OPTIMIZATION OF THE SUPERFINISHING PROCESS

R. Cebalo, D. Bajić and B. Bilić

Abstract

By superfinishing process, ultra precision machining process of cylindrical, conical, spherical and even flat surface machine parts is performed. These parts must have high wear resistance and low coefficient of friction.

In this paper impact factors on surface roughness are determined. According to the factors test plan and regression analysis, extend equation for mean arithmetical roughness is given. In other to get minimum values of the surface roughness, optimization of the mathematical model is done and optimal values of the examined factors are determined. The obtained results are, according to the experiment plan, valid for the testing of material 34 CrNiMo6. The test results are to be probably applied to other materials, however, has to be proved for each separate case.

Keywords: surface roughness, superfinishing, mathematical modelling, optimisation

  1. Introduction

Surface finish quality influence to the exploitation characteristics of machine parts (sliding clearances, noise, lubrication etc.). Ideal straight and smooth surface cannot be obtained, but there are several processes that improve finish quality. Smooth surface provide a very close fit between two parts in contact and for parts exposed to the dynamic load. Supefinishing, figure 1., is a surface–improving process that removes undesirable effects in exploitation. Due to the high costs of the process the shape and dimensions of the workpiece are to be obtained by a procedure of fine machining (e.g. fine turning, milling, grinding), leaving a machining supplement which depending on the surface quality obtained by the previous machining, does not go over 0.005 to 0.01 mm. Roughness Ra from 0.012 m to 0.0025 m can be obtained by superfinishing. The basic features of the process are the following:

  • slow tangential circumferential speed of the workpiece,
  • medium pressure,
  • little heat,
  • large area of workpiece and tools contact,
  • short machining time,
  • silent machining.

During the procedure the workpiece surface roughness falls very quickly. Simultaneously, the workpiece surface effects the stone surface causing the wear of the stone. Full automatisation of machines for superfinishing process is obtained by building in the elements for acceptance, supply and further transport of workpieces and by building in devices for their automatic control. Besides great advantages obtained by such a process it is chiefly applied in mass production since it is difficult to decide on optimal conditions for a particular process without long lasting experiments. In piecemeal machining in which theprocess was not optimalised machining costs would be too high.

Figure1. Illustration of superfinishing process

During machining there is practically no warming up of the workpiece and consequently the workpiece surface does not alter characteristics obtained by previous heat machining. This is why the possibility of heat exertions of surface in the course of machining is excluded.

  1. Influencing factors on the surface roughness

During machining, figure 2., main motion is a complex motion from workpiece rotation with tangential circumferential speed vo and stone vibration parallel to the workpiece axis with speed vv. Feed speed vfis also parallel to the workpiece axis. Stone presses upon workpiece with the force Fn and the pressure between the workpiece and stone appears, the high of which enables the chip formation process. The result of such motion is a sinusoidal curve, which appears at the surface of the workpiece in the form of a coil. Because of such motions an abrasive grain seldom covers the same way, thus taking off roughness without new scratches. Relative motion of tools (stone) in relation to the workpiece can be decomposed into three orthogonal speed components, figure 2:

-speed components parallel with workpiece surface:

1.axial speed va (feed velocity vf and oscillation velocity vv),

2.tangential circumferential speed vo,

-speed component vertical to workpiece surface:

1.supply speed vn.

The result of superfinishing is influenced by two sets of units, figure 3:

  • Machining system
  • Operating parameters

So-called “statical units” which do not change in the course of processing define machining system:

  • Machine (work effect, power, dynamic properties),
  • Workpiece (geometry, material, quality of roughing out),
  • Tools (geometry, type of abrasive, grain size, hardness, porosity, bond, finish quality, accuracy),
  • Cooling (type, viscosity, concentration, quantity, mode of supply, pressure).

Operating parameters are variable units, which alter according to the needs and within the limits of the construction capabilities of the operating system:

  • Pressure,
  • Amplitude of axial sinusoidal oscillation,
  • Workpiece rotational speed,
  • Axial oscillation frequency.

