TECHNICAL AND VOCATIONAL TRAINING INSTITUTE

FACULTY OF MECHANICAL ENGINEERING

POST GRADUATES PROGRAM
DEPARTMENT OF MANUFACTURING TECHNOLOGY

Course: Metal Forming – (MAT-613)

NAME ID NUMBER
KASSAHUN FERDA 038/15
SUBMITTED TO: A.PROFESSOR ASMAMAW

Addis Ababa-Ethiopia
 DECEMBER
2023

2
Table of Contents
Hot Drape Forming Analysis of Thermoset Materials..................................................................................................................2
1. ABSTRACT..........................................................................................................................................................................2
2. INTRODUCTION.................................................................................................................................................................3
3. Experiment............................................................................................................................................................................6
3.2 Methodology.....................................................................................................................................................................7
3.2.2 Design of Experiment...................................................................................................................................................8
3.2.3 Experimental Method..................................................................................................................................................10
4. Results and Discussion........................................................................................................................................................14
5. Mechanical Testing.............................................................................................................................................................22
6. Conclusions.........................................................................................................................................................................24
7. References...........................................................................................................................................................................26

1
Hot Drape Forming Analysis of Thermoset Materials

1. ABSTRACT

Composite materials are becoming prevalent in aerospace industries but a full

understanding of how processing these materials leads to potential defects still needs to

be fully characterized. This paper investigates the effect of forming temperature, ply

count, fiber type, and fabric type on out-of-plane wrinkle development for a hot drape

formed, double-joggle c-channel. Tensile tests were performed on wrinkled samples to

understand the effect that the wrinkle size has on mechanical properties. Influential

parameters were identified as the forming temperature affecting the resin viscosity, the

weave type, and the fiber type. A decrease of the resin viscosity reduced the generated

wrinkles. This is attributed to the improved interlaminar slippage with lower viscosity.

The same effect was also observed with increased fiber twist. The 8 harness weaves

eliminated out-of-plane wrinkles as compared to tape or plain weave spars. Increasing

the ply count increases the complexity and thus the likelihood of wrinkle formation but

was not identified as a driving factor for wrinkles. The results show that the primary

influencing factor of hot drