BUDAPEST UNIVERSITY OF TECHNOLOGY AND ECONOMICS

FACULTY OF MECHANICAL ENGINEERING

DEPARTMENT OF MANUFACTURING SCIENCE AND ENGINEERING

ANALYSIS OF MACHINING INDUCED MICRO- AND

MACRO-GEOMETRICAL DAMAGES OF

UNIDIRECTIONAL CARBON FIBRE REINFORCED

POLYMER (UD-CFRP) COMPOSITES, AND THE

DEVELOPMENT OF ITS CHARACTERISTICS

MEASURES

PhD dissertation booklet

AUTHOR: NORBERT GEIER

M.Sc. Mech. Eng.

SUPERVISOR: DR. TIBOR SZALAY

associate professor

BUDAPEST

2019

1

1

TABLE OF CONTENTS

1. Description of the research topic, main objectives.................................................... 2

2. State of the art, research methods ............................................................................. 4

2.1. Machining induced macro-geometric defects .................................................... 4

2.2. Machining induced micro-geometrical defects .................................................. 6

2.3. Research methods .............................................................................................. 7

3. Summary of the research and description of theses .................................................. 7

3.1. Micro-geometry of machined UD-CFRP surfaces ............................................. 7

3.2. Helical milling of UD-CFRP ............................................................................. 9

3.3. Special tool path for edge trimming of UD-CFRP ........................................... 13

4. Exploitation of achievements .................................................................................. 16

5. References .............................................................................................................. 17

6. List of own publications related to the work topic .................................................. 18

The reviews of the doctoral dissertation and the minutes of the defence are available

at the Dean's Office of the Budapest University of Technology and Economics,

Faculty of Mechanical Engineering.

2

1. DESCRIPTION OF THE RESEARCH TOPIC, MAIN OBJECTIVES

Carbon fibre reinforced polymer (CFRP) composite materials are favoured due to

their excellent specific mechanical properties in industries where low weight and high

strength are required. In the aerospace industry, as well as in the automotive, wind

turbine, military, sports, and aerospace industries, manufacturers strive to laminate

CFRP components in a single operation (moulding and curing). they often require

subsequent machining(s) prior to assembly [1]. These may include: (i) removing

material build-up in the dividing plane of the laminating tool, (ii) removing excess

material from the flange of the laminating tool(s), (iii) smoothing the mating surfaces

of the laminated composites, and (iv) drilling the parts. These subsequent machining

needs are typically met with different machining technologies.

However, the machining of CFRP composite materials is complex and expensive:

(i) due to the inhomogeneity and anisotropy of the material, the characteristic

geometrical errors caused by the machining and the chip formation mechanisms are

significantly dependent on the machining directions [2]; (ii) carbon fibres have a strong

abrasive abrasion effect, which should be considered for each component of the MKGS

system; and (iii) heat dissipation is also problematic due to the low thermal conductivity

of polymers and the dangers of using coolant lubricants (polymer wicking) [3]. Because

of these cutting features and conditions, CFRP materials are referred to as difficult-to-

cut, which can result in a variety of micro- and macro-geometric material errors. Such

typical geometrical errors caused by machining are shown in Fig. 1.

Fig. 1 Typical machining hole defects in CFRP materials: (a) delamination (material discontinuity

from laminate layer separation) and matrix burn, (b) uncut fibres and (c) residual powdery chip

shavings, where vector k shows direction of reinforcing fibres

Although centuries of experience in the field of metalworking have been

accumulated, this theoretical and practical knowledge cannot be directly applied to the

engineering of fibre reinforced technical polymer composite materials that have been

researched for decades.

The main objective of my doctoral research was to investigate the machinability of

unidirectional carbon fibre reinforced polymer (UD-CFRP) composite materials, with

special emphasis on the integrity of machined shape properties, which included micro-

and macro-geometric properties of drilled and milled CFRP edges and surfaces:

characteristics of (i) surface roughness and of (ii) uncut fibres. My goal was the

technological optimization of the CFRP machining process, which included (i) specific

measures generation, (ii) process monitoring and diagnostics.

3

The objectives of my doctoral research were specified in the following points:

(i) The surface roughness depth (Rz/Ra) specific for arithmetic surface roughness in a

UD-CFRP composite material is spread over a wider range than for quasi-

homogeneous materials. The local high-roughness peaks caused by reinforcing and

bonding fibres (hereinafter referred to as "uncut fibres") that remain on the

machined surface cause high values of the Rz/Ra ratio, which average value depends

on the cutting technology: Rz/Ra=4-14 (μm/μm). The degree of deviation of the

Rz/Ra metric from a target value can be used to decide whether or not the machined

surface needs additional operations.

(ii) I developed (a) a parameter describing the amount of uncut fibres of a machined

through hole in a UD-CFRP composite material, and (b) its measurement

methodology, which characterizes the holes with good reproducibility.

(iii) During helical milling of UD-CFRP, the amount of uncut fibres is significantly

dependent on the abrasion state of the cutting edge and the extent of helical tool

path pitch. As the abrasion condition deteriorates and the thread pitch increases, the

amount of uncut fibres increases. By systematically influencing these two

parameters between helical milling operation elements (by the process monitoring

and diagnostic algorithm I developed) the amount of uncut fibres can be

significantly reduced.

(iv) During edge trimming of UD-CFRP, the fibre orientation angle significantly

influences the cutting force. The cutting force as a function of the orientation angle

has a minimum when using both milling strategies (climb and conventional miling).

The minimum cutting force location for climb and conventional milling is not the

same.

(v) I developed a parameter describing the milled edge-specific amount of uncut fibres

remaining at the outer edges of the UD-CFRP composite.

