between potential energy of the hammer in the initial condition and in the final condition gives us the entity of the energy dissipated on the specimen. The notch on top of the hammer can have different shapes depending on the regulations ( KCU , KCV , KIZ , …) used and the shape of the notch affects the result of the test. Charpy KCV notch Charpy KCU notch The result of the test is expressed in J oule because it represents the energy dissipated by the notch on the specimen. International standards fix at 27 Joule the limit to consider a material resilient /tough , but optimal values are around 40 - 60 Joule. Impact tests must be performed at a specified temperature and matching the specimen dimensions prescribed by the international standards in order to obtain valid results. The test must be repeated many times and the final results must be computed statistically . 13 The temperature at which the test take place must always be specified because at a certain temperature mat erials undergo a transformation from fragile to brittle behavior called transition temperature . We want materials to work above the transition temperature so that they show ductile behavior which is safer and avoid brittle failure, because Brittle beha vior is characterized by very fast crack propagation so it is a source of danger. K IC - Linear elastic fracture K IC test is a kind of test used to measure toughness based on energy considerations: the fracture is always featured by elasticity . The stress on the resisting surface in the specimen before fracture is not distributed homogeneously and it increases at the t ip of the crack ; a material can bear only a certain amplification factor K , which is the ratio between the stress in the material far from the crack and the stress at the tip of the crack (highest stress in the material before failure). By measuring the amplification factor that a material can stand we can have an index of the toughness of the material (idea on the basis of the K IC test) . As we can see the elastic stress distribution 9 $ assumes a constant value far away from the crack while approaching the crack the value of stress increases exponentially. The amplification factor K , which is the ratio between the stress in the material far from the crack and the stress at the tip of the crack (highest stress in the material before failure), can be computed by an experimental formula: ! = # ∙ √ ' ( ∙ ) * ( + , -./0 1) 2 ≥ 2 ∗ [Other information on K IC test in chapter I of the book] 14 ! = # ∙ √ ' ( ∙ ) * ( + , # : the stress induced in the material measured as if it was uniformely distributed and without the presence of the crack, so it is the ideal stress computed as F/A ; ( : width of the fracture [ : length of the fracture ) * " # , : functio n which depends the shape factors of the crack (width and length) The formula for the amplification factor K is valid only if the thickness of the specimen B is large enough, otherwise the measurements done to obtain the factors which appear in the K formula are not valid: 4-.5161-. : 2 ≥ 2 ∗ = 2 . 5 ; ! $ # % < & K Q is the amplification factor measured experimentally while K IC is the amplification factor computed with the mathematical formula. In fact K Q is equal to the K IC if the constraint on the thickness of the specimen B is verified, otherwise the amplification factor measured K Q is not equal to the one computed by the formula K IC ; this is why the formula can only be applied under this condition. The material is loaded with a cyclic laod in order to obtain the highest amplification factor possibl e, then a tensile load perpendicular to the fracture is applied and the displacement reached by the material until fracture is measured by an extensimeter. By representing the result on a force - displacement diagram it is possible to compute the stress # by finding the tangent of the curve, we lower it down by 5% and we obtain the value of the force in the point of rupture as intersection between the result curve and the line created (5% line). Then we can divide this force by the resist a nt area of the specimen in order to obtain the failure stress # which we can use to compute the amplification factor K . 15 FATIGUE Fatigue is a failure mechanism that happens when a material is sub jected to a cyclic load which causes in the material a stress lower to its yielding stress, so the failure happens even if the material is supposed to be in its safe zone (this is why the failure is unexpected) . The fail ure due to fatigue it’s typically brittle . When the material i s subjected to fatigue there appear some line in the region where the cracking begins; each of those traces indicate also the end of one cyclic loading and the beginning of the next loading. Woehler curve The Woehler diagram points out the fatigue limit , which is the maximum stress that can be applied to a material before the fatigue failure takes place (lower than the yield strength). Some material has a sharp transition between two zones , one in which the fatigue curv e decreases and one in which it remains constant, while for other materials the curve is smooth . For material with sharp transitions we use as fatigue strength limit the value of the stress which ensure no failure for 5 10 7 cycles , while for material characterized by smooth curves we take as endurance limit the stress which ensure no failure for at least 1 mil l ion cycles , which is also the value at which the curve cha nges and becomes constant. 