Twenty-five single beads are deposited over mild steel substrate (as shown in Fig. 4) at different parameter combinations as single beads have maximum influence over the geometrical accuracy of multilayered deposits.
The interrelation between comparative and reference sequences becomes stronger when a high grey relational grade value is achieved by the experiment set. Therefore from Table 4 and Fig. 3(a), it has been recognised that experiment number-8 is the optimised parameter set. Again to determine the significance of different deposition parameter level values on entire response characteristics and their ranking, a response table (Table 5) and graph (Fig. 3(b)) has been prepared. Response table and graph is prepared by taking the average of grey relational grade values conforming to each parameter level. The maximum grey grade value for each process parameter level represents the optimal condition at which better control over response characteristics is achieved. As per the above explanation, ‘A2, B5, C1, D5’ is the best parametric condition and at this setting a confirmation test (as shown in Table 6) is performed and its response characteristics are compared as with experiment number-8. It also conforms from Table 5 that WFS has maximum control over bead geometry and droplet transfer behaviour followed by welding voltage, TS and current. Therefore WFS is termed as a key deposition parameter during multi-criteria analysis.
Table 6
Confirmation test at ‘A2, B5, C1, and D5’parametric combination
Symbol | Deposition Parameters | Deposition Characteristics |
A | Welding voltage (U) = 19.0 V | Bead width deviation (Wd) = 0.1849 mm |
B | Welding current (I) = 85 A | Height deviation (Hd) = 0.15 mm |
C | Travel speed (TS) = 0.048 m/min | Droplet diameter (D) = 2.61 mm |
D | Wire feed speed (WFS) = 4.5 m/min | Droplet detachment frequency (fn) = 13.6 Hz |
Better performance characteristics are obtained from the confirmation test as compared to initial experiment (experiment number-8). It is due to the enhancement in welding current value, which improves heat generation at the arcing point that leads to proper melting of wire electrode and generates less droplet diameter with higher detachment frequency. Similarly, under low TS, droplets are properly deposited on the substrate surface and give rise to low bead width and height deviation. The process of droplet transfer at ‘A2, B5, C1, and D5’parametric combination is depicted in Fig. 5 (a-f). It describes all phases of droplet transfer like its generation (Fig. 5 (a)), path of movement (Fig. 5 (b-e)), and final deposit (Fig. 5 (f)) on the metal surface. Now at ‘A2, B5, C1, and D5’parametric combination single (Fig. 6 (a)) and multilayered deposits (Fig. 6 (b)) are prepared over SiO2 substrate for ease in part removal after complete deposition.
4.1. Parametric influence on bead width deviation
The influence and percentage contribution of deposition parameters on a particular response attribute is determined from the ANOVA analysis (as shown in Table 7). For a definite response characteristic, the percentage contribution of deposition parameters is determined by taking the ratio between Adj. SS. values for the deposition parameters to the total Adj. SS. conform to that response attribute.
Table 7
Adj. SS. values of deposition parameters for different deposition characteristics
Deposition parameter | Adj SS for Wd | Adj SS for Hd | Adj. SS for D | Adj. SS for fn |
Welding voltage (U) | 14.0795 | 2.2836 | 1.2517 | 9.344 |
Welding current (I) | 0.1763 | 1.4526 | 5.3791 | 1.216 |
Travel speed (TS) | 1.6997 | 0.3172 | 0.9388 | 4.272 |
Wire feed speed (WFS) | 3.4208 | 0.5573 | 16.8746 | 333.952 |
Total SS | 19.3763 | 4.6107 | 24.4442 | 348.784 |
It has been recognised that bead width deviation is highly controlled by welding voltage followed by WFS, TS, and current. To study the variation of bead width deviation with deposition parameters, few graphical representations are prepared (as shown in Fig. 7 and Fig. 8). It has been observed that under high welding voltage, the bead width deviation increases. Thermal energy generated at the arcing zone depends upon voltage drop and current density [2, 13, 30, 31]. Current flow through electrons and ions is constant, so heat generation varies with respect to the voltage drop at two electrodes [30, 31]. As reported by Cao et al. [13], under high voltage, thermal energy generation becomes more which enhances the melting rate and droplet detachment. It also aggravates the plasma force and generates an unstable plasma jet. Under high voltage, arc length between two electrode increases. Therefore droplet forms at wire tip is highly oscillating between the inter-electrode gaps due to unstable stable plasma jet. So the droplets continuously come in contact with the tungsten electrode and give rise to high tungsten diffusion and wear (as shown in Fig. 9 (a-h)). In such a case percentage of tungsten in the deposits increases. When tungsten starts to diffuse, the arc plasma starts oscillating from left to right at a high velocity; therefore, no consistency was achieved in the droplet transfer process [24–27]. This results in poor bead uniformity (high deviation).