(vi) I have developed a novel tool path system for contour milling of UD-CFRP

composites, which uses less uncut fibre left on the milled edge than traditional

contour parallel milling technology.

CFRP cutting experiments, investigations and technological optimizations were

designed to describe, estimate, influence and minimize micro- and macro-geometric

errors caused by machining.

4

2. STATE OF THE ART, RESEARCH METHODS

In this section I present a (i) literature review of the micro- and macro-geometric

errors of machined shape features closely related to my doctoral research and the (ii)

methods used in my research.

2.1. Machining induced macro-geometric defects Macro-geometric defects caused by machining include quality parameters

characteristic of the machined shape properties that characterize the shape and / or size

deviations from the nominal geometry (e.g. burr, delamination). During my scientific

research, I analysed the characteristics of the uncut reinforcing and bundling fibres that

were formed during the cutting of CFRP and which were not cut at the nominal location.

In the case of improper machining conditions and features, the machined CFRP

may have excess material left over. While reinforcing and / or bundling fibres remaining

on the cut edges or surfaces, not cut at the nominal location (hereinafter referred to as

"uncut fibres"), do not directly reduce the inherent mechanical properties of the

component, they may require post-machining, which involves considerable time and

cost [4].

Fig. 2 Uncut fibres remaining after cutting UD-CFRP: (a) along the circumference of a hole [5],

(b) at the edge milled with conventional milling using "A" technology [6] and (c) at the edge milled

with straight milling technology "B" [7]. The "A" and "B" technologies and the CFRP matrix

material were also different.

Uncut fibres, often visible to the eye (Fig. 2), are typically formed at the machined

edges, where the reinforcing and bundling fibres are easily bent by the cutting tool

rather than being cut. Factors that significantly influence the degree of bending of the

fibres are: (i) fibre cutting angle [4], (ii) tool edge geometry (edge radius, rake angle,

clearance angle) [2], (iii) presence of support material relative to the fibres bending

direction [8], (iv) the wear condition of the cutting tool [4], (v) the feed rate [9], and

(vi) the milling direction (conventional or climb milling) [6]. Xu et al. [9] observed that,

when drilling UD-CFRP, uncut fibres exhibit certain characteristics along the

circumference of the hole: two burr zones (S1 and S2) were designated, in which zones

the amount of uncut fibres was significant, as shown in Fig. 2 (a). The location of these

zones is significantly dependent on the fibre cutting angle (θ) around the borehole and

typically occurs at a nominal position θ=135°± δ ° (Type IV chip formation mechanism)

[9].

5

Uncut fibres are examined by digital image processing of optical microscope

images [4, 9, 10]. The characteristics of the uncut fibres are summarized in Table 1

below.

Table 1 Metrics commonly used in the literature to characterize uncut fibres

Reference Measures Equation Nomenclature

Slamani et al. [11]

Wang et al. [4]

Maximal high of

uncut fibres

x n – number of measurements

A0 – nominal hole area

Si – burr area l – machined edge length

Geier et al. [7] Average maximal high of uncut fibres

X=n-1∑xi

Hrechuk et al. [10]

Maximal length of uncut fibres

lh

Hrechuk et al.

[10]

Average maximal

length of uncut fibres

L=n-1∑lhi

Hrechuk et al.

[10]

Average width of

uncut fibres

W=n-1∑wi

Xu et al. [9] Burr area S=S1+S2

Geier et al. [12]

Xu et al. [13] Hrechuk et al.

[10]

Specific area of

uncut fibres

α=S/A0

A=S/l

Measurements on length (lh, L), width (w, W), height (x, X) or area (S) of uncut

fibres are most commonly used by researchers to describe and classify the quality of

the cut edges. Measuring and / or calculating these geometric features is quick and easy,

but using these metrics does not create clear grading and grading methods that can

define bore diameters and / or cut edge lengths independent of cut length (e.g. S=2 mm2

burr zone is significant for a hole with a diameter of d=3 mm but negligible for a hole

with a diameter of d=100 mm). Therefore, it is necessary to specialize the listed metrics

into a certain geometric feature:

(i) Specializing the area of the uncut fibres to the nominal hole diameter (α) for

the cut hole.

(ii) In the case of a cut edge, specification of the area of the uncut fibres to the

cut length (A).

With these specific metrics, the percentage of uncut fibres can be estimated and

compared to the relative amount of uncut fibres left after cutting shapes with different

geometric dimensions.

The influence of cutting technology process parameters on the characteristics of

uncut fibres has already been investigated by many researchers in order to consciously

influence and minimize their quantity. Ramirez et al. [14] observed that the amount of

uncut fibres is maximal at θ=135°, and they also proposed a novel measure for the

specific classification of uncut fibres by specifying the area of uncut fibres on the

surface of the entire bore. Xu et al. [15] demonstrated that the axial cutting force

component significantly influences the expansion of the burr zone by increasing the

6

burr zone size. Heisel et al. [16] investigated the effect of feed rate on burr height during

CFRP drilling experiments with various angular auger drills. It has been observed that

increasing the feed rate also increases the height of the uncut fibres at the leading edges

of the holes, but no clear relationship has been found between the amount of uncut

fibres remaining at the exit edge of the hole and the feed rate.

Hrechuk et al. [10] proposed a novel combined quality parameter that takes into

account delamination in addition to the amount of uncut fibres. Poulachon et al. [17]

conducted drilling experiments with uncoated carbide drilling tools in CFRP composite

materials and observed that: (i) with novel sharp tools, the amount and distribution of

uncut fibres mainly depends on the cutting angle, and (ii) with increasing tool wear,

decreases to the amount of uncut fibres. Wang et al. [18] carried out drilling

experiments with four different drilling tools (coated and uncoated: twist and double

point angle drills) and observed that the bore zone size is smaller for uncoated drill bits

than for drilled bits, while the tool wear rate was minimal.