16 Growth rate of fatigue crack The g rowth rate of the crack 8+ 8' is also a very useful parameter and it is a function of ∆ \ , which is the difference between the highest and the lowest amplification factor experienced by the material during the cyclic loading, and a function of m , which is an experimental coefficient. TC T@ = , ( ∆ N ) % ∆ N = N %+9 − N %,' = _ ∆ 9 √ aC The growth rate can be drawn as a function of ∆ N and we can identify different regions: D epending on the application it could be useful to increase the toughness or the strength of the material in order to avoid fatigue failure : j) W ith a strong material (high strength) we can avoid the nucleation of the crack 4 so we can also avoid fatigue and failure . Usually this strategy is preferred because we avoid failure in the first place but we cannot always achi ev e h igh strength ; k) W ith tough material we can have a safer failure, because the material fails in a ductile w a y , so even if it fails it is in a safer way wrt to the brittle failure due to fatigue , but if we have a tough material which is not very strong the nucleation of the crack takes place and we have to control the crack growth . This approach is called fail safe approach . 4 Nucleation is the first ste p in which the crack appears and starts to grow. If we avoid the nucleation we also avoid the crack propagation. 17 Practical safety rule The following experimental relation gives an approximation of the endurance stress we can achieve in a material that still avoids the fatigue phenomenon: = : = b 0 . 25 c - + - $ e + 50 f gMC Where - is the fracture strength of the material and - $ its yield strength S afety conditions Also the type of cyclic loading applied to the material can affect the fatigue phenomenon so it is useful to consider not only the amplitude of the load but also it s shape in order to study accurately the fatigue phenomenon. The parameters we have to take into account which descri b es completely the periodic loading cycle are: 1) h ;5< : maximum stress achieved during the cycle 2) h ;=> : minimum stress achieved during the cycle 3) h ; : average stress achieved during the cycle We can t hen draw a graph in which we can identify the safety region (inside the curve) and the dangerous region (outside the curve). If we are on the vertical axes we are considering a sinusoidal cycle with the same amplitude of compression and tension with a zero mean value, while if we move to the right or to the left we obtain a cyclic loading with a non - zero mean value of stress applied to the material. In the extreme of the graph the region becomes unsafety (dashed lines) because the loading on the materi al is no more cyclic but as we tend to the extremes the loading becomes a simple tension or compression test, in fact the two lines which delimitate the safety zone and represent the cyclic load converge. In order to void the fatigue failure we have to take into account also the surface quality of the material: low roughness of the material decreases the amplification factor on the surface and avoid the nucleation of a crack . This is very important in case of tors ional loading or bending to avoid fatigue, in which it is better to improve the surface quality than other properties to avoid failure Another very useful measure is to avoid sharp edges which amplificated the stresses and then also the amplification fact or . 18 1. FOUNDMENTAL OF WELDING METALLURGY Welding 5 metallurgy is the study of welding materials in order to avoid their failure and it is focused on the solidification process of materials , which is the transition from a liquid phase to a solid one. Localized melting of material and the following solidific a tion produces a rigid joining of different parts , so it is a very useful practice because it allows to produce separately the different components of a product, but the welding point can be the origin of fa t igue phenomenon , so it as to be anal y zed in order to avoid failure. So let’s start with the study of the solidification process. SOLIDIFICATION AND COOLING Pieces obtaine d by casting are pieces that come from a solidification of a liquid metal ; t he solidification can cause microstructure transformation which can lead to brittle behavior so it has to be analyzed in order to avoid failure. If we take a look at the phase diagram we can see that in t he region between the liquidus line and the solidus line the material contains both liquid portions and solid portions that coexist ; moreover a t a given temperature the chemical composition of the solid portion and the liquid portion are different, because the liquid portion is richer of alloy elements (liquidus line is translated to the right wrt the solidus line). 