Study and Experimental Investigation of the Effect of Progressive Feed Rate on Surface Roughness in CNC End Milling Process Using RSM

The challenge of contemporary machining industries is mainly interested in the achievement of high-quality products in terms of surface finish, dimensional exactness in a stipulated time. End milling is one of the most commonly used metal cutting process occupying cores of manufacturing industries like aerospace, defense, automotive, marine, and other fields. The mechanism behind the formation of surface finish in CNC end milling process is very dynamic, complicated and process dependent. The quality of surface of a machined component plays an important role in its performance. A good quality milled surface significantly improves fatigue life, corrosion resistance, and creep life. Surface finish is of great important, particularly in case of fighter aircraft structural components, as they will be subjected to high level of fatigue loads. Some of fighter aircraft detail parts are very intricate in shape and needs extensive end milling process to realize the final component.

The ability to control the machining process by setting proper input factors, for a better surface finish and in turn a good quality product is of vital importance. Many researchers have concentrated on various optimization and predictive modeling techniques to determine the optimal input parameters, and to achieve the better surface finish. Hashimoto et al. [1] described the characteristics of surfaces created by various finishing methods, and discussed the influence of surface finish on its performance. According to Ghosh et al. [2] better surface finish can be obtained by a combination of high speed, low feed, and low depth of cut. They concluded that RSM-PSO is an efficient optimization technique for keyway milling. The advantage of dry machining over the wet machining has been clearly established by selecting proper cutting tools and tool geometry [3, 4]. The optimization, carried out in their work gives an opportunity for the user to select the best tool geometry and cutting conditions so as to get the required surface quality. Varghese et al. [5] presented a detailed investigation of effects of cutting parameters and different cooling environments on the machinability of austenitic stainless steel AISI 304. ANOVA and Regression analysis, through a DoE full factorial design (4³), were conducted to relate the surface roughness of face milled high strength steel to the most common machining parameters [6]. The optimization resulted in spindle speed of 1250rpm, depth of cut of 1.0 mm, and table feed rate of 67mm/min with a composite desirability of 0.83, to give R_a and MRR values of 0.15 and 2333mm³/min respectively. The objective of the study is to determine the optimum combination of parameters for minimum surface roughness during MQL assisted milling process on AISI-1045 alloy [7]. Surface roughness model is developed by both ANN and RSM. It is found that RSM coupled with PSO gives better result and it is validated by confirmation tests. It is perceived that the feed has significant effect on surface roughness followed by spindle speed and Depth of cut has small effect on roughness. Kumar and Rajamohan [8] developed a mathematical model using RSM, with the data obtained by end milling of Al 6063-T6. The surface roughness and flatness values predicted from the proposed model were found in good agreement with the experimental results. Investigations shows output response increases rapidly with an increase of feed rate and decreases with an increase in cutting speed. A fuzzy model was developed for predicting the surface roughness for a set of input parameters: cutting speed, feed rate and depth of cut [9]. Therefore, the operator can predict the quality of the response in advance with that. Vakondios et al. [10] examined the influence of the milling strategy selection on the surface roughness of an aluminum alloy Al7075-T6. They performed 96 experiments and the results are processed using regression analysis and ANOVA. The models were statistically validated and experimentally verified are in good fit. Surface topography of super alloy GH4169 work piece machined by milling and grinding was studied and their result shows that R_a and R_z values from milling are 2-5 times and 1-3 times as high as those of grinding [11]. Ramanujam et al. [12] conducted CNC end milling experiments on Inconel 718 super alloy based on L₉ orthogonal array and optimized the cutting parameters by Taguchi and desirability function. They developed a regression model for R_a and MRR. ANOVA shows that the cutting velocity is the most significant machining parameter. Pang et al. [13] carried out an experimental investigation on end milling of halloysite nanotubes (HNTs) based on Taguchi’s L₂₇ orthogonal array. Results from their study shows that the application of the Taguchi method can determine the best combination of machining parameters that can provide the optimal machining responses: surface roughness and cutting force. CNC milling of brass was conducted by Ranganath et al. [14] have used RSM to accomplish the objective of the experimental study. The result shows that, tool diameter is the most significant factor followed by feed rate. Yahya et al. [15] optimized machining parameters based on surface roughness prediction for AA6061 using RSM. Cutter flutes have a more significant influence on surface roughness followed by feed rate and depth of cut. Julie et al. [16] were carried out ANOVA analysis to identify the significant factors affecting surface roughness and also found that optimal cutting combination by seeking the best surface roughness by SN ratio. Rashmi et al. [17, 18] were optimized the machining parameters for AA6061 using RSM& PSO, and calculated the cutting force by an indirect method. ANOVA results reveals that the effect of spindle speed is much more evident than the effect of feed rate and depth of cut on surface roughness.

