Analysis of milling the parts with thin walls using CAM-FEM methods.

Ghionea, Gabriel Ionut ; Ghionea, Adrian Lucian ; Tanase, Ioan 等

1. INTRODUCTION

The use of modern CAD-CAM techniques in the parts manufacturing processes has a large application in the mechanical, automotive, airspatial industry domains (Altintas & Merdol, 2007). The specialty literature (Constantinescu et al., 2006) also highlights results of FEM applications in order to determine the elastic deformations values and the stresses inducted in parts by the contact with the tools during the working processes. The influences of the technological system (machine-tool, tools, fixture devices) determine dimensional variations and irregularities of the processed surfaces, dynamical behavior of the machine-tool in the cutting process and also the premature wear of the cutting edges (Childs et al., 2000). The results of the surfaces generation by simulation, with the aid of CAM techniques (tool path, division pattern of the stock left for machining, different working strategies, etc.) offer a large number of data: processing times, surface accuracy (roughness size), behavior of the machine-tool in working conditions (power and cutting torque) etc., both for roughing and finishing operations. The aim of the presented analysis is to show the working simulation possibilities (CAM and FEM) using specialized software. For the optimization purpose of the technological process there are considered and applied various criteria, software environments, manufacturing data. The paper presents a method to calculate the cutting power and force, to identify the spots where the force is applied on the part's thin walls, but also the stress deformation values using simulations.

2. ASPECTS OF THE CAM SIMULATION

Preparing the piece for processing on a machine-tool with numerical control involves the generation of command information, all data is then stored in a preset order within a storage device. On modern approaches, the NC codes can be generated using a CAD-CAM program and a virtual model of the piece (Ghionea, 2009). Defining the piece in a CAD environment is used as the entry date to generate the NC program codes with one of the complex existing programming languages (Kief, 2001/02). Thus, the simulation is justified to optimize the process. For the purposes of this paper, it is considered a part having its 3D model made in CATIA Part Design module and presented in figure 1. The overall dimensions of the stock part are 100x100x32 mm.

[FIGURE 1 OMITTED]

From the analysis of the part's surfaces it is identified the cavity delimited by a closed contour whose walls have thickness of 2 mm and 20 mm height. To express (by comparison) the material influence over the part's stiffness in manufacturing conditions, there are considered two materials: high-alloyed steel and aluminum alloy, having the yield strength of 2.5x[10.sup.8] N/[m.sup.2], respectively, 9.5x[10.sup.7] N/[m.sup.2].

To manufacture the part it is used a 3-axis CNC vertical milling machine-tool having the following main characteristics: spindle speed: 20000 rpm with variable speed range (direct drive spindle), the power of the principal electric engine: P = 15 kW (continuous rating), machining feedrate max: 7000 mm/min and rapid feedrate: 20000 mm/min. The machine-tool has a CNC Sinumerik controller. The tools used in manufacturing simulation are chosen from a company catalogue (Sandvik, 2008). Also, the toolholders are in correspondence with the spindle nose and the holding system of the machine-tool.

The main steps of the CAM simulation include the following operations: Face mill, External roughing profile contouring mill, External finishing profile contouring mill, Internal roughing profile pocketing mill, Internal finishing profile contouring mill and Drill four holes.

In figure 2 it is shown the part during the profile contouring mill, finish surface. The parameters are: two passes, tool diameter [D.sub.c] = 12 mm, average chip thickness [h.sub.m] = 0.03 mm, cutting speed [v.sub.c] = 735 m/min, specific cutting force [k.sub.c1] = 700 N/[mm.sup.2], exponent [m.sub.c] = 0.25, spindle speed [n.sub.c] = 19500 rpm, feed speed [v.sub.f] = 8500 mm/min, cutting power for removal of chips [P.sub.c] = 4.4 kW, feed per cutting edge [f.sub.z] = 0.11 mm, cutting depth [a.sub.p] = 20 mm, working engagement [a.sub.e] = 1 mm, number of cutting edges [z.sub.c]= 4, machining time [t.sub.m] = 27s, total time [t.sub.t] = 31s.

[FIGURE 2 OMITTED]

The values previously presented are established for the case of aluminum alloy machining, with 75 HB.

In the case the part would be processed of high-alloyed steel, with 200 HB, the data above will be modified. Thus, the parameters become: two passes, [D.sub.c] = 12 mm, [h.sub.m] = 0.03 mm, [v.sub.c] = 235 m/min, [k.sub.c1] = 1950 N/[mm.sup.2], [m.sub.c] = 0.25, [n.sub.c] = 6235 rpm, [v.sub.f] = 2750 mm/min, [P.sub.c] = 4.3 kW, [f.sub.z] = 0.11 mm, [a.sub.p] = 20 mm, [a.sub.e] = 1 mm, [z.sub.c] = 4, [t.sub.m] = 30 s, [t.sub.t] = 34 s. To determine the cutting power for this operation, it is used the next equation:

[P.sub.c] = [a.sub.p] x [a.sub.e] x [v.sub.f] x [k.sub.c]/60 x [10.sup.6] x [eta], (kW) (1)

where: [k.sub.c] = [k.sub.c1] - [h.sub.m.sup.mc] = 4685 N/[mm.sup.2] and [eta] = 1.