5 Welding = saldatura Casting = getto colato e solidificato Foundy shop = fonderia 19 Segregation Let’s consider a piece of material in which both solidus and liquidus coexist. When the solid portion is deformed it ejects alloying element to the liquid phase so the liquid phase undergo an enrichment of the alloying elements and when this part becomes s olid it will have an higher concentration of the alloying element th a n the one solidified before ; this phenomenon is called segregation . The chemical composition of the solidified material varies (increases) due to the s egregation phenomenon , that crea tes a n hetero geneous chemical composition in the material while solidifying and this causes also an h eterogen e ity in the mechanical properties of the different regions . ß In the first region there is a lower chemical concentration of the alloy element while in the next part, which solidifies after, the concentration is higher due to the elements ejected before by the solid part. Due to segregation the chemical composition of the material is not cons tant and equal to the nominal value but it varies in the material: depending on the actual value of the concentration we distinguish a negative segregation zone (if the %C is lower than the nominal value) and a positive segregation zone (if the %C is highe r than the nominal value) In the figure on the left we can distinguish different regions: a) Lines that separates the material to be joined (base material) characterized by grains with a certain orientation from the region in which the material was melted and has already solidified; b) Fully solidified material ; c) Solidifying material , which starts solidifying forming some tips (column); d) Liquid pool . Passing from a to d we can observe segregation , so we will observe different concentrations passing from the different regions. 20 The solidification microstructure is imposed by the orientation o f the crystals of the base material on which the columns are growing. This phenomenon is called epitaxial growth and it describes the fact that as the liquid material starts to solidify the direction of the forming crystals is the same as the direction of the grains of the base material on which the liquid solidifies. There is also a phenomenon called micro - segregation which is a particular type of the segregation which takes place between different grains at the microscopic scale, because some grain solidify first and eject alloying element, so the grains that will solidify later will have an higher concentration of th at element, therefore creating an heterogeneous chemical composition also at the microscopic level. Isothermal lines in the welding process During the welding the heat source (in the figure below represented by the point 0 ) moves along the material i n order to form a rigid joint. Around the heat source we can identify some isothermal lines which separates the region in which the material is still liquid, called liquid pool , and the region where the material has already solidified. The pool of liquid metal moves together with the heat source while the region outside the isothermal line of the solid part is not affected by the melting process (HAZ). Moreover t he shape of the welding waves gives information about the speed at which the welding in performe d. 21 G rowth of the grains The grains which grow when the material starts to solidify can have different characteristics depending on the cooling rate of the material. On the left we can observe the formation of grains in very slow solidification process while on the right the formation of grains in faster solidification process , which leads to column solidification . Those type of solidification refer to an equilibrium situation. Since the metals have high conductivity coefficients the transfer of heat inside the material is very fa st and the solidification happen in non - equilibrium conditions . This can lead to the d evelopment of lateral branches called dendrites , which formation is due to thermodynamic reasons: the heat is transferred from the liquid pool, which is at an higher temperature, to the solid portion, and in order to increase the rate of heat exchange and the rate of entropy variation the material solidifies building column in order to increase the surface which can exchange heat. 22 Heat affected zone (HAZ) There is another critical point that is a region adjacent to the solidified zone (welding zone) but does not belong to the region which was liquid and then solidified. This region can be brittle even if it doesn’t belong to the properly welding zone and it is called heat affected zone 6 : this zone does not melt but it undergo es a heat treatment because it reaches the temperature A 3 and it becomes austenite , but if the transformation is very fast we have also quenching which creates martensite , so it becomes a very brittle zone. T he conductivity of the metal is very high ther e fore also the adjacent zone is alterated by the welding and we have to study both the heat affected zone and the melted zone. So we distinguish 2 different zones: a) Molten solidified zone ; b) Heat affected zone surrounding the molten solidified zone , which undergoes a heat treatment ( quenching ), because in this region the temperature is so high that the material becomes austenite and then a very high cooling rate due to the high conductivity of the metal causes quenching and the formation of martensite. Fresh martensite is very brittle because it has not undergone a tempering procedure which make s it ductile , so this phenomenon has to be avoided. Cooling rate of HAZ and related microstructures If we consider the CCT curves of the transformation we can obtain the formation of very different microstructur es , such as: 1. Martensite , for very high cooling rate ; 2. Martensite surrounded by ferrite and iron carbides , for intermediate cooling rate ; 3. Ferrite , perlite and iron carbides , for low cooling rate , which is the most desired because it is the most ductile . 