2.2. Machining induced micro-geometrical defects Micro-geometric defects caused by machining of CFRP materials include

deviations from the nominal geometry that can be determined from the measurements

of the surface roughness profile.

Structural engineers most often use standard surface roughness parameters (Ra, Rz,

Rt, etc.) to determine the maximum roughness properties of a surface (whether cast,

rolled, drawn, laminated, chipped, etc.). These metrics, measurement conditions and

methods are detailed in DIN EN ISO 4287:1998 and DIN EN ISO 4288:1998. However,

the measures and methods of these standards have evolved on the basis of extensive

knowledge and experience in metal technology and their application is limited to the

surface roughness characterization of non-metallic materials [17]. Thus, in my research

I paid special attention to the examination of a composite roughness parameter, which

is the amount of specific roughness height (Rz) calculated for a given measurement

length, standardized roughness height (Ra) (Rz/Ra). For quasi-homogeneous and quasi-

isotropic materials, this ratio is significantly dependent on the machining mode (e.g.

turning, milling, grinding, rolling, etc.) and has an average value of Rz/Ra=4-7 μm/μm.

A high Rz/Ra value indicates that the surface has high local roughness peaks and/or deep

roughness valleys relative to the average surface roughness parameter. In the case of

fibre reinforced polymer composites, uncut fibres or stretched fibres can cause such

local roughness peaks and valleys, which are one of the focal points of my research.

The surface roughness of CFRP composites is typically evaluated by researchers

using the standard Ra, Rz, Rt parameters. The surface roughness of CFRP has been

investigated by analysis of variance by Gao et al. [19]. It was shown that machined

surface roughness is most influenced by the cutting angle, followed by cutting speed,

depth of cut, and then the rake angle. Surface roughness of conventionally drilled holes

in CFRP has been investigated by Poulachon et al. [17]. It has been observed that the

surface roughness characteristics in the range of θ=135 ± 18 ° filament angles are worse

than those of the outer surface measured elsewhere. Poulachon et al. as a result of their

scientific work, they suggested that a novel roughness index should be used to evaluate

the surface roughness of CFRP, which has a higher weight and physical sense due to

uncut fibres [17].

7

2.3. Research methods The experiments to achieve the aims of my PhD research were designed using full

factorial and central composite inscribed experimental design methods. The factors,

optimization parameters, factors and variation intervals of the studied cutting processes

were determined on the basis of a priori information. From the measured data, the

optimization parameters were generated by (i) digital image processing, (ii) fast Fourier

transformation and (iii) frequency filtering, which were investigated by (i) regression

analysis, (ii) response surface methodology and (iii) analysis of variance. I supported

the systematic design and evaluation tasks with Minitab, AutoCAD, Microsoft Excel,

Wolfram Mathematica, Stream Essentials, Gwyddion, IrfanView, LabView and the

DoDEx software I developed.

3. SUMMARY OF THE RESEARCH AND DESCRIPTION OF THESES

3.1. Micro-geometry of machined UD-CFRP surfaces I drilled holes in a UD-CFRP composite on a Kondia B640 three-axis, vertical

spindle NCT100 controlled machine using traditional drilling and helical milling

techniques. I removed the chip from the cutting space with an NILFISK GB733 (p=

14.7 kPa) industrial vacuum cleaner. For drilling experiments, I used a Ø11,138 SECO

SD205A-11.138-53-12R1-C1 double-ended diamond-coated drill bit and an Ø10

TIVOLY 8236651 1000 single-edged carbide end mill. The roughness tests of the

circumferential surface of the machined holes were performed with a Mitutoyo SJ-400

contact profilometer using a cut-off λc = 2.5 mm and a feed rate v = 1 mm/s. Surface

roughness was calculated according to DIN EN ISO 4288:1998. The applicability of

the diamond-pointed contact profilometer was investigated for the roughness test of

machined CFRP surfaces.

The factors and their levels are shown in Table 2. In the conventional drilling

experiments, I also examined the effect of cutting speed (vc) and feed rate (f), while in

the case of helical strategy I also examined the effect of helical tool path pitch (h).

Table 2 Factors and their levels

Tools Factors Levels

-2n/4 -1 0 +1 +2n/4

Twist drill

n=2

Cutting speed (m/min) 50 65 100 135 150

Feed rate (mm/rot) 0.035 0.043 0.064 0.078 0.093

End mill n=3

Cutting speed (m/min) 50 70 100 130 150

Feed rate (mm/rot) 0.020 0.028 0.040 0.051 0.060

Screw pitch (mm) 0.10 0.68 1.55 2.41 3.00

The cutting tools used for the milling experiments were: (i) Ø10 SECO 871100.0-

DURA diamond-coated compression end mill, (ii) Ø10 TIVOLY 82366511000 single-

edged carbide milling cutter, (iii) Ø10 FRAISA 20340.450 coarse tooth Ø10.4 and (iv)

coarse toothed carbide compression end mills. All four tools have been subjected to

conventional milling experiments with ϕ=0°/180° filament parallel tool paths with fixed

technological parameters: vc=160 m/min, vf=1528 mm/min, ap=15 mm, ae=5 mm, climb

milling, dry machining. The surface roughness parameters measured with the Mitutoyo

8

contact profilometer were validated with an ALICONA Infinite Focus type confocal

microscope.