3. THE FEM SIMULATION AND ANALYSIS

In the FEM simulation it is considered a finishing end mill with the diameter [D.sub.c] = 12 mm, [z.sub.c] = 4, pitch of the helical cutting edge [P.sub.hc] = 37.4 mm, helix angle [omega] = 45[degrees]. On the contact line between the end mill and the part there are created four cutting spots placed on the height of the helix pitch. Two of the spots corresponding to the cutting depth ([a.sub.p] = 20 mm) are shown in figure 3. In order to establish these spots where the tangential medium cutting force is applied it was necessary to determine the contact length of each helical edge with the part. The dimensions of these spots and also the contact angle [phi] between the part and the cutting edge are calculated with the following equations:

[b.sub.c] = [square root of [D.sub.c.sup.2]/4 - [([D.sub.c]/2 - [a.sub.e]).sup.2]] = 3.32, (mm) (2)

[L.sub.ch] = [square root of [D.sub.c] x [a.sub.e]]/tg[bar.[omega]] = 3.46, (mm) (3)

[phi] = 2 x [square root of [a.sub.e]/[D.sub.c]] = 0.577.(rad) (4)

The medium cutting force acting on a contact spot for the contact angle [phi] is determined by the relation 5 and then compared with the values given by the relation 6.

[F.sub.m[phi]] = [L.sub.ch] x [h.sub.m] x [k.sub.c], (N) (5)

[F.sub.tm] = 60000 x [P.sub.c]/[v.sub.c] x (N) (6)

With the previously determined values for cutting power (aluminum alloy and high-alloyed steel) it is calculated the tangential medium cutting force, relation 6, and has the values: 360 N, respectively, 1098 N.

The medium radial cutting force (which is practically applied on the two spots) is [F.sub.rm] = (0.3 ... 1) * [F.sub.tm], considered [F.sub.rm] = 0.45 x [F.sub.tm]. As follows, there are presented some results of FEM simulations and analysis in the cases of the part being processed of aluminum alloy and high-alloyed steel. Thus, the medium radial cutting force is different for each case ([F.sub.rm] = 164 N - aluminum alloy and [F.sub.rm] = 536 N - high alloyed steel).

Figure 4 shows the Von Mises stresses calculated by the FEM analysis in the case of aluminum alloy and flat surface. For a force of 164 N applied on the two spots (Fig. 3) the results are: max. stress = 4.68x[10.sup.7] N/[m.sup.2] and max. elastic deformation = 0.062 mm (Fig. 5) with an error of 48.8 %. That error is too high, so a new analysis is done after a refinement of the part structure. After the refinement, the error is 32.7 % and the values become: max. stress = 6.22x[10.sub.7] N/[m.sup.2] and max. elastic deformation = 0.083 mm.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

In the high-alloyed steel case (flat surface S1, [F.sub.rm] = 536 N), for an error of 48.8 % the max. stress = 1.53x[10.sup.8] N/[m.sup.2] and max. elastic deformation = 0.07 mm. Also, for an error of 31 % the values are: max. stress = 2 x [10.sup.8] N/[m.sup.2], max. elastic deformation = 0.09 mm.

The user is interested in the outputs of the maximum stress that must not exceed the admissible material yield strengths.

4. CONCLUSION

The simulations' results show that the stresses inducted in the part's thin walls did not exceed the elastic domain. The maximum elastic deformations of the processed surfaces are greater in the case of the high-alloyed steel than in the case of the aluminum alloy. The data obtained in such simulations are very useful in real manufacturing processes.

5. REFERENCES

Altintas, Y.; Merdol, D.S. (2007). Virtual high performance milling. Annals of the CIRP, Vol. 56, No.1, (2007), pp.81-84, ISSN 1726-0604

Childs, T.; Maekawa, K.; Obikawa, T.; Yamane, Y. (2000). Metal machining. Theory and applications. Arnold Publishing, ISBN 0 340 69159 X, London, NW1 3 BH

Constantinescu, N. I.; Sorohan, St.; Pastrama, St. (2006). The practice of finite element modeling and analysis. Printech Publishing, ISBN 978-973-718-511-2, Bucharest

Ghionea, I. (2009). CATIA v5. Applications in mechanical engineering. BREN Publishing, ISBN 978-973-648-843-6, Bucharest

Kief, B. H. (2001/02). NC/CNC Handbuch. Carl Hanser Verlag, ISBN 3-446-21756-8, Wien

*** (2008) Cutting tools from Sandvik Coromant. Main catalogue. Milling. Drilling. Boring. Toolholding, C-2900: 6 ENG/01, Sweden, www.coromant.sandvik.com