Similarly, under a higher value of WFS, the deviation in bead width was found to be reduced. High WFS raises the arc current, which enhances the heat input and magnitude of electromagnetic pinch force at the wire tip [2, 12, 14, 32]. Therefore molten droplets (of small size) are detached at a faster rate rather than oscillating at wire tip. Therefore the chance of contact of molten droplet at tungsten side reduces that results uniform bead. So WFS must be set to higher values at high welding voltage for producing fine and uniform beads. TS symbolises the speed at which worktable moves. TS basically maintains heat input and deposition per unit length [33–37]. For a definite value of welding voltage, current and WFS, the droplet detachment frequency is constant. So when table moves faster (TS is more), then the droplets do not adhere properly to the substrate and gives rise to poor bead uniformity and reduces the geometrical accuracy of the form part [25].
4.2. Parametric influence on bead height deviation
ANOVA analysis (from Table 7) convey that welding voltage and current are the most crucial deposition parameters that affect the difference in bead height between start and exit point of arc. To study the variation of bead height difference with deposition parameters, few graphical representations are prepared (see Fig. 7 and Fig. 10). It has been recognised that under the high value of welding voltage, height deviation increases due to unstable metal transfer results from unstable plasma jet [24–27]. Heat generation at arcing zone is determined from the product of welding voltage and current [2, 12, 13, 30, 31]. Therefore at higher current and voltage, the heat generation is aggravated [38–41], also heat content of the plasma jet increases [25]. As the process goes on, temperature of substrate increases and results in a reduction of thermal gradient, which retains the deposits in molten state for longer duration of time due to reduction in viscosity of molten pool under high temperature [2]. Therefore the flowability of molten metal is enhanced that gives rise to high bead width at low height towards exit point of arc as a result adequate difference in bead height is noticed [2, 6, 24–27, 41].
It has been noticed from Fig. 7 that height deviation is raised with a rise in TS, but it again reduces after a certain limit of TS [5, 6, 12, 14]. Under the high value of current and voltage, the plasma jet temperature increases [25]. The interaction time between plasma jet and substrate reduces upon increase in TS that reduces the heat input [26, 33–39, 42]. So the substrate temperature is not increasing so rapidly, which enhances the temperature gradient and promotes faster solidification [12]. Therefore the deposits are solidified quickly rather than free-flowing, which gives rise to a reduction in height difference between start and exit point of the arc. Under high WFS, an excessive amount of wire get melted and deposited over the substrate, so the chance of difference in the height of bead reduces [43, 44]. In general, better control over height deviation is achieved in the NTA approach compared to GMAW based WAAM.
4.3. Parametric influence on droplet diameter
Droplet diameter is highly controlled by WFS followed by welding current and voltage (observed from Table 7). The dependence of droplet diameter with deposition parameters is presented in Fig. 11–12. It is noticed that under the low magnitude of welding current, the droplet diameter is increasing. The magnitude of electromagnetic pinch force is directly related to the squared value of current [2, 24–27, 32]. So under a low level of current, the magnitude of pinch force is curtailed, and surface tension force is increased [45]. Therefore instead of detachment, its diameter continuously increases. But due to the increase in droplet diameter and reduction in surface tension force, the droplet starts oscillating under the action of buoyancy and promotes the arc between droplet and tungsten electrode. So the arc and plasma jet starts oscillating due to the oscillation of droplet. As a result, consistency in droplet transfer was not achieved. Therefore droplet comes in frequent contact with tungsten electrode at high droplet diameter, which aggravates the tungsten diffusion and wears. The adhesion of droplets on tungsten electrode raises the iron content on tungsten side, which is evident from the EDS analysis (see Fig. 13). The EDS report of tungsten electrode (wt% of different elements) after depositing a bead under a low level of current is presented in Table 8.