Fig. 3 summarizes the Rz/Ra parameters for (i) conventional drilling strategy and

(ii) helical milling strategy. The diagram shows that the average Rz/Ra= 8.53 μm/μm for

conventional drilling and Rz/Ra= 11.13 μm/μm for helical milling. These ratios are

averaged from n=65 (conventional drill) and n=94 (helical milling) test results. Further

evidence of the presence of residual uncut fibres is provided by the uncut reinforcing

and bundling fibres shown in the optical microscope images shown in Fig. 4.

Fig. 3 Rz/Ra parameter based on roughness data measured on the perimeter surface of holes made

with conventional drilling and helical milling strategy

Fig. 4 Images taken with an Olympus optical microscope: (a) SECO 871100.0-DURA (b)

TIVOLY 82366511000 (c) FRAISA 20340.450 és (d) FRAISA 20360.450 tools

Based on the surface roughness characteristics of (i) conventional drilling, (ii)

helical milling, and (iii) contour milling experiments in the UD-CFRP composite, the

following new scientific statement was declared:

9

Thesis 1: The Rz/Ra parameter of a surface machined in a unidirectional carbon

fibre reinforced polymer (UD-CFRP) composite material is scattered over a wider

range than in the case of quasi-homogeneous materials. The local high height

peaks caused by reinforcing and bundling fibres not cut at the nominal depth on

the cut surface cause a high value of the Rz/Ra ratio, which may be Rz/Ra= 4-14

μm/μm, depending on the machining technology. It is recommended to supplement

the study of the machinability of UD-CFRP composite materials by studying the Rz/Ra

parameter.

Own publications related to the first thesis: [S1–S7]

3.2. Helical milling of UD-CFRP Helical milling technology was used to drill holes into the UD-CFRP composite in

the experimental environment described in the previous section. For helical milling

experiments I used a Ø8 TIVOLY 82366510800 single-edged carbide end mill and a

Ø8 TIVOLY 80308810808 single-edged HSS end mill. The edges of the tools were

photographed with a Dino-Lite AM4013MT digital microscope with magnification

xn=25x from direction of +Z.

The characteristics of the uncut fibres were investigated by digital image

processing. During my doctoral research I divided the digital images of the machined

features into grey and black pixels using a grey histogram (Fig. 5).

Fig. 5 The main steps of digital image processing are: (a) a hole-lighted digital image of a hole,

(b) a segmented digital image of the hole, (c) the measurement setup

I used segmented images for later processing, in which (i) the number of white

pixels was counted, and then (ii) specified for the pixels of the nominal hole diameter.

The resultant α quantitative dimensionless dimension can be used to characterize the

quality of the cut holes in terms of uncut fibres. If the area factor is α=1, there are no

reinforcing and/or bundling fibres in the circumference of the machined bore that are

within the nominal bore contour.

With fixed technological parameters (vc=75 m/min, f=0.08 mm/rpm, h=3 mm,

climb milling), I drilled holes with helical milling with the single-edged HSS mill and

investigated the effect of cutting length on (i) tool wear and (ii) the area factor. I studied

tool wear with a digital image processing method similar to the area factor.

Experimental results showed that the number of machined holes (i) increases the rate

of tool wear and (ii) decreases the value of the area factor. This phenomenon was to be

expected, because as the tool edge radius increases (tool wear), the thread cutting

10

mechanism becomes a "bending dominant", so that the fibres are not cut at the nominal

position but bent, thereby increasing the value of the area factor.

In a further series of experiments, I investigated the effect of helical tool pitch (h)

on the amount of uncut fibres. The value of the technological parameters was recorded

so that the value of the optimization parameter did not change: vc= 75 m/min, vf= 0.08

mm/min, climb milling. The thread pitch was examined on three levels (h= 1; 2; 3 mm),

and the experiments were repeated five times on each level. In the examined thread

pitch range it was clearly observed that as the thread pitch increased, the value of the

area factor decreased, and the quality of the machined bores decreased.

Based on the above experimental results and observations, the effect of the

following two factors is significant on the quality of the machined bores: (i) helical tool

path pitch (h) and (ii) milling depth (z). Smaller h results in lower axial cutting force,

less delamination, less uncut fibre and lower tool wear rate. Furthermore, if the

programmed point of the tool is sent deeper than would be justified by the conventional

approach, then the non-worn sections of the edge strip are more likely to cut the

reinforcing and braiding strands than the already worn lower edge sections. I have

developed an algorithm based on the systematic change of these parameters, the quality

of the holes drilled by the algorithm is better than that of the fixed technological

parameters, as can be seen in Fig. 6.

Fig. 6 Area factor for fixed parameter helical milling and process controlled milling

11

Helical milling experiments were performed on UD-CFRP composite with (i) fixed

technological parameters, (ii) systematic alteration of helical tool path pitch and (iii)

milling process monitoring, which resulted in the following new scientific findings:

Thesis 2: The amount of reinforcing and bundling fibres remaining at the edges of

a through hole drilled in a unidirectional carbon fibre reinforced polymer (UD-

CFRP) composite material is characterized by the following parameter:

𝜶 =𝟏

𝒏∑

𝑨𝒊

𝑨𝟎

𝒏

𝒊=𝟏

where α (-) is the area factor, n (-) is the number of repeated experimental settings, Ai

(pixels) is the number of white pixels calculated from the digital image taken during the

ith experiment, A0 (pixels) is the white pixels calculated from the nominal hole number.