A lot of research has been observed through literature on conventional constant feed machining of end milling process and minimizing the surface roughness using different statistical techniques. But only few articles [19, 20] are available on progressive feed machining, wherein the feed is applied progressively rather than instantly to minimize cutting force and tool wear. However, to the best of author’s knowledge there is no literature observed on minimizing surface roughness through progressive feed method of machining in end milling process. Especially machined components assembled in fighter aircraft structures like: longerons, spars, central frames, side frames, shear walls, trouser ducts, integral fuel walls and various supporting ribs of the fuselage etc., will be subjected to high level of fatigue loads during maneuvering. In the case of these parts surface finish plays a crucial role in their fatigue life and in-turn durability of the aircraft. Hence it is very important to fabricate a component with good surface finish to enhance the life of each detail part of aircraft. The primary objective of the present study is to adopt progressive feed method of machining in end milling of BS L168-T6511alloy, in order to achieve better surface finish on the machined components. A comparative analysis has been made between constant feed machining and progressive feed machining besides finding the influence of cutting parameters on output response (R_a). The chosen material for this research work is a widely used for high technology applications like aerospace, defense, marine and automotive etc. due to its high level of technical merits.

Cutting parameters and their levels are finalized based on the trial runs and literature review. The various machining parameters and their levels are shown in Table 1 and Table 2.

3.1 Workpiece details

BS L168-T6511 aluminum alloy work piece samples of size 100mm×80mm×20mm are considered for the experimental study. Test samples are prepared from a single billet to have identical properties throughout the specimen. The chemical composition of the work piece material is shown in Table 3, confirmed by spectroscopy.

3.2 Cutting tool and machine tool details

Precision measuring instruments Co. made X-PM16807-2000 solid carbide end mill cutter of 20mm diameter, with two flutes, has been used for this purpose (Figure 2a). The machine tool shown in Figure 2b, used for conducting experiments is the AMS MCV-450 model, three axes CNC vertical machining center. The major specifications of the machine are: Axis travel: X-800mm, Y-450mm, Z-500mm. Maximum spindle speed-60000rpm; Maximum feed rate-10m/min; Basic power supply–18kVA, Positional accuracy-0.015mm, Repeatability ± 0.005mm.

2a.png

(a) Solid carbide end mill cutter with two flutes

2b.png

(b) AMS MCV–450 Vertical milling machine

2c.png

(c) Dry cutting condition

Figure 2. Machine tool and cutting tool used for research

3.3 Experiment details

Experiments were conducted as per Taguchi’s L27 orthogonal array [21, 22] and detailed as in Table 4. Spindle speed (s), feed rate (f) and axial depth of cut (d) are the input process parameters and surface roughness (R_a) is the output response. Conventional climb milling with dry machining (Refer Figure 2c) condition was chosen to study the clear cut analysis of the constant feed and progressive feed machining. The measured surface roughness (R_a) values against each experimental run are depicted in Table 4. Column ‘x’ shows the starting percentage of feed rate (f) and column ‘Y’ displays the percentage of improvement in surface finish against each experimental run. S/N ratio of R_a from constant feed and progressive feed machining are presented in last two columns of the Table 4 respectively.

The effect of spindle speed, feed rate, and depth of cut on the aforesaid response while machining of BS L168-T6511 alloy in end milling process has been investigated. Experiments were conducted as per the Taguchi’s L₂₇ factorial method in order to estimate the effect of input factors on average surface roughness (R_a) in both constant feed and progressive feed machining environments. From attained results, subsequent conclusions were derived:

(1) Progressive feed method of machining offered considerable reduction in average surface roughness (R_a) around 19.206% over conventional constant feed machining (Table 4). It is due to reduction in sudden impact of cutting tool with work piece, reduction in cutting tool deflection, minimal tool vibration and less chattering in progressive feed machining contributed to substantial drop in the response (R_a).

(2) The predicted values of Ra from the developed regression mathematical models are in good agreement with the experimental results. Errors, in constant feed and progressive feed machining are only 2.56% and 5.08% respectively. These errors are mainly due to uncontrollable variables like: chip loads, chip formation, tool wear and unexpected variations in power supply., etc.

(3) From the ANOVA analysis, feed rate has a foremost contribution on the considered response (R_a) both in constant feed machining as well as in progressive feed machining.

(4) The plot of residual errors (Figure 6a) was normally distributed, since the model residuals are following the path of straight line. Main effect plots, Interaction plots, and 3D surface plots reveal that considered response (R_a) increases drastically with the increase of feed rate in both the methods of machining.

(5) Better surface finish can be attained by incorporating lower feed rate, with higher spindle speed and higher depth of cut for the aforesaid material. This work also emphasizes that proper selection of cutting parameters along with progressive feed machining eliminates the use of secondary finishing operations and hence saves the manufacturing time and cost.