A sharp fall in droplet diameter is noticed upon increasing the WFS. High WFS promotes a high flow of arc current between two electrodes, which generates high thermal energies at arcing zone [2, 12, 14]. That helps in generation of concentric plasma jet (as shown in Fig. 14) and effective melting of wire electrode. Therefore droplet diameter reduces at high temeperature [25]. Also at high temperature, the surface tension force decreases [45]. As a result, the droplet is not able to retain at the wire tip for a longer duration of time. Therefore droplet detachment occurred at a faster rate with a reduction in diameter.
Table 8
Wt% of elements present on tungsten electrode surface at a low level of welding current
Element | Fe | W | Th |
Wt% | 48.49 | 50.97 | 0.54 |
4.4. Parametric influence on droplet detachment frequency
ANOVA analysis (from Table 7) convey that WFS is the key deposition parameter that affects the droplet detachment frequency most, whereas other parameters have negligible impact. High WFS conduct maximum arc current that produces high amount of heat energy at arc spot [2, 3, 14]. It also enhances the magnitude of electromagnetic pinch force at wire tip [2, 32]. Electromagnetic force pinches the droplet at wire tip that helps in detachment of droplets. Also, due to high-temperature generation, the surface tension force is diminished at the wire electrode tip that prevents the long duration adherence of droplet at wire tip and promoting faster detachment [45]. Therefore under high WFS, droplet detachment frequency is enhanced (as shown in Fig. 11 and Fig. 15). Arc length between tungsten and wire electrode depends upon the value of welding voltage. Small voltage results a small confined arc length. Therefore there is less space available for the growth of droplets between two electrodes. So the droplets continuously come in contact with the tungsten electrode and leads to a reduction in droplet detachment frequency and aggravates the tungsten wear (as shown in Fig. 9). But upon increasing the welding voltage, space available between two electrodes increases, which promotes free detachment of droplets without colliding with tungsten electrode. But after a certain limit with the increase in welding voltage, the droplet transfer rate reduces. As very high welding voltage generates an unstable arc that promotes high plasma oscillation, in such case, droplet again starts oscillating and makes frequent contact with tungsten electrode. That reduces the droplet transfer rate with increased chances of tungsten wear.
There also exist another factor (argon gas flow rate) that have some effects on droplet transfer rate. To study the impact of gas flow rate on droplet transfer frequency, depositions are made under different gas flow rate settings, but other parameters remain same as with the confirmation test. It is noticed from Fig. 16 that with a hike in gas flow rate, the droplet transfer rate increases. As it again enhances the plasma temperature, that leads to higher melting of wire electrode. As a result, droplet diameter reduces with high detachment frequency [24]. Under a very high gas flow rate (22–26 l/min), the plasma jet velocity increased rapidly that makes unstable droplet transfer due to the reduction in droplet size. In absence of shielding gas, the tungsten electrode wear and diffusion is aggravated. Also, powders of yellowish in colour are formed on the tungsten electrode surface under the absence of shielding gas, as oxygen reacts with tungsten and forms tungsten oxide, which is determined under EDS analysis (as shown in Fig. 17). During some practices, a 2.4 mm tungsten electrode is also used for studying the influence of electrode diameter on droplet transfer behaviour. It is observed that when the gas flow rate is more than 22 l/min, in such case tungsten electrode starts to burn. But at a gas flow rate of 26 l/min, the tungsten electrode bends after some time due to higher localised heat concentration under high plasma temperature. Application of argon gas for shielding also assists in maintaining the geometrical accuracy of the fabricated part [46].
4.5. Microstructural analysis
Microstructural analysis is performed on the multilayered deposited over SiO2 substrate. After deposition, small blocks are cut from the top, bottom, and interface between two adjacent beads through wire electrical discharge machining (WEDM). Specimen preparation is the most crucial aspect for microstructural analysis. All three samples undergo grinding (using a belt grinder) to produce a flat surface. Then polished with emery paper (1/0, 2/0, 3/0, and 4/0) and Al2O3 paste (through cloth polishing) to generate a scratch-free mirror-like surface. Finally, etchant (nital solution (4% HNO3 and 96% C2H5OH)) is used to remove the non-metallic inclusions and impurities present on the sample surface and to reveal phases and grain boundary. Now the specimens are observed under a Scanning electron microscope (SEM) to identify the different phases. Microstructure of the top, bottom, and interface zone is depicted in Fig. 18(a-f). The enlarged view of microstructure is presented in Fig. 18 (d-f).