The area factor can be determined by the following method: (i) Lower illumination of

the cut hole by diffused LED light source. (ii) Take a digital image of the hole opposite

to the light source. The nominal axis of the hole should be parallel to the optical axis of

the camera unit. Only one hole may be included in the recording. (iii) Convert a colour

image to a grayscale image. (iv) Segmentation of the image based on the grey colour

spectrum. (v) Determining the number of white pixels, and then (vi) specifying the

number of white pixels for the number of white pixels per hole of nominal geometric

size. The measuring method is suitable for the quantitative evaluation of uncut fibres

with a p= 1,2% confidence.

Own publications related to the second thesis: [S1–S3, S8–S14]

12

Thesis 3: The amount of reinforcing and bonding fibres (hereinafter referred to

as "uncut fibres") remaining at the edges of the machined hole during helical

milling of a unidirectional carbon fibre reinforced polymer (UD-CFRP)

significantly depends on the wear condition of the cutting edge and the screw pitch

of the helical tool path. With the deterioration of the abrasion state and the

increase of the thread pitch (h) the amount of uncut fibres in the examined factor

space (h= 1-3 mm) increases. By systematically influencing these two parameters

between helical milling operation elements, the amount of uncut fibres can be

significantly reduced by the following algorithm:

where α (-) is the area factor, Ai (pixels) is the number of white pixels calculated on the

digital image taken in experiment no. i, A0 (pixels) is the number of white pixels

calculated from the nominal hole digital image, α1 (-), α2 (-) and α3 (-) context-dependent

constants, B (-) denotes the number of holes to be made, T (-) denotes the number of

available tools, h (mm) denotes the helical tool path pitch, z (mm) denotes the

machining depth, I (mm) denotes the variation interval, index 0 denotes the initial value,

while max and min indices represent the highest and lowest values, respectively.

Own publications related to the third thesis: [S1 – S3, S5, S9 – S12, S14 – S16]

13

In order to validate the second and third theses, I performed repeated experiments

under the same experimental conditions, but with technological parameters of vc= 150

m/min and f= 0.06 mm/rot., which gave the same results.

3.3. Special tool path for edge trimming of UD-CFRP Milling experiments were performed in UD-CFRP composite on a VF 22 type

vertical spindle milling machine. I removed the chip from the cutting zone with the

NILFISK GB733 industrial vacuum cleaner. For milling experiments, I used a high-

speed steel end mill with D= Ø50 mm diameter, zf= 5 cutting edges (αo = 10°, γo= 25°).

The cutting force was measured with a KISTLER 9281 Type B three-component force

measuring cell and collected using a Labview measuring program at a sampling

frequency of fm= 18 kHz for t= 10 s per experimental setting. Frequency filtering was

applied at the measured force values with noise using a low pass filter with a cut-off

frequency of fh= 400 Hz. During the evaluation, the models were adequately examined

by regression analysis. The experimental results are illustrated in Fig. 7.

Fig. 7 Cutting force in CFRP milling as a function of strand orientation angle, in the case of

conventional and climb milling

Az UD-CFRP kompoziton végzett egyen- és ellenirányú palástmarási kísérletek

során a szálirány hatását vizsgáltam a forgácsolási erő egyes komponenseire, mely

alapján a következő új tudományos megállapítást fogalmaztam meg:

In the UD-CFRP composite experiments, the effect of the fibre direction on the

individual components of the cutting force was investigated, based on the following

new scientific findings:

Thesis 4: The analysis of the cutting force function F(ϕ)=[Ff (ϕ),Fr(ϕ),Fp(ϕ)] ,

ϕ∈[0°; 180°) for the conventional and climb milling of unidirectional carbon fibre

reinforced polymer (UD-CFRP) composite shows that:

(i) During UD-CFRP milling, the fibre orientation angle significantly influences

the cutting force. The cutting force has a minimum in both milling strategies.

14

(ii) The minimum cutting force location is not the same for the conventional and

the climb milling: ϕopt= 156° (+26°, -33°) for conventional milling and ϕopt=

123 °(+21°, -23°) for climb milling.

where F (N) is the cutting force, f is the feed direction, r is the radial direction, p is the

passive direction and ϕ (°) is the fibre orientation angle.

Own publications related to Thesis 4: [S2, S6, S8, S12, S17–S25]

Cutting at the force minimum is a cardinal task for polymer matrix composites,

since (i) reducing the rate of tool wear, (ii) reducing heat generation due to frictional

forces at the cutting edges (matrix burn) and (iii) reducing delamination [20].

When designing the movement path (toolpath) for cutting force-optimized milling

of quasi-homogeneous materials, it is advisable to optimize the chip cross-section

primarily [21, 22]. In the case of UD-CFRP composite, due to anisotropy, the direction

of the reinforcing fibres significantly influences the milling paths optimized for the

force minimum, so it is advisable to take this into account when milling UD-CFRP.

Further coat milling experiments were performed on UD-CFRP composite on

Kondia B640 machine tool. I removed the chip from the cutting zone with the NILFISK

GB733 industrial vacuum cleaner. For milling experiments, I used special CFRP

composite milling cutters: (i) Ø10 FRAISA 20340.450 coarse tooth and (ii) Ø10

FRAISA 20360.450 medium tooth carbide compression end mills. The roughness tests

of the machined peripheral surfaces were performed with a Mitutoyo SJ-400 contact

profilometer using λc= 2.5 mm cut-off and v= 1 mm/s feed rate. The values of the

technological parameters were fixed so that the optimization parameter under their

influence does not change: ϕ= 180°, vc= 160 m/min, vf= 1528 mm/min, ap= 15 mm, ae=

5 mm, unidirectional milling. The edges were photographed with a Dino-Lite

AM4013MT digital microscope at xn= 70x magnification, from the direction of +Z.