6 ZTA = zona termicamente alterata 23 (1) (2) (3) Martensite Martensite surrounded Ferrite, perlite by ferrite and iron carbides and iron carbides To avoid complete or partial quenching in the region surrounding the s ea m 7 (welding) which lead to formation of fresh brittle martensite we have to take some precautions. Hydrogen solubility in the different structural constituents The presence of hydrogen could cause some problem such as brittleness and fracture because the hydrogen can be mant a ined in solution in austenite but cannot be kept into martensite , therefore it can cause points of brittleness. The formation of the hydrogen ( H + H à H 2 ) causes internal stresses because the volume of a molecul e of hydrogen H 2 is very different from the volume of two separated atoms of hydrogen H summated, therefore the volume difference can cause the nucleation of a fracture in the steel . 7 Seam = saldatura 24 The hydrogen migrates in the part of the material which contains aust enite because the gamma phase can solve hydrogen , so the concentration of hydrogen in the gamma phase increases , while martensite cannot solve hydrogen . This high concentration of hydrogen accelerate s the formation of the molecul e of H 2 ( H + H à H 2 ) , which can lead to the formation of fracture due to its higher volume When the austenite transforms into martensite or ferrite the hydrogen cannot be solved anymore in the phase . This phenomenon is dangerous especially in the steels with a high concentration of alloy ing elements , because the higher the concentration of alloying element the easier the formation of austenite, which absorb H and lead to the formation of H 2 . In steel with a lower concentration of alloying element s the formation of martensite is possible only for very very high cooling rate (dashed lines), while steel with a high concentration of alloying element the CCT curve is shifted in the right - lower region of the diagram , so the martensite can form also for low cooling rate (continuous lines). Therefore for steel s with a high concentration of alloying element we have to take into account the cooling rate in order to avoid the formation of martensite . 25 Equivalent carbon parameters There are some parameters which allow us to prevent the formation of martensite in steels with a high concentration of alloying element depending on the concentrations of alloying elements: we define the carbon equivalent CE as the sum of the actual concentration of carbon C in the steel and the concentrations of the others alloying elements multiplied by different coefficients depending on the importance of the element considered wrt the martensite formation. To compute the carbon equivalent CE we use t wo differ e nt expressions valid for different value of carbon concentrations C; t he other factors in the formulas are the concentrations of the other alloying elements in the steel and indicate when we have to implement procedures in order to slow the cooling rate of the steel to avoid martensite : Yi , > 0 . 16% Yi , < 0 . 16% CE = carbon equivalent C = carbon concentration In order to avoid the formation of martensite we act in different ways depending on the CE value: a) If #j > % . k we can weld steel without following a specific procedure to decrease the cooling rate because martensite does not form ; b) If % . k < #j < % . l a pre - heating has to be applied to the base material on which the welding is performed in order to lower the heat rate extraction. The pre - heating is in the order of 100 - 250°C ; c) If #j > % . l both a pre - heating and a post - heating ha ve to be performed on the based material, so the temperature has to be maintained in the range 100 - 250°C to avoid the formation of martensite during welding . IRSID diagram There are also some rules that prescribe which pre - and post - heating temperature has to be applied on the base material if ,A > 0 . 4 ; such rules are described by a diagram defined by IRSID from which we can find out the pre - and post - heating temperature to be applied to the material in order to avoid the formation of martensite given the different informations we know about the material and the process (thickness, electrical energy used, angle of joint, welding technology, …). The IRSID diagram has the following characteristics: a) On the vertical axis the graph shows the thickness of the base material for the welding [mm] ; b) The circles on the lines in the first quadrant indicate the interval of time which is used to decrease the temperature from 700°C to 300°C; this time interval is important because martensite usually forms between 380 - 320°C so this is a good index for the martensite formation, in fact the longer this interval the lower is the possibility of formation of martensite, which forms only for high cooling rate ; 26 c) O n the left horizontal axis the graph shows the electrical energy needed to perform the welding 8 ; d) The diagonal line in the IV quadrant (down left) represents the angle j formed between the two heads of the materials that have to be joined by the welding proces s . In case the joining is not head - to - head , so we do not have angle to compute , we refer to the thickness a and s of the two pieces (edge welding) ; e) The diagonal line in the III quadrant (down right) represents different welding technologies that can be us ed . 