The Effect of Machining Fluid in the Process of Steam-Treated Pine and Beech Wood Turning on Selected Surface Roughness Parameters

Abstract

In this study, the influence of different cutting conditions on the quality of the machined surface during the turning process of steam-treated pine wood (Pinus sylvestris L.) and beech wood (Fagus sylvatica L.) was investigated. The cutting conditions analysed included dry turning, turning with compressed air cooling, and turning with oil mist cooling. The turning processes for all analysed cutting conditions were carried out for three different feed per revolution values. The carried out studies allowed the observation of a noticeable increase in the values of the analysed surface roughness parameters (Ra, Rz, Rq, Rt, and Rp) with the increase in feed per revolution for pine and beech wood. In addition, a noticeable decrease in the values of these parameters was observed with the use of compressed air and oil mist (MQL) for pine wood. However, in the case of beech wood, the lowest surface roughness values were observed during dry turning, while the use of MQL leads to increased values of surface roughness, especially at high feeds per revolution.

1. Introduction

Wood, as a natural construction material, has been used by humans for thousands of years. Modern thermal and chemical treatment technologies make it possible to reduce the weaknesses of wood that previously limited its range of applications [1]. Pine wood (Pinus sylvestris L.) is the most widely used species of softwood in Poland and Central and Eastern Europe as a construction material. Beech wood (Fagus sylvatica L.) is one of the most commonly used hardwood species in Europe and is widely applied in processing.

The treatment processes mainly include thermal processes such as drying, steaming, and high-temperature treatment, as well as chemical processes such as impregnation, painting, and the application of biological coatings [2]. Thermal treatment improves wood properties in terms of higher resistance to biological factors (insects, fungi) and reduces its hygroscopicity. However, these processes also affect the mechanical and machinability properties of wood. Research has shown that basic drying processes influence the mechanical properties of pine and beech wood [3], but also cutting power [4], quality of machined surface [5,6,7], as well as the amount of dust and chips generated during machining [8]. Chemical treatments similarly influence mechanical properties, cutting forces and power [9], quality of machined surface [10], as well as the granulation of chips and dust produced during machining.

Steaming is an important process in the treatment of structural wood materials, as it significantly improves the possibility of shaping wooden elements through bending. However, it also affects the mechanical properties of wood, cutting forces and power, and surface roughness parameters during machining [11,12,13]. It can be assumed that changes in material properties influence the tribological behaviour in machining processes. The quality of the cutting process (dimensional accuracy, surface roughness) of wood and wood-based materials depends on: the type of processed material [14,15,16], the type of cutting process and geometry of cutting edge [17,18,19], the parameters of the cutting process [20,21,22,23], and the level of cutting edge wear [24,25,26,27]. However, with a stable machining process, for selected cutting parameters, cutting tool type and geometry of cutting blade, the degree of wear of the cutting edge has a major effect on the surface roughness [24,27,28,29,30,31,32]. Therefore, if the steaming process can affect the wood properties in such a way that the tribological conditions during the machining process change, and as a result, can also affect the quality of the machined surface, then external factors should be used to reduce the friction coefficient in the machining process. The most common method of reducing the friction coefficient in the cutting process is the use of machining fluids. This solution is widely used in metalworking [33,34,35]. In woodworking, machining fluids have sometimes been used in narrow-sawing processes cutting through wet wood to reduce the adhesion of resins, which are contained in softwood, to the cutting blades. Chuchala et al. [36] analysed the effects of using machining fluids in dry woodworking. The study showed that the fluids had little effect on the colour and chemical properties of the machined surface. Staszczyk et al. [37] also proposed a formula for a machining fluid dedicated to the machining of glued wood, demonstrating its biodegradability, negligible impact on wood properties, and no weakening of the adhesive joints.

The minimum quantity lubrication (MQL) system is an environmentally friendly system for using machining fluids. This system consumes small amounts of machining fluid, which has a positive impact on the environment, but also means that less fluid will affect the machined wood. Chuchala et al. [36] showed that with this method of applying machining fluid, the impact of the fluid on colour and chemical changes is negligible.

The aim of this study was to examine how the use of the MQL system during longitudinal turning process affects the surface quality of steamed pine and beech wood.

2. Materials and Methods

2.1. Material

The studies were carried out on Scots pine (Pinus sylvestris L.) and beech (Fagus sylvatica L.) wood obtained from the Pomeranian region in Poland. After sawing the logs into lumber, they were subjected to a steaming process, which was followed by drying the wood to a moisture content MC = 12%. The steaming process was carried out in accordance with the procedure described in the works of Klement et al. [38] and Konopka et al. [39].

