*3.2. Machined Surface Roughness Analysis*

The average Ra values obtained for the performed tests can be observed in Table 6.



As shown in Table 6, the machined surface roughness values were considerably higher for tests conducted at higher values of cutting length. This was expected, as Inconel 718 induces high levels of tool wear over a relatively low cutting length. As such, this tool wear negatively impacted the surface roughness values obtained. Regarding the performed measurements, it was noticed that, in the workpiece's periphery, the values were slightly higher than those at its center. This can be attributed, once again, to an increase in tool wear throughout the test, which was lower at the start of the machining path and, consequently, higher at its periphery (end of the toolpath). This variation was observed to be more

significant for tools that exhibited higher tool wear (consequently, this variation induced higher standard deviation values).

Apart from tool wear, it was noticed that machining parameters clearly influenced the machined surface's quality, especially the feed per tooth (fz) parameter [45]. In Figure 6, the obtained results regarding the machined surface roughness for the various test parameters can be observed.

**Figure 6.** Comparison of surface roughness values for all test conditions (divided into three series based on the selected fz parameter).

As shown in Figure 6, there was a common tendency for machined surface roughness variation with the feed-per-tooth value. Usually, at lower values of this parameter, the machined surface quality is better than that obtained at higher values of feed per tooth [45,46]. This tendency was observed for both 100 m/min and 125 m/min in the tests conducted at a 5 m cutting length. However, concerning the tests conducted at a 15 m cutting length (L15), the tendency was slightly different for the case of the machined surface quality produced by tools machining at a 100 m/min cutting speed. Indeed, there was an increase in surface roughness from 75% to 100% feed per tooth in the tests conducted at a 100 m/min cutting speed. This value, however, decreased slightly from 100% to 150% feed per tooth.

In the tests conducted at a 125 m/min cutting speed and 15 m cutting length, the tendency was the same as that registered for the tests witha5m cutting length. This means that the machined surface quality deteriorated as the feed-per-tooth value increased. However, the machined surface roughness values obtained at a 125 m/min cutting speed tended to be slightly higher than those obtained at a 100 m/min cutting speed. This is not commonly observed, as cutting speed tends to produce a slightly better machined surface quality (although the feed per tooth is a more influential parameter in the machined surface quality) [47]. This is probably related to the fact that the tools' wear, in this case, was higher than that registered at a 100 m/min cutting speed (this will be explained in more detail in Section 3.3).

The fz parameter had a clear influence on the produced machined surface's roughness. Higher feed-per-tooth values produced higher Ra values, except in the case of S100L15 tools tested at 150% fz, which produced a better surface quality than that registered at 75% fz, which was the indicated value to obtain lower values of machined surface roughness. Regarding the cutting speed, as previously mentioned, a decrease in the surface roughness value commonly occurs when the cutting speed value increases. However, this was not the case for what was observed. This was probably due to the amount of tool-wear sustained by the tools at this cutting speed [48].

It is also worth noting that the values for the 75% fz produced at 100 m/min and 125 m/min were quite similar, only having a noticeable difference at 150% fz. Regarding the standard deviation registered for all surface roughness measurements, it seems that the value was higher for tests conducted at 150% fz, especially at 125 m/min. This was probably due to higher sustained tool wear, which resulted in a greater difference in the registered Ra from the center of the workpiece to its periphery (where the tool wear would be higher) [49,50].