8 The welding is performed by the formation of an electric arc between the electrodes (belonging to the tool used) and the material; the power generated by the electric arc is the product between the tension featuring the arc U and the current through the arc I, while the electrical energy is obtained by dividing the power by the speed v at which the electron are moved along the material during the welding ( P = U I , E = P/v ) . 27 The welding technologies that can be used are : a) TIG ( = Tungsten Inert Gas ) , which means the elect rodes are made of tungsten and are not consumable, so the electrode melts a n external metal road that is placed between the base material and the electrode; “inert” because the atmosphere around the welding zone is saturated with argon in order to avoid th e oxidation of the melted metal. It is used especially for high performance components ; b) MIG ( = Metal Inert Gas ) , which is instead a consumable wire used as an electrode and it is used especially in big manifacturies 9 ; c) Electrode enrobe sous flux , which uses an electrode coated with ceramic materials and it is used especially when someone has to work outside the shop . How to use the IRSID diagram: In order to know which type of pre - and post - heating we have to perform on the material we follow this procedure: 1. First we individuate on the vertical axis the thickness of the material and we draw a horizontal line in correspondence of the thickness value, so that we intersect all the time - interval lines; 2. Then we go to the horizontal axis on the left and we draw a vertical line downwards in correspondence to the value of electrical energy used in the process; 3. We intersect the vertical line drawn with the diagonal line which correspond to the geometry of th e welding performed (angle of joining or thickness of the parts joined); 4. From the point of intersection on the diagonal line we draw a horizontal line which intersect the diagonal on the III quadrant, in correspondence of the line which represents the wel ding technology used; 5. From the point of intersection found we draw a vertical line which intersect the horizontal line in correspondence of the thickness of the welding performed; the closest time - interval line to this new point of intersection is the ti me - interval which is needed by the material to decrease the temperature from 700°C to 300°C and it is an index of the possibility of martensite formation; 6. If the time - interval found is too small, we have to apply a pre - /post - heating to avoid martensite formation; to discover the temperature of the treatment we chose the time - interval we want the material to have during the cooling 10 , we intersect it with the horizontal line in correspondence of the thickness of the welding performed and from the point of intersection we draw a vertical line downwards; we intersect the two vertical line in the first quadrant with the diagonal pattern below the axis and we find a point of intersection ; 7. From the point of intersection on the diagonal pattern we project the point on the axis which represents the pre - /post - heating temperature , so we finally find the information we wanted. 9 So the big difference between the technologies is that in TIG the electrode is not consumable while in MIG the electrode is a wire and is consumable; the wire melts and fill the gap between the pieces to be joined . 10 A good time interval that can be chosen is between 30 s and 100 s to be sure to be in a safety region (l ast 4 curves). 28 The welding is critical as the thickness of the base material increases due to the very good thermal conductivity of steel, because the higher the thickness, the worse is the possible formation of martensite. Example: In this example the time - interval chosen for the cooling is 15 seconds and the pre - /post - heating temperature that has to be applied is 800°C 29 DESIGN OF THE WELD JOINING 11 Classification s of welding types The classification of weldings could be performed following many principles. Classification as a function of the position where the joining is performed : 1) Flat welding 2) Horizontal welding 3) Vertical welding 4) Overhead welding : difficult because the melting material can drop down due to gravity Classification as a function of the relative position of the pieces joined : 1) Head - to - head welding : the two faces of the pieces are joined (angle of 180°) 2) Edge coupling ( edge welding ): formation of angle that are not 180° Classification as a function of the type of penetration between the pieces joined: 1) Fully penetrated : complete filling of the gap between the metals; this is the safest condition but it cannot always be achieved easily (for example because not all the gap is accessible and there is no room enough to perform the welding overall). Fully penetration is necessary for high safety requiring conditions ; 2) Partially penetrated : uncomplete filling of the gap betwee n the metals, which leave a notch on the surface of the material which can lead to the nucleation of a crack because it is a very critical point . 