From the dried boards, laboratory samples of square cross-section were cut: 49 mm × 49 mm for pine and 37 mm × 37 mm for beech, with a sample length of 220 mm. In the next stage of processing, the samples were turned to diameters of ∅48 and ∅36 mm, respectively, without the use of any lubricants. The samples were stored in laboratory conditions for 72 h and the MC was stabilised at levels of 10% ± 0.8 for beech and 9 ± 0.6 for pine. The densities of the analysed samples, after oven drying, were as follows: for beech ρ = 631.3 kg·m⁻³ (SD 17.7—standard deviation) and for pine ρ = 416.6 kg·m⁻³ (SD 33.2).

2.2. Machine Tools, Cutting Tools, and Cutting Parameters

The machinability tests were conducted on a universal lathe with a chuck. The wood samples were mounted in the chuck of the lathe. The turning knife was specially made of high-speed steel, and the basic geometric parameters of the cutting edge were as follows: tool side rake angle γ_f = 18°, tool side clearance angle α_f = 27°, tool main cutting edge angle κ_r = 90°, tool minor cutting edge angle κ_r′ = 10° (Figure 1a,b).

The turning process was carried out under three different cutting conditions: dry turning, turning with compressed air cooling, and turning with the MQL system. For each of the mentioned processes, tests were performed at three different feed per revolution values (Figure 1c), while the remaining cutting parameters remained unchanged (

Table 1. Wood turning process parameters and parameters of the compressed air and MQL cooling systems.

Parameter	Symbol	Value	Unit
Cutting speed	vc	2.95	m·s−1
Feed per revolution	f1	0.07	mm
	f2	0.14	mm
	f3	0.28	mm
Depth of cut	ap	0.5	mm
Air flow rate	qa	0.4	m3·h−1
Air pressure	p	0.5	MPa
Fluid flow rate	qf	0.18 × 10−3	m3·h−1

).

The feed speed vector direction was parallel to the wood fibres direction. However, the cutting process was carried out in a 90–90 configuration in accordance with Kivimaa [40].

The machining fluid sprayed through the MQL system was a mixture of different fractions of heavy petroleum oil with a boiling point in the range of 155–217 °C (60%–80% m/m), aromatic hydrocarbons with carbon numbers predominantly in C8–C10 and boiling point in the 135–210 °C (10%–30% m/m) range, and white mineral oil with viscosity at 40 °C (10%–30% m/m) in the range of 14.2–17.0 mPa·s. It did not contain silicone and wax. The density of the used machining fluid was 0.794 g/mL at 20 °C. The same machining fluid was analysed by Chuchala et al. [36] and Staszczyk et al. [37].

2.3. Surface Roughness Measurement

Surface roughness measurements were carried out using a stylus-type device, namely the Hommel Tester T500 surface roughness tester (Hommelwerke GmbH, VS-Schwenningen, Germany). Surface roughness measurements were carried out in the direction of feed movement, which corresponded to the direction parallel to the wood fibres for both species analysed. Measurement parameters were set according to the recommendations of Gurau and Irle [41] as follows: measurement length ln = 12.5 mm, cutoff value λc = 2.5 mm, cutoff ratio λc/λs = 300, sample spacing 1.5 μm. The stylus tip was conical with a cone angle of 60° and a tip radius of 2 μm. Five selected surface roughness parameters were analysed: Ra—arithmetic mean height, Rz—maximum height, Rq—root mean square height, Rt—total height, Rp—mean peak height, according to ISO 21920-1 [42]. When preparing surface roughness measurements, the conclusions from the analyses conducted by Sandak and Negri [43] were taken into account.

2.4. Statistical Analysis

To stabilise the empirical data, a Grubbs test was performed to eliminate outliers. The experimental results are presented as a sample of the central part of the distribution. In order to verify the significance of differences for the analysed surface roughness parameters depending on the feed per revolution and cutting conditions, a variance (ANOVA) analysis was conducted for the significance level 0.05% [44]. Moreover, for statistically significant groups, a post hoc Tukey test (Tukey HSD, Honest Significant Difference) was conducted. To analyse the obtained data, the lower 25th percentile, the median at the 50% level (values indicated in the graphs), and the upper 75th percentile of the data distribution were used, which allows for the assessment of the central tendency of the data and their dispersion. Additionally, the graphs show the minimum and maximum observed values that do not exceed the range of the normal data distribution (whiskers).