In all the micrographs, complete ferrite colonies are observed with few pearlite phases (volume fraction is low with respect to ferrite phase). The ferrite that formed is known as proeutectoid ferrite as it forms before the starting of eutectoid reaction. It is commonly observed in hypo-eutectoid steel. Similar micrographs are obtained during GMAW based WAAM using ER70S6 wire [18, 47, 48]. In Top surface grains (see Fig. 18(a)) are slightly bigger than the bottom surface (see Fig. 18(b)); this could be due to the difference in the rate of cooling between top and bottom surface of the deposits that occurred due to reduced heat dissipation rate in top surface of the deposit with increased number of layers [41]. With respect to top and bottom surfaces, the grain size is quite smaller in the intermediate zone between two layers (as shown in Fig. 18(c)). During depositing a layer, some amount of heat is being transferred to the previously solidified layer [49]. So the previous deposit again undergoes recrystallisation and forms smaller and fine grains as compared to top and bottom layer. Fine pearlites are observed along with coarse pearlite in the top surface and interface layers. Whereas only coarse pearlites are observed in the bottom surface of the deposit. A reduction in size and enhancement in fineness of pearlite is noticed with increase in layer thickness (with increase in number of layers). This could be due to the reduction in cooling rate in the top and interface which enhances the growth and nucleation rate of pearlite [50]. Also under low cooling rate, the deposits remain at higher temperature for longer duration of time as a result the inter-lamellar spacing between Fe3C lamellae reduces, which gives rise to the development of fine pearlites [50].
After deposition of a layer, a small droplet is solidified at the wire tip, whose microstructure is depicted in Fig. 19. It also exhibits a similar microstructure as with the deposits. Therefore it can be concluded that the type of microstructure shown in the final fabricated part is completely depends upon cooling characteristics and temperature gradient maintained during the deposition process [51]. The elemental composition at wire, top, bottom, and interface layer is identified through EDS analysis. The element identified on wire, top, bottom, and interface layer is depicted in Fig. 20 and Table 9. It has been observed that iron content is maximum in the deposits; also composition of the deposit is almost same in top, bottom, interface zone and holds good agreement with the wire (ER70S-6) composition.
Table 9
Chemical composition of wire and deposits
Sample | | Wt% of elements |
Si | P | S | V | Ti | Cr | Mn | Fe | Ni | Cu | Zr | W |
ER70S6 | 1.19 | 0.00 | 0.03 | | 0.00 | 0.05 | 1.73 | 96.73 | | 0.16 | 0.10 | |
Top | 0.36 | | 0.11 | | | 0.22 | 0.99 | 97.67 | 0.31 | 0.24 | | 0.10 |
Bottom | 0.33 | 0.07 | 0.02 | | 0.04 | 0.04 | 1.06 | 97.63 | 0.16 | 0.22 | | 0.43 |
Interface | 0.55 | | | 0.06 | 0.05 | | 0.94 | 97.67 | 0.56 | | | 0.17 |
4.6. XRD analysis
XRD analysis is conducted through BRUKER D8-Advance®, which uses Cobalt (K-α radiation wavelength of 0.179 nm) as an X-ray source and a Lynxeye detector. The voltage (35 KV) and current (25 mA) are kept constant during all measurements. The two-theta (2θ) angle range is selected from 30°-126°. At the same time, scan rate and step size are selected as 5°/min and 0.02, respectively. The XRD raw data is analysed through X’Pert High Score software to identify the diffraction planes. Diffraction patterns observed on wire, top and bottom surface is depicted in Fig. 21. The identified diffractions planes are {110}, {200}, {211}, and {220} at 2θ angle of 52.3°, 77.2°, 99.6°, and 123.9° respectively as per the JCPDS pattern [06-0696] [52]. Similar results are also reported in previous studies [18, 47, 48]. In all samples, ferrite colonies are observed. But in wire, few copper peaks like {111} and {200} are identified at 50.3° and 59.2° respectively, which conforms to the JCPDS pattern [03-1005] [53]. Copper is basically shown due to the presence of copper coating on filler wire. But their intensities are very low as compared to the intensities of ferrite peaks. But these copper peaks are not presents in the deposited samples, which indicates that the copper coating is evaporated during the melting of wire under the high arc temperature. Instead of scherrer equation, the Williamson hall (W-H) plot is adopted for computing crystallite size and lattice strain, as broadening of diffraction peaks (ΔK, nm− 1) is defined (EQs. 1) using the effect of both crystallite size and lattice strain [54]. Here a scatter plot is prepared between 4Sin(θ) (in X-coordinates) and aCos(θ) (in Y-coordinates), and linear fitting is applied to it as shown in Fig. 22(a-c). Crystallite size is determined from EQs. 2, and slope of the plot represents the lattice strain. Crystallite size and lattice strain for all samples is presented in Table 10. The copper peaks present in wire sample are not considered during evaluation of crystallite size and lattice strain, due to their very low intensities compared to ferrite peaks.