During the experiments I investigated the effect of two categories factors (x1: tool type

and x2: tool path type) on the characteristics of uncut fibres. The two tool path types are

illustrated in Fig. 8.

Fig. 8 Milling tool paths tested in the experiments: (a) ϕ= 180°, conventional contour milling and

ϕ= 100°, trochoidal tool path, (b) trochoidal tool path sections

With conventional contour milling (Fig. 8 (a)), the unfavourable technological

condition (ϕ= 180°) is met. The basic idea of the special trochoidal toolpath (Fig. 8 (b))

15

is that the thread orientation angle in the main cutting path section is ϕ= 100°, at which

angle the cutting conditions can be expected based on preliminary experimental results.

The images of the milled edges were analysed by digital image processing. The

preparation of the study was carried out in the main steps shown in Fig. 9 (segmentation,

cutting). The processed images of milled edges (Fig. 10) clearly show that the amount

of uncut fibres remaining on special milled edges is minimal.

Fig. 9 Major steps in preparing digital images for image processing. (a) original image (b)

segmented image (c) segmented image narrowed to evaluation area

Fig. 10 Binary images of UD-CFRP composite edges milled with conventional and special tool

paths. (a) conventional toolpath (ϕ= 180°) with coarse-toothed compression end mill, (b) special

toolpath (ϕ= 100°) with coarse-toothed mill, (c) conventional toolpath (ϕ= 180°) with medium-

toothed mill, (d) special tool path (ϕ= 100°) with medium toothed mill

Based on the measurement method and results described in this chapter, I have

made the following new scientific findings:

Thesis 5: The quantity of reinforcing and braiding fibres (hereinafter referred to

as "uncut fibres") left on the cut edges, which are not cut at the nominal location,

is characterized by a quantitative optimization parameter for unidirectional

carbon fibre reinforced polymer (UD-CFRP):

𝐴𝑏 = 𝑐1

𝑙

1

𝑛∑𝑝𝑚𝑖

𝑛

𝑖=1

=1

𝑙

1

𝑛∑𝑎𝑏𝑖

𝑛

𝑖=1

where Ab (mm2/m) is the area of uncut fibres specific to milled edge length, c

(mm2/pixel) is the area specific to pixel score, l (m) milled edge length, n (-) number of

16

windows, pm (pixel) measured pixels number of pieces, ab=cpm (mm2) area of black

pixels in a window. Measure Ab shows the projection area of the reinforcing and

braiding fibres not perpendicular to the measuring direction per unit milled edge. The

ab parameter can be determined by the following method: (i) diffused illumination of a

cut edge and a solid colour background behind the deburring edge. (ii) Make n digital

images of the milled edge, with equal distance between the images. The optical axis of

the camera unit should be parallel to the normal vector of the mantle plate. (iii) Cutting

images along the nominal milled edge. Only the uncut fibres should be visible on the

cut. (iv) Convert trimmed images to grayscale. (v) Segmentation of recordings based

on grey colour spectrum. (vi) Determining the number of black pixels, and then (vii)

specifying the number of black pixels per unit edge length.

Own publications related to Thesis 5: [S2, S17, S20, S22, S24, S26]

Thesis 6: The amount of reinforcing and bonding fibres remaining in the

unidirectional milling of unidirectional carbon fibre reinforced polymer (UD-

CFRP) composite can be minimized by a trochoidal toolpath that has the feed rate

vector of the main section of the climb milling system:

𝑣𝑓(𝛽, 𝐾, 𝜙) = 𝑣𝑓 [𝑐𝑜𝑠(𝛽 + 𝐾 + 𝜙)𝑠𝑖𝑛(𝛽 + 𝐾 + 𝜙)

] , 𝜙𝜖[0; 180)

where vf (mm/min) is the feed rate, β (°) represents the angle of the workpiece to be

milled with the X axis of the workpiece datum, K (°) represents the angle of the

workpiece reinforcing fibres to be milled, and ϕ (°). The value of vf must be selected

for a given tool type (geometry, material, coating, condition), while ϕ=123°(+21°,-23°).

Own publications related to Thesis 6: [S2, S17, S20, S22, S24, S27]

4. EXPLOITATION OF ACHIEVEMENTS

For companies cutting large quantities and/or high quality carbon fibre reinforced

polymer composite materials, optimizing the cutting processes of their products is

crucial. My research results help in this endeavour, because with the measures and

methods I recommend, it is possible to characterize, evaluate and compare the

individual cut features with good reproducibility indices. With the help of these metrics

I have developed a direct process monitoring and diagnostic method based on digital

image processing, which can produce drills of significantly better quality than with

fixed technological parameters.

By experimental work I have proved that using the new contour milling (non-

parallel) tool paths, number of uncut fibres can be drastically minimised, which can be

used by the industry, however, it needs further optimisation.

17

5. REFERENCES

[1] S. Vigneshwaran, M. Uthayakumar, and V. Arumugaprabu, “Review on

Machinability of Fiber Reinforced Polymers: A Drilling Approach,” Silicon, vol.

10, no. 5, pp. 2295–2305, Sep. 2018.

[2] J. Ahmad, Machining of Polymer Composites. Springer US, 2009.

[3] V. Krishnaraj, R. Zitoune, and J. P. Davim, Drilling of Polymer-Matrix

Composites. Berlin, Heidelberg: Springer Berlin Heidelberg, 2013.

[4] F. Wang, J. Yin, J. Ma, Z. Jia, F. Yang, and B. Niu, “Effects of cutting edge radius

and fiber cutting angle on the cutting-induced surface damage in machining of

unidirectional CFRP composite laminates,” Int. J. Adv. Manuf. Technol., vol. 91,

no. 9–12, pp. 3107–3120, Aug. 2017.

