3.1 Powder size effect
In fabrication of FGM by means of pre-mixed powder, it is probable that the powder flow of two different materials do not reach the meltpool at the same time and with the expected ratio. This is because of different sizes and densities which causes different acceleration of particles. To overcome this issue, since the powder density is constant, the size of powder particles should be chosen in a way that all the particles have the same acceleration. As Li et. al (9), explained in their study, particle acceleration equation can be written as following:
\(\frac{d\stackrel{⃑}{x}}{dt}={u}_{p}\)
|
(3-1)
|
\(\frac{d{u}_{p}}{dt}=\frac{18\mu }{{\rho }_{p}{d}_{p}^{2}}\frac{{C}_{d}Re}{24}\left({u}_{ar}-{u}_{p}\right)+\frac{g\left({\rho }_{p}-{\rho }_{ar}\right)}{{\rho }_{p}}\)
|
(3-2)
|
Where, \({u}_{p}\), \({u}_{ar}\), \(\mu\), \({C}_{d}\), \({\rho }_{ar}\), \({\rho }_{p}\), \(Re\) and \({d}_{p}\), is particle velocity, argon gas flow velocity, viscosity, drag coefficient, argon density, powder density, Reynolds number and particle diameter. In Eq. (3-2), the first and the second term on the right side represent the acceleration caused by drag force and the gravity. Since particle density, is much greater than argon gas density, the second term can be simplified as gravitational acceleration, g. Thus, in order to obtain the same acceleration for particles of the two different materials, the first term should be equal for both of them. The only coefficient that varies with different types of powders is the product of density and diameter square. Therefore, the particle size satisfies the following equation:
\({\rho }_{p-1}{d}_{p-1}^{2}={\rho }_{p-2}{d}_{p-2}^{2}\)
|
(3-3)
|
\(\frac{{d}_{p-1}}{{d}_{p-2}}=\sqrt{\frac{{\rho }_{p-2}}{{\rho }_{p-1}}}\)
|
(3-4)
|
In order to investigate whether choosing powder size based on Eq. (3-4) would prevent from separation of pre-mixed powder or not, four sets of pre-mixed powder was prepared (Table 2).
Mixture
|
Volume Ratio
|
WC-12%Co Diameter
|
SS Diameter
|
Table 2
Characteristic of pre-mixed powder used in experiments
A
|
36% WC-12Co − 64% SS316
|
26–30
|
45–63
|
B
|
36% WC-12Co − 64% SS316
|
26–30
|
90–95
|
C
|
17% WC-12Co − 83% SS309
|
26–30
|
45–63
|
D
|
17% WC-12Co − 83% SS309
|
26–30
|
63–88
|
A single layer was deposited with feeding each mixture. The shielding gas flow rate was 3 lit/min in these experiments. |
The layer A and B shown in Fig. 4, are deposited from mixture A and B, which had the same materials and ratio, but different SS316L powder size.
SEM photos, Fig. 5 and Fig. 6 shows two sections of these layers in different locations along X direction.
As it can be seen in SEM photos, WC powders are not melted due to their high melting point (2870 ⸰C). The brighter color of WC particles made it possible to calculate the percent area it occupied, in each section, with assistance of ImageJ software (Fig. 7).
Figure 7 and also comparing Fig. 5 and Fig. 6, shows that in layer A, which was deposited by the pre-mixed powder with the proper powder size, the both sections have approximately same amount of WC. In layer B, the SS316L powder size was bigger than the proper diameter to satisfy Eq.(34), which causes unstable ratio of WC along the layer. The lighter particles of WC-12Co (due to smaller size) reached the meltpool faster at first, thus their amount in the mixture descend at the end of deposition.
Figure 8 shows the layers deposited from mixture C and D. They have the same materials ratio. In layer C particles diameter are chosen based on Eq.(34), while in layer D, SS309L particles are larger than it should be.
SEM photos of different sections along X direction, in layer C and D, are shown in Fig. 9 and Fig. 10.
The percent of the area of each section which was filled with brighter particles of WC was calculated in ImageJ software. The result is shown in Fig. 11.
It can obviously be seen in Fig. 9 and Fig. 10, that the wrong choice of powder size in layer D caused inconsistency in ratio of WC and SS309L. At the beginning of deposition, the lighter particles of WC were more and gradually it became less. On the other hand, in layer C, where the product of density and diameter square for both material is approximately equal, the amount of WC relative to SS309L is more consistent along X direction.
3.2 Shielding gas flow rate effect
The authors found out in their previous study(29) that the shielding gas, flowing in the axial channel of the nozzle, has effect on powder stream distribution. Whatever particles is lighter, they are blown more off the axial axis and gather in a ring with larger diameter. Thus, if there are two kinds of materials in the powder stream, it may cause separation. There is the chance that the lighter particles in the bigger ring, fall out of the meltpool and consequently their cooperation in building the layer is less than expected. To investigate this issue, two layers were deposited based on Table 3. To eliminate the effect of different acceleration in carrier gas flow, the powder size for two materials is chosen based on Eq.(3-4).
Table 3
Characteristic of pre-mixed powder used in experiments
Layer
|
Volume Ratio
|
WC-12%Co Diameter(µm)
|
SS Diameter (µm)
|
Shielding Gas Flow rate(lit/min)
|
E
|
36% WC-12Co − 64% SS316
|
26–30
|
45–63
|
3
|
F
|
36% WC-12Co − 64% SS316
|
26–30
|
45–63
|
5
|
SEM photos of layer E and F are shown in Fig. 12 and Fig. 13. |
The area occupied with WC in each section is calculated with ImageJ software and it is shown in Fig. 14.
Since particle size satisfies Eq. (34) in both layers, the materials ratio is consistent along X axis. However, the increase in shielding gas flow rate causes the lighter particles of WC to fall out of meltpool. Consequently, the amount of WC in layer F (Fig. 13) is obviously less than layer E (Fig. 12).