11 Not all the slide s (3C) on welding design are to study, only the first and the last part 30 DESIGN OF THE COUPLING We have to study a resistance criterion in order to define the size of a welding s ea m 12 that has to be defined as a function of the material to be joint. The stresses into the material can be represented by 3 type of stresses orthogonal to each other, which can be studied with 2 different approaches: a) Real neck section appro ach : The f irst approach is based on the identification of the resisting section as the neck section of the welding seam , which is the section underlined in the figure below and is the section corresponding to the center of the seam: The neck section is identified in different ways depending on the type of welding seam To define the size of the seam we have to identify the stresses on the neck section ( # , > ) and then we confront them with the defined safety parameters using the formulas prescribed by the standards : ⎩ ⎪ ⎨ ⎪ ⎧ C # ' & + 3 ( > / / & + > ' & ) ≤ # ) I J * & # ' ≤ 0 , 9 # ) J * & Resisting criterion for the real neck section approach 9 ? = yield strength of the material (= resistance) _ , ! @ A = safety parameter s ! @ A = 1 . 25 The factor 3 which multiply the shear stresses means that the shear stresses are more severe than normal stresses so they have to weight more into the relation. This criterion is easy to applicate for head - to - head weldings while for edge weldings the overturned neck section approach is easier to use. 12 Seam = saldatura 31 _ varies mainly as a function of the type of steel used and it becomes more severe as the yiel d strength of the steel increases (up to _ = 1 ) 13 b) Overturned section approach : T he second approach is based on the identification of the overturned neck section, which is obtained rotating the neck section on one side of the seam 14 (not projection!) and it is the approach used for edge weldings : As we did for the real neck section approach we define the stresses imposed on this new section ( . , 6 ) 15 and we confront them with the defined safety parameters using the formulas prescribed by the standards : M C . ' & + 6 / / & + 6 ' & ≤ I , # %- | . ' | + | 6 ' | ≤ I & # %- Resisting criterion for the overturned neck section approach # %- = # ) = yield strength of the material (= resistance) I , , I & = safety parameters which depends on the type of steel considered 13 It is more severe if beta increases because it is at the denominator of the formula so the safety condition becomes more restrictive. 14 Rotation, NOT projection! 15 Notations: ( ! , # ) are referred to the real neck section approach while (n,t) to the overturned neck section approach 32 N.B. Classification of the steels (alphanumerical coding): Letter which represents the type of application of the steel + number which represents the yielding stress of the steel in MPa. Letters: S stands for steel certified for structural application (the S DOESN’T stand for steel but for the application), E stands for mechanical application (E=engineering) while L stands for st eel certified for pipes constructions) Example: S 325 H = steel for structural application with yield strength of 325 MPa. GENERAL BEHVIOR OF PARALLEL AND PERPENDICULAR SEAMS Depending on the orientation of the seam its behavior changes when a load is applied: a longitudinal seam parallel to the load has a more ductile behavior to a seam perpendicular to the load, so the first case is preferred because it avoids brittle failure. This rule is general and valid for every type of welding (head - to - head, edge, …) DESIGN CONSTRAINT FOR EDGE COUPLING In edge coupling in order to have a fully penetrated welding we have to perform two seams, one on each side. There are to constraints to be observed for edge welding : a) The sum of the length of the seam s in contact with the vertical component must be higher than the thickness of t he vertical component ; b) The distance between the two sea m s must be lower than the minimum value between the thickness of the vertical component divided by 5 or 3 mm. 33 DESIGN OF OVERLAPPED WELDING Overlapped plates can be jointed in many different ways (the seams can be displaced in many positions) and every combination will have different resistance properties depending on the type of seams performed (longitudinal or perpendicular) . [ See behavior of every different type of welding on file “3B) Welding design” ] Exercise s on welding : See exercise notebook 16 [ See also exercises on welding - file “3C) Welding design ONLY EXERCICES” ] 16 Similar to the exercise in the exam 34 2. SLIDING SYSTEMS – TWINNING – INVERSE POLAR FIGURE – EULER ANGLE THE DISLOCATIONS The boundary between plastic behavior and elastic behavior on the macroscopic scale can be pointed out with tensile test by the changing of behavior in the stress - strain graph, but on the microstructural scale it is due to an irreversible movement of the d islocations , which signal the metal have reached the yielding point . So when we talk about dislocations we are in the plastic deformation area. Dislocations are atomic planes or single atoms present in excess or not in the exact position wrt to the id eal geometrical structure