3. Results and Discussion

The distribution of the experimental surface roughness data presented above is shown in Figure 2 for pine and in Figure 3 for beech.

The values obtained for the analysed parameters of pine wood surface roughness are similar to those obtained by Chuchala et al. [10] for the frame saw cutting process with similar feed per tooth values. Surface roughness parameter values for beech wood, which are similar to those presented in these studies, were presented in Gurau’s work [45] for the analysis of the sanding process with P160 grain size. However, the Ra and Rz parameter values obtained in this study for beech wood are slightly higher than those presented in Sogutlu’s work [46] for the beech wood planing process, both up-milling and down-milling, with similar feed per tooth values (approximately f_z = 0.2 mm ≈ f₃).

An increase in surface roughness parameter values was observed with increasing feed per revolution. This phenomenon is well-known and widely scientifically proven for a lot of machining processes [10,11,28]. It occurs in all three analysed cases, i.e., under the three different machining conditions: dry, with compressed air, and with the MQL system, as well as for both analysed materials. The mentioned increase in the values of the analysed surface roughness parameters is statistically significant to a considerable extent (

Table 2. The p-value for various surface roughness parameters measured for steam-treated pine wood when turning at different feed per revolution (Tukey HSD test).

Pine Wood	Ra	Rz	Rq	Rt	Rp
	p-Value
f1d, f2d	0.09	0.25	0.10	0.50	0.45
f1d, f3d	0.01	0.001	0.01	0.03	0.001
f2d, f3d	0.99	0.59	0.99	0.92	0.23
f1a, f2a	0.02	0.01	0.03	0.10	0.06
f1a, f3a	0.001	0.001	0.001	0.001	0.001
f2a, f3a	0.02	0.02	0.01	0.01	0.34
f1o, f2o	0.25	0.06	0.12	0.12	0.28
f1o, f3o	0.003	0.001	0.001	0.001	0.03
f2o, f3o	0.77	0.96	0.77	0.87	0.99

d—dry machining, a—machining with compressed air, o—MQL.

and

Table 3. The p-value for various surface roughness parameters measured for steam-treated beech wood when turning at different feed per revolution (Tukey HSD test).

Beech Wood	Ra	Rz	Rq	Rt	Rp
	p-Value
f1d, f2d	0.54	0.31	0.72	0.60	0.91
f1d, f3d	0.02	0.09	0.05	0.25	0.69
f2d, f3d	0.84	1.00	0.86	1.00	1.00
f1a, f2a	1.00	1.00	0.96	0.14	0.37
f1a, f3a	0.10	0.11	0.13	0.58	0.21
f2a, f3a	0.38	0.48	0.75	1.00	1.00
f1o, f2o	0.98	0.97	0.86	0.60	0.99
f1o, f3o	0.002	0.005	0.01	0.003	0.03
f2o, f3o	0.96	0.12	0.39	0.50	0.26

d—dry machining, a—machining with compressed air, o—MQL.

). Table 2 and Table 3 present the analysis for pairs of feed values per revolution. It is worth noting that the use of the MQL system did not change the phenomenon of increasing roughness parameters with increasing feed per revolution. In the case of statistically insignificant differences between the average values of roughness parameters, these occur both for the MQL system and other conditions, e.g., pine wood, f₂–f₃, dry conditions and MQL (Table 2), also beech wood, f₁–f₂, dry condition, compressed air, and MQL (Table 3).

In the case of pine wood, at feeds per revolution f₁ and f₂, the values of the analysed surface roughness parameters for air and MQL machining are similar. As the feed per revolution increases, the surface roughness in air machining becomes higher than in the MQL method. In MQL machining, roughness also increases with feed per revolution, but to a lesser extent. Dry machining is characterised by the highest roughness values, which is particularly evident for the lowest feed rate per revolution f₁ (Figure 2).

However, in the case of beech wood, the lowest values of the analysed surface roughness parameters were observed during dry machining. In contrast to pine, the MQL method provided higher surface roughness values for beech, especially at the highest feed per revolution values tested (Figure 3).

The results of the ANOVA analysis of significance differences are presented in

Table 4. Values of F (F-ratio) and p-value for various analysed surface roughness parameters measured for steamed pine wood when turning at different feed per revolution and different cutting conditions (ANOVA test).

Pine		f1d, f1a, f1o	f2d, f2a, f2o	f3d, f3a, f3o	f1d, f2d, f3d	f1a, f2a, f3a	f1o, f2o, f3o
Ra	F	1.43	0.74	2.97	11.07	16.06	10.97
	p	0.26	0.48	0.07	3.08 × 10−4	2.54 × 10−5	3.27 × 10−4
Rz	F	2.15	0.45	4.60	11.67	18.04	14.19
	p	0.14	0.64	0.02	2.23 × 10−4	1.06 × 10−5	6.15 × 10−5
Rq	F	2.48	1.21	2.52	9.85	18.86	16.56
	p	0.10	0.31	0.10	6.12 × 10−4	7.50 × 10−6	8.40 × 10−5
Rt	F	3.30	1.94	1.65	4.72	18.53	17.85
	p	0.05	0.16	0.21	1.75 × 10−2	8.60 × 10−6	1.15 × 10−5
Rp	F	3.34	1.83	3.46	6.63	22.04	9.34
	p	0.05	0.18	0.05	4.55 × 10−3	2.11 × 10−6	8.26 × 10−4

d—dry machining, a—machining with compressed air, o—MQL.

for pine wood and in

Table 5. Values of F (F-ratio) and p-value for various analysed surface roughness parameters measured for steamed beech wood when turning at different feed per revolution and different cutting conditions (ANOVA test).

Beech		f1d, f1a, f1o	f2d, f2a, f2o	f3d, f3a, f3o	f1d, f2d, f3d	f1a, f2a, f3a	f1o, f2o, f3o
Ra	F	3.76	0.36	1.23	5.79	6.85	7.87
	p	0.04	0.70	0.31	8.10 × 10−3	4.07 × 10−3	2.03 × 10−3
Rz	F	1.84	0.45	1.38	3.74	7.26	8.03
	p	0.18	0.64	0.27	3.68 × 10−2	3.13 × 10−3	1.83 × 10−3
Rq	F	2.10	0.54	0.97	4.16	4.85	8.09
	p	0.14	0.59	0.39	2.67 × 10−2	1.62 × 10−2	1.85 × 10−3
Rt	F	0.14	0.41	2.19	2.08	5.56	15.54
	p	0.87	0.66	0.13	1.45 × 10−1	1.04 × 10−2	5.78 × 10−5
Rp	F	0.31	0.96	1.95	1.73	3.04	8.97
	p	0.73	0.40	0.16	1.97 × 10−1	6.46 × 10−2	1.03 × 10−3

d—dry machining, a—machining with compressed air, o—MQL.

for beech wood. The presented ANOVA analysis of groups of experimental data shows that statistically significant differences occur with the change in feed per revolution. However, the machining conditions, at constant feed per revolution, are statistically insignificant in most cases. These relationships are similar for both analysed wood species: pine and beech. Furthermore, for beech wood, statistical differences can be observed in the cutting conditions only for the first analysed feed per revolution (f₁) and only for the surface roughness parameter (Ra) (Table 5). In the case of pine wood, a similar situation occurs only for the (Rz) parameter for the third analysed value of feed per revolution (f₃) (Table 4).

In the analysis of the influence of various turning conditions, a decrease in the values of the analysed surface roughness parameters was observed when using compressed air cooling and the MQL system. The median values of most of the analysed surface roughness parameters using the MQL system were lower or comparable to the values obtained when machining only with compressed air. Both of the above-mentioned systems allowed for the obtainment of noticeably lower values of the analysed surface roughness parameters compared to dry turning. However, none of these differences were statistically significant.

4. Conclusions

Based on the experimental studies and analysis of the results, the following conclusions can be drawn:

• For steam-treated pine wood, MQL provides lower surface roughness values compared to compressed air and dry turning, especially at higher values of feed per revolution.
• In the case of steam-treated beech wood, the lowest surface roughness values are observed during dry turning, while the use of MQL leads to an increase in the values of surface roughness, especially at high feeds per revolution.
• The use of different turning conditions has little effect on changes in surface roughness values. At the same time, increasing the feed per revolution (0.07–0.28 mm) under the same conditions results in statistically significant differences in surface roughness parameters for both pine and beech.
• In the case of pine wood, statistically significant differences were found between adjacent feed rates (f₁–f₂ and f₂–f₃) during turning with compressed air cooling. Under dry and MQL turning conditions, the differences between adjacent feed per revolution were not statistically significant.
• For beech wood, with dry conditions and compressed air cooling, the changes in surface roughness parameters between feeds per revolution f₁ and f₃ were not statistically significant. However, during MQL machining, significant differences in surface roughness were observed for the same feeds per revolution.