Figure 2. Superfinishing stone in contact with the workpiece

Figure 3. Impact factors and superfinishing result

3.Testing conditions

Testing were executed using universal lathe “Prvomajska”, type D420/1500. Preparation of the specimen were done to remove influence layer, rust, grooves, damage in material, admixtures in surface and to get specimen with defined dimensions and initial roughness. Superfinish device “Supfina SE-40” was fixed in tool holder on lathe, so the lathe was used for superfinish process also. Sandvik Coromant tools and inserts were used for specimen preparation (inserts TNMG 16 04 08 – PR and TNMG 16 04 08 – PF as well as tool PTGNR 20 20 K 16).

Superfinishing stones for this investigation contained 400 grit silicon carbide abrasive. Each stone had a rectangular cross section 60 mm wide (axial direction) and 28 mm thick (circumferential direction). Each workpiece was a cylinder of 34 CrNiMo6, 70 mm outer diameter and 420 mm length, subdivided into 5 test specimens 58 mm wide. Device “Surtronic 3+” was used for measurement of surface roughness. Beside device, Data Processing Module was used. Data processing Module connected directly to the “Surtronic 3+” via the RS232 port, has an integral thermal printer for hard copy output of measured profiles and results lists. During superfinishing a mixture of mineral oil and kerosene in ration 60/40% was applied.

4.Results of experiments and optimisation

As it was mentioned before, surface roughness during superfinishing is influenced by lots of impact factors. Due to limited number of factors that could be examined in the same time, in this paper are chosen:

  • tangential circumferential speed vo,
  • air pressure p,
  • machining time t,
  • initial roughness Ra0.

According to the design of experiment, in table 1. experimental results of measured surface roughness are given. Statistical analysis of data, using software Design Experiment 6.0, was done. Using obtained response functions models, singular values for certain conditions were calculated and set in table 1. High value of coefficient of determination shows high parity of calculated values with the measured values. The response function is well modeled by a non-linear function of the independent variables, and the approximating functions for Ra is the second order model:

(1)

with coefficient of determination R2=0.9749.

Applying the partial derivation as an optimization method on the function (1), optimal value of influencing parameter can be determined:

(2)

Thus, following equations has to be solved:

(3)

Optimal parameters obtained applying this procedure are:

