212
S. Gajjar et al.
Fig. 2 Fabricated fixture
Fig. 3 Forming tool
experiments are performed using a different number of forming stages (single stage,
two stages, three stages). Table 1 shows the number of stages and corresponding wall
angles. Figure 4 shows the drawing of desired formed parts with wall angle 40°, 55°
and 70°, respectively.
All eighteen experiments are carried out using Taguchi L 18 orthogonal array. The
process variables and their levels are listed in Table 2 and Taguchi design matrix is
presented in Table 3. Influence of three process parameters namely number of stages
(N), pitch size (P) and feed rate (F) on responses namely geometric accuracy (using
RMSE - root mean square error method) and surface roughness (R a value) is studied.
The response geometric accuracy is measured using RMSE - root mean square
error method. Formed parts are masked using Spotcheck’s SKD-S2 developer spray
and scanned using Dr. PICZA 3D laser scanner and it is as shown in Fig. 5. The part
is scanned and the data is processed and saved in point cloud data form which is then
imported in SolidWorks modelling software as depicted in Fig. 6. After importing
this point cloud data in SolidWorks software an outer profile of the formed part is
produced, which is the deformed part profile. This profile is compared with the target
part profile. The deviation between target and formed part profile is recorded at 42
Table 1 Forming stages and respective wall angle
Number of forming stages
1
2
3
Respective wall angle (in degrees)
70°
55° and 70°
40°, 55° and 70°
S. Gajjar et al.
Fig. 2 Fabricated fixture
Fig. 3 Forming tool
experiments are performed using a different number of forming stages (single stage,
two stages, three stages). Table 1 shows the number of stages and corresponding wall
angles. Figure 4 shows the drawing of desired formed parts with wall angle 40°, 55°
and 70°, respectively.
All eighteen experiments are carried out using Taguchi L 18 orthogonal array. The
process variables and their levels are listed in Table 2 and Taguchi design matrix is
presented in Table 3. Influence of three process parameters namely number of stages
(N), pitch size (P) and feed rate (F) on responses namely geometric accuracy (using
RMSE - root mean square error method) and surface roughness (R a value) is studied.
The response geometric accuracy is measured using RMSE - root mean square
error method. Formed parts are masked using Spotcheck’s SKD-S2 developer spray
and scanned using Dr. PICZA 3D laser scanner and it is as shown in Fig. 5. The part
is scanned and the data is processed and saved in point cloud data form which is then
imported in SolidWorks modelling software as depicted in Fig. 6. After importing
this point cloud data in SolidWorks software an outer profile of the formed part is
produced, which is the deformed part profile. This profile is compared with the target
part profile. The deviation between target and formed part profile is recorded at 42
Table 1 Forming stages and respective wall angle
Number of forming stages
1
2
3
Respective wall angle (in degrees)
70°
55° and 70°
40°, 55° and 70°