$$\Delta K=\frac{{0.9}}{D}+\Delta \mathop K\nolimits_{D}$$
1
$$D=\frac{{K \times \lambda }}{I}$$
2
It has been observed from Table 10 that there is less variation in crystallite size between top and bottom surface of the deposits as compared to wire, which indicates a reduction in anisotropy. It is attributed to grain refinement in the bottom surface due to a relatively higher rate of cooling in bottom surface with respect to top surface. The lattice strain in both top and bottom layer is compressive in nature.
4.6.1. Dislocation density measurement
Dislocation density is also computed through X-ray diffraction method. It is computed using a modified Williamson hall plot [55–58]. In modified W-H plot, K is substituted by KC1/2 as per EQs. 3–5 [57, 58]. Here a linear fit plot is prepared between ∆K and KC1/2 (Fig. 23(a-c)), and the value of slope is determined to compute the dislocation density. The computed dislocation density for all samples is presented in Table 10. The graphical variation of crystallite size, lattice strain and dislocation density for all three samples is presented in Fig. 24.
$$\Delta K \cong \frac{{0.9}}{D}+\left[ {\left( {\frac{{\pi \mathop M\nolimits^{2} \mathop b\nolimits^{2} }}{2}} \right) \times \mathop \rho \nolimits^{{\frac{1}{2}}} \times \left( {K\mathop C\nolimits^{{\frac{1}{2}}} } \right)} \right] \pm \left[ {O \times \mathop {\left( {K\mathop C\nolimits^{{\frac{1}{2}}} } \right)}\nolimits^{2} } \right]$$
3
Where, \(\Delta K=\frac{{a \times \cos \left( \theta \right)}}{\lambda }\)and \(K=\frac{{2\sin \left( \theta \right)}}{\lambda }\) (4)
By neglecting the higher terms \(\left[ {O \times \mathop {\left( {K\mathop C\nolimits^{{\frac{1}{2}}} } \right)}\nolimits^{2} } \right]\) EQs. (2) becomes
$$\rho =\frac{{2 \times \mathop \beta \nolimits^{2} }}{{\pi \mathop M\nolimits^{2} \mathop b\nolimits^{2} }}$$
5
Where,
θ | Bragg’s diffraction angle | b | Burger’s vector (0.2485×10− 9 m) |
K | Scherer’s constant | ρ | Dislocation density (1/m2) |
λ | Wavelength of the X-ray source | C | Contrast factor |
I | Intercept of the linear fit plot between 4Sin(θ) and aCos(θ) | O | Higher-order component of KC1/2 |
D | Crystallite size | a | FWHM (full width half maximum) |
M | Material constant varies between 1 and 2 M = 1 represents no deformation in the material M = 2 represents a material with deformations | β | Slope of linear fit plot between ∆K and KC1/2 |
Table 10
Crystallite size, lattice strain, and dislocation density of samples
Sample | Crystallite size (nm) | Lattice strain | Dislocation density (1/m2) |
Wire (ER70S-6) | 23.04931507 | 0.000446959 | 3.2663×10^13 |
Top Surface | 18.38907104 | -8.21E-04 | 3.8992×10^12 |
Bottom Surface | 16.27272727 | -1.49E-03 | 7.3769×10^13 |
Dislocation density observed in the deposits is more as compared to wire material, which means the strength of the deposits is better than that of wire. It has been observed that small crystallite size exhibits a larger dislocation density.
4.7. Residual stress measurement
Residual stress is measured using the XRD method through BRUKER D8-Advance®. In the XRD method, residual stress is computed by analysing the peak shifting phenomenon (as shown in Fig. 25). The standard d-Sin2(ѱ) method is followed for calculating the value of residual stress present on the sample surface [59–64]. The procedure for residual stress calculation is based on the work reported in [59]. In d-Sin2(ѱ) method, a graph (as shown in Fig. 26) is plotted between interplanar spacing (d) and Sin2(ѱ), from which residual stress is computed [59–64]. For residual stress measurement, current and voltage are set at 40 mA and 40 KV respectively for performing initial 2θ scanning from 30° to 126° for all samples. In this 2θ range the observed diffraction peaks are {110}, {200}, {211}, and {220}. From the above planes, {211} plane is selected, which appeared at 99.6°, and residual stress is determined about this peak for all samples [60]. Here diffraction peaks are recorded at different ѱ values (from − 40° to 40°), and their corresponding interplanar spacing is determined. The obtained residual stress among the wire, top and bottom layer of the deposit is presented in Table 11 and Fig. 27.
Table 11
Residual stress value for wire, top, and bottom surface of the deposit
Sample | Residual stress (MPa) |
Wire (ER70S-6) | -227.3 ± 7.9 |
Top Surface | 100.8 ± 2.9 |
Bottom Surface | -139.1 ± 2.5 |
Residual stress observed in all three samples (from Table 11 and Fig. 27) is compressive in nature due to negative slope. As compared to wire, the magnitude of compressive residual stress is decreasing in the deposits, which means large residual stress present in the wire is relived during the deposition process. Both top and bottom layer of the deposit exhibits compressive residual stress, which means through the NTA approach, compressive residual stress is developed in the fabricated part. As compared to previous kinds of literature without any external post-processing, tensile residual stress is present in the deposition, which reduces mechanical performance of the fabricated part [16, 18]. In the NTA process, a layer is fabricated through drop by drop transfer of molten metal, which act as a thermal load during the deposition process. Also, when a detached droplet starts to cool at that time, another droplet is deposited adjacent to the previous one. As a result, continuous cooling and heating cycle is going on during the deposition process. Therefore compressive residual stress is developed in the deposits. The development of compressive residual stress will further improve the mechanical performance of the fabricated part in terms of high strength and fatigue life [18]. Less variation of residual stress between top and bottom layer symbolises the isotropic nature of the deposit.
4.8. Microhardness analysis
Vickers microhardness tester is used to perform the microhardness analysis at 500 gf load, and 10s dwell time [16, 19]. Before conducting the microhardness test, the samples are polished to get a flat surface. Microhardness is measured in top, bottom and lateral surface of the deposit. Indentations are made on the central axis of deposit. For the top and bottom surface, a 5mm gap is provided between each indentation, but for the lateral surface, a 3mm gap is provided between each indentation from the top surface. Microhardness values in all three samples are presented in Table 12–13. The variation of microhardness value with increasing distance is presented in Fig. 28.
Table 12
Variation of microhardness in the top and bottom surface of the deposit
Distance (mm) | Microhardness in top surface (HV) | Microhardness in bottom surface (HV) |
5 | 254.7 | 214.2 |
10 | 270.9 | 239.7 |
15 | 220.9 | 262.6 |
20 | 231.5 | 213 |
25 | 195.9 | 250.1 |
30 | 225.9 | 226.5 |
35 | 209.2 | 215.2 |
40 | 222.1 | 224.5 |
45 | 235 | 209.9 |
50 | 226.4 | 210.8 |
55 | 214.4 | 204.6 |
60 | 253.7 | 220.4 |
65 | 224 | 225.3 |
70 | 234.5 | 257.1 |
75 | 238.6 | 291.3 |
80 | 268.7 | 215 |
85 | 225.2 | 266 |
90 | 222.6 | 245.5 |
Average | 231.9 | 232.872 |
Table 13
Variation of microhardness in the lateral surface of the deposit
Distance from top surface (mm) | Microhardness in lateral surface (HV) |
3 | 226.7 |
6 | 224.3 |
9 | 238.9 |
12 | 232 |
15 | 204.1 |
18 | 203.3 |
21 | 206.7 |
24 | 224.4 |
27 | 231.3 |
30 | 241.5 |
33 | 261.7 |
Average | 226.809 |
It has been observed from Table 12–13 and Fig. 28 that the average microhardness value corresponding to top, bottom, and lateral surface of the deposits is 231.9 HV, 232.872 HV, and 226.809 HV, respectively. The microhardness value is marginally more in the bottom layer with respect to top surface of the deposit. It is due to the grain refinement that occurred in the bottom surface of the deposit due to better cooling action. There is less variation in microhardness value in between lateral and longitudinal directions. Therefore it can be concluded that the NTA approach generates isotropic deposits.