vo = 1.672m/s / p= 0.178MPa / t = 18.67s / Ra0= 0.95m

Table 1. Measured values of machined surface roughness

FACTOR / vom/s / pMPa / ts / Ra0m
FACTOR CODE / X1 / X2 / X3 / X4
BASIC LEVEL / (0) / 1.5375 / 0.15 / 15 / 1.21
UPPER LEVEL / (+1) / 2.050 / 0.2 / 20 / 1.52
LOWER LEVEL / (-1) / 1.025 / 0.1 / 10 / 0.89
LOWER LEVEL OF MIDDLE AXIS / (-2) / 0.513 / 0.05 / 5 / 0.58
UPPER LEVEL OF MIDDLE AXIS / (+2) / 2.560 / 0.25 / 25 / 1.84
R.B. / X1 / X2 / X3 / X4 / Ram / Rmaxm / Rzm
1 / -1 / -1 / -1 / -1 / 0.35 / 3.05 / 2.71
2 / +1 / -1 / -1 / -1 / 0.37 / 3.04 / 2.73
3 / -1 / +1 / -1 / -1 / 0.24 / 2.46 / 1.92
4 / +1 / + / -1 / -1 / 0.14 / 1.60 / 1.11
5 / -1 / -1 / +1 / -1 / 0.16 / 1.94 / 1.31
6 / +1 / -1 / +1 / -1 / 0.20 / 2.25 / 1.65
7 / -1 / +1 / +1 / -1 / 0.11 / 1.42 / 1.03
8 / +1 / +1 / +1 / -1 / 0.08 / 1.01 / 0.67
9 / -1 / -1 / -1 / +1 / 0.47 / 3.35 / 2.78
10 / +1 / -1 / -1 / +1 / 0.50 / 3.40 / 2.85
11 / -1 / +1 / -1 / +1 / 0.29 / 2.76 / 1.98
12 / +1 / +1 / -1 / +1 / 0.17 / 2.03 / 1.47
13 / -1 / -1 / +1 / +1 / 0.26 / 2.67 / 2.11
14 / +1 / -1 / +1 / +1 / 0.38 / 3.13 / 2.76
15 / -1 / +1 / +1 / +1 / 0.15 / 1.80 / 1.32
16 / +1 / +1 / +1 / +1 / 0.11 / 1.43 / 1.06
17 / -2 / 0 / 0 / 0 / 0.25 / 2.53 / 1.90
18 / +2 / 0 / 0 / 0 / 0.17 / 2.01 / 1.42
19 / 0 / -2 / 0 / 0 / 0.45 / 3.13 / 2.77
20 / 0 / +2 / 0 / 0 / 0.14 / 1.63 / 1.15
21 / 0 / 0 / -2 / 0 / 0.36 / 3.09 / 2.71
22 / 0 / 0 / +2 / 0 / 0.10 / 1.34 / 0.95
23 / 0 / 0 / 0 / -2 / 0.07 / 0.98 / 0.63
24 / 0 / 0 / 0 / +2 / 0.25 / 2.54 / 2.08
25 / 0 / 0 / 0 / 0 / 0.11 / 1.41 / 1.05
26 / 0 / 0 / 0 / 0 / 0.10 / 1.34 / 0.91
27 / 0 / 0 / 0 / 0 / 0.11 / 1.39 / 1.03
28 / 0 / 0 / 0 / 0 / 0.10 / 1.30 / 0.88
29 / 0 / 0 / 0 / 0 / 0.10 / 1.32 / 0.89
30 / 0 / 0 / 0 / 0 / 0.11 / 1.40 / 1.05
31 / 0 / 0 / 0 / 0 / 0.11 / 1.40 / 1.07

Figure 4 shows dependence of surface roughness upon stone pressure and tangential component of cutting speed for optimal values of machining time and initial roughness.

Figure 4. Dependence of surface roughness upon stone pressure and tangential component of cutting speed for optimal values of machining time and initial roughness

4.Conclusion

In this paper effect of tangential circumferential speed vo, air pressure p,machining time t, initial roughness Ra0on surface roughness during superfinishing process has been examined. The strongest effect on roughness has air pressure, and than follows machining time, initial roughness and tangential circumferential speed. Model that includes only impact factors was not satisfied because of low coefficient of determination. Because of that, interaction between factors should be included. The strongest effect has interaction between air pressure and initial roughness, as well as interaction between air pressure and machining time. However, interaction between tangential circumferential speed and initial roughness, as well as interaction between machining time and initial roughness has not significant effect on surface roughness. Applying the partial derivation as an optimization method on the function (1), optimal value of influencing parameter is also determined. The obtained results are, according to the experiment plan, valid for the testing of material 34 CrNiMo6. The test results are to be probably applied to other materials, however, has to be proved for each separate case.

References

[1]Cebalo,R. “Optimalno područje brušenja u puno vatrootpornih Ni-legura za plinske turbine”, Ph.D. Thesis, FSB, Zagreb, 1985.

[2]Bajić, D. “Doprinos poboljšanju obradivosti kod kratkohodnog honovanja”, Ph.D. Thesis, University of Zagreb, FSB, Zagreb 2000.

[3]Cebalo,R., Bajić,D. “Fundamentals of the Superfinishing Process, 5th International Scientific Conference on Production Engineering, Opatija, 1999. Str. I-095.

Roko Cebalo, Ph.D., Prof., retired,

FSB, University of Zagreb, Machine tools, I. Lučića 1 , 10000 Zagreb, Croatia, ++38516198558

Dražen Bajić, Ph.D., Ass. Prof.

FESB, University .of Split, Ruđera Boškovića bb, 21000 Split, Croatia,

Boženko Bilić, PhD., Ass. Prof.

FESB, University of Split, Ruđera Boškovića bb, 21000 Split, Croatia,

1