of the crystal ; d islocations can be detected by TEM ( T ransmission E lectron M icroscop e ) and they are very important for the study of the properties of the material because w ithout dislocations the plastic behavior could not be poss ible because the strength to be applied to displace a portion of a crystal would be too high, so the movement between the different portions of grains is only possible trough movement of dislocations , which instead require lower energy. In correspondence to dislocations the load to be applied to move a portion of the crystal is much lower wrt to a portion in which no dislocation is present (perfect structure) . Dislocations are also very important because strengthening mechanism done to improve the strength of the material are based on the attempt to stop the dislocations on the material or to free the movement of the dislocations by removing any obstacles they could meet. We can have different type of movement of dislocations, for example they can climb several plane of the crystals or they can slide (glide) along the material. The dislocations move until the portion which surround the dislocation are completely displaced and the alignment of the atomic plane becomes again simi lar to the geometrical one. Analogy between caterpillar and dislocation motion The dislocation has two properties, a line direction , which is the direction running along the bottom of the extra half plane, and the Burgers vector which describes the magni tude and direction of distortion to the lattice. 35 Moreover we can define different types of dislocations: a) Edge dislocation : localized lattice 17 where an extra half - plane of atoms is introduced midway through the crystal, distorting nearby planes of atoms ; w hen enough force is applied from one side of the crystal structure, this extra plane passes through planes of atoms breaking and joining bonds with them until it reaches the grain boundary . In an edge dislocation, the Burgers vector is perpendicula r to the line direction ; b) S crew dislocation : topological defect of a crystal lattice where the lattice plane shifts by one layer (or more layers), like a spiral staircase. The Burgers vector of a screw dislocation is parallel to the dislocation line ; c) Mixed dislocation : dislocations which have both edge and screw components. Dislocation can move only along preferential plane and directions in the crystal, which are the plane and directions with the highest atomic density . However t his causes anisotropy in the material because there are some areas along which dislocations cannot move therefore some parts of the material will behave differe ntly when a load is applied. Anisotropy is the different beh a vior of a material depending on the type of loading applied (direction of the load) . Dislocation are featured with mathematical symbols such as: - ^ or T - ® or ¬ - … When 2 dislocations featured by opp o site mathematical symbols meet they annihilate and we obtain again a perfect crystal . 17 Lattice = reticolo 36 TYPE OF ELEMENTARY CELL S When we deal with metal alloys we can have 3 main crystal kinds: 1. FCC (Face Centered Cube) : the crystal has a unique cell featured by atoms located at the vertexes of a cube and at the center of each face; repeating periodically this unit cell in the 3 direction of the space we can obtain the complete crystal. 2. BCC (Body Centered Cube) : cubic cell featured by atoms at the vertexes of the cube and one atom in the center of the cube. 3. HCP (Hexagonal Compact elementary cell that is a hexagonal pris m Prismatic cell) : with atoms on the vertexes, atoms in the center of the upper and lower basis and on an intermediate plane with atoms disposed on a triangle. SLIDING SYSTEMS In order to identify the direction and the plane where the dislocation can move, which is called sliding system , we use the following criterion: t he directions and planes where the dislocations can slide are featured by the highest atomic density in the cr ystal, so we can identify them as a function of their atomic density . Sliding system = plane where the dislocation can move + direction of movement IDENTIFICATION OF PLANE AND DIRECTIONS OF THE SLIDING SYSTEMS Each of the different type of cell a metal alloy can present is characterized by different sliding systems, which can however be described with the same notation ( Miller indexes ) obtained thorough mathematical computations based on the geometry of the crystal represented by the coordin ates in the space . First of all to identify planes and direction we have to fix a reference system ; for FCC and BBC cells we need a reference system with 3 ax e s (a,b,c) along 3 edges of the cube, which origin coincides with one of the vertex of the cube . Instead for HCP cell we need a different reference system, made of 4 ax e s (3 on the basis and one along the height). 37 In order to identify a sliding system we need a plane and a direction on that plane, along which the dislocation can slide 18 : a) Notation for planes: the plane on which a dislocation moves can be identified by writing the coordinates of the normal unit vector to the plane, but if we use a cartesian reference system with 3 axes the normal vector can be also written by finding the po ints in which the plane intersects the axes of the reference systems;