[5] N. Ibriksz and N. Geier, “Analysis of uncut fibres at machined holes in carbon

fibre-reinforced plastics (CFRP) using digital image processing,” Bánki Rep., vol.

1, no. 3, pp. 11–14, 2018.

[6] M. Li, M. Huang, X. Jiang, C. Kuo, and X. Yang, “Study on burr occurrence and

surface integrity during slot milling of multidirectional and plain woven CFRPs,”

Int. J. Adv. Manuf. Technol., vol. 97, no. 1–4, pp. 163–173, Jul. 2018.

[7] N. Geier, T. Szalay, and I. Biró, “Trochoid milling of carbon fibre-reinforced

plastics (CFRP),” Procedia CIRP, vol. 77, pp. 375–378, Jan. 2018.

[8] J. C. Aurich, D. Dornfeld, P. J. Arrazola, V. Franke, L. Leitz, and S. Min, “Burrs—

Analysis, control and removal,” CIRP Ann., vol. 58, no. 2, pp. 519–542, Jan. 2009.

[9] J. Xu, Q. An, and M. Chen, “A comparative evaluation of polycrystalline diamond

drills in drilling high-strength T800S/250F CFRP,” Compos. Struct., vol. 117, pp.

71–82, Nov. 2014.

[10] A. Hrechuk, V. Bushlya, and J.-E. Ståhl, “Hole-quality evaluation in drilling fiber-

reinforced composites,” Compos. Struct., vol. 204, pp. 378–387, Nov. 2018.

[11] M. Slamani, S. Gauthier, and J.-F. Chatelain, “Comparison of surface roughness

quality obtained by high speed CNC trimming and high speed robotic trimming

for CFRP laminate,” Robot. Comput.-Integr. Manuf., vol. 42, pp. 63–72, Dec.

2016.

[12] N. Geier, T. Szalay, and M. Takács, “Analysis of thrust force and characteristics

of uncut fibres at non-conventional oriented drilling of unidirectional carbon fibre-

reinforced plastic (UD-CFRP) composite laminates,” Int. J. Adv. Manuf. Technol.,

vol. 100, no. 9–12, pp. 3139–3154, Oct. 2018.

[13] J. Xu, C. Li, S. Mi, Q. An, and M. Chen, “Study of drilling-induced defects for

CFRP composites using new criteria,” Compos. Struct., vol. 201, pp. 1076–1087,

Oct. 2018.

[14] C. Ramirez, G. Poulachon, F. Rossi, and R. M’Saoubi, “Tool Wear Monitoring

and Hole Surface Quality During CFRP Drilling,” Procedia CIRP, vol. 13, pp.

163–168, Jan. 2014.

[15] J. Xu, Q. An, X. Cai, and M. Chen, “Drilling machinability evaluation on new

developed high-strength T800S/250F CFRP laminates,” Int. J. Precis. Eng.

Manuf., vol. 14, no. 10, pp. 1687–1696, Oct. 2013.

[16] U. Heisel and T. Pfeifroth, “Influence of Point Angle on Drill Hole Quality and

Machining Forces When Drilling CFRP,” Procedia CIRP, vol. 1, pp. 471–476,

Jan. 2012.

18

[17] G. Poulachon, J. Outeiro, C. Ramirez, V. André, and G. Abrivard, “Hole Surface

Topography and Tool Wear in CFRP Drilling,” Procedia CIRP, vol. 45, pp. 35–

38, Jan. 2016.

[18] X. Wang, X. Shen, C. Zeng, and F. Sun, “Combined influences of tool shape and

as-deposited diamond film on cutting performance of drills for CFRP machining,”

Surf. Coat. Technol., vol. 347, pp. 390–397, Aug. 2018.

[19] C. Gao, J. Xiao, J. Xu, and Y. Ke, “Factor analysis of machining parameters of

fiber-reinforced polymer composites based on finite element simulation with

experimental investigation,” Int. J. Adv. Manuf. Technol., vol. 83, no. 5, pp. 1113–

1125, Mar. 2016.

[20] F. Girot, F. Dau, and M. E. Gutiérrez-Orrantia, “New analytical model for

delamination of CFRP during drilling,” J. Mater. Process. Technol., vol. 240, pp.

332–343, Feb. 2017.

[21] A. Jacso and T. Szalay, “Analysing and Optimizing 2.5D Circular Pocket

Machining Strategies,” in Advances in Manufacturing, 2018, pp. 355–364.

[22] A. Jacso, T. Szalay, J. C. Jauregui, and J. R. Resendiz, “A discrete simulation-

based algorithm for the technological investigation of 2.5D milling operations,”

Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci., vol. 233, no. 1, pp. 78–90, Jan.

2019.

6. LIST OF OWN PUBLICATIONS RELATED TO THE WORK TOPIC

[S1] N. Geier és T. Szalay: Optimisation of process parameters for the orbital and

conventional drilling of uni-directional carbon fibre-reinforced polymers (UD-

CFRP). Measurement, vol. 110, pp. 319–334, 2017. (IF.: 2,359)

[S2] N. Geier, J. Paulo Davim és T. Szalay: Advanced cutting tools and technologies

for drilling carbon fibre reinforced polymer (CFRP) composites: a review.

Compos. Part A, vol. 125, p. 105552, 2019. (IF.: 6,282)

[S3] N. Geier és Gy. Mátyási: Machinability Study of Unidirectional CFRP Using

Central Composite Design of Experiments. Óbuda Univ. E-Bull., vol. 6, no. 1,

pp. 21–27, 2016.

[S4] N. Geier és Cs. Pereszlai: Analysis of characteristics of surface roughness of

machined CFRP composites. Period. Polytech., (benyújtva, bírálat alatt).

[S5] Cs. Pereszlai és N. Geier: Szénszál erősítésű polimer (CFRP) kompozit

anyagok speciális fúró szerszámainak áttekintése. OGÉT Konf., pp. 428-431.,

2019.

[S6] N. Pálfi és N. Geier: Teljes faktoriális és central composite kísérlettervvel nyert

információk elemzése, optimumkeresés. OGÉT 2017 Konf., pp. 299-302., 2017.

[S7] N. Geier és Gy. Mátyási: Egyirányú CFRP forgácsolhatósági vizsgálata

frakcionális faktoriális kísérlettervvel. OGÉT 2016 Konf., pp. 143-146., 2016.

[S8] N. Geier, T. Szalay és M. Takács: Analysis of thrust force and characteristics

of uncut fibres at non-conventional oriented drilling of unidirectional carbon

fibre-reinforced plastic (UD-CFRP) composite laminates. Int. J. Adv. Manuf.

Technol., vol. 100, no. 9–12, pp. 3139–3154, Oct. 2018. (IF.: 2,601)

[S9] N. Geier, Gy. Póka és T. Szalay: Direct monitoring of hole damage in carbon

fibre-reinforced polymer (CFRP) composites. IOP Conf. Ser. Mater. Sci. Eng.,

vol. 448, no. 1, p. 012003, 2018.

19

[S10] N. Ibriksz és N. Geier: Analysis of uncut fibres at machined holes in carbon

fibre-reinforced plastics (CFRP) using digital image processing. Bánki Rep.,

vol. 1, no. 3, pp. 11–14, 2018.

[S11] Cs. Pereszlai és N. Geier: A comparative analysis of wobble milling, helical

milling and conventional drilling of CFRP. Int. J. Adv. Manuf. Technol.,

(benyújtva, állapot: major revision). (IF.: 2,601)

[S12] N. Geier, T. Szalay és Gy. Mátyási: A comparative experimental study of the

machinability of UD-CFRP using right-hand-cut and left-hand-cut end mills.

Int. Conf. Innov. Technol. -TECH, pp. 221-224., 2016.

[S13] D. Poór, N. Geier, Cs. Pereszlai és N. Forintos: A pilot experimental research

on drilling of CFRP under tensile stress. Int. Conf. Innov. Technol. -TECH

2019, (benyújtva, elfogadva).

[S14] N. Geier, G. Póka és Cs. Pereszlai: Monitoring of orbital drilling process in

CFRP based on digital image processing of characteristics of uncut fibres.

Procedia CIRP, (benyújtva, minor revision).

[S15] N. Geier és I. Gyurika: Development of parametric CNC program and decision

algorithm into product line made from natural stones. IESB 2014 Conf., 2014.

[S16] N. Ibriksz és N. Geier: Szénszál erősítésű polimer (CFRP) forgácsolhatósági

vizsgálata különböző furatkészítési stratégiákkal. OGÉT, pp. 216–219, 2018.

[S17] B. Somoskői és N. Geier: Machinability analysis of carbon fibre-reinforced

plastics (CFRP) using compression tools. Bánki Rep., vol. 1, no. 3, pp. 5–10,

Sep. 2018.

[S18] N. Geier és Gy. Mátyási: Szénszállal erősített polimer kompozit (CFRP)

szálvágási szögének hatása a forgácsoló erőre. GépGyártás, vol. LV, no. 2, pp.

104–108, 2015.

[S19] N. Geier és T. Szalay: Analysis of the cutting forces in machining of uni-

directional carbon fiber reinforced plastics (UD-CFRP). Poceedings 7th Int.

Technol. Conf. CVUT, pp. 42-46.

[S20] N. Pálfi és N. Geier: A Comparative, Experimental Study of Full Factorial and

Central Composite Designs, Through Machinability Analysis of Aluminum

Alloy. Óbuda University e-Bulletin, vol. 9, no. 1, pp. 11-17, 2019.

[S21] I. Biró, T. Szalay és N. Geier: Effect of cutting parameters on section borders

of the empirical specific cutting force model for cutting with micro-sized uncut

chip thickness. Procedia CIRP, vol. 77, pp. 279–282, Jan. 2018.

[S22] Cs. Pereszlai és N. Geier: Edge trimming of unidirectional carbon fibre-

reinforced polymer composite. Technol. Forum 2019, (benyújtva, bírálat alatt).

[S23] N. Geier: CAD/CAM alkalmazások integrációja forgácsolt alkatrészgyártásba.

OGÉT 2014 Konf., pp. 127-130., 2014.

[S24] B. Somoskői és N. Geier: Szénszállal erősített polimer kompozit (CFRP)

forgácsolhatósági vizsgálata kompressziós szármaróval. OGÉT, pp. 422–425,

2018.

[S25] N. Pálfi és N. Geier: A comparative study of full factorial and central composite

designs through the machining of aluminium alloy. Proc. 8th Int. Eng. Symp.

Bánki, p. 9, 2016.

[S26] Cs. Pereszlai, N. Geier és D. Poór: Influence of fixturing setup on quality of

edge trimmed UD-CFRP. Int. Conf. Innov. Technol. -TECH 2019, (benyújtva,

elfogadva).

[S27] N. Geier, T. Szalay és I. Biró: Trochoid milling of carbon fibre-reinforced

plastics (CFRP). Procedia CIRP, vol. 77, pp. 375–378, Jan. 2018.

20

21

God is all I don't know.

That's why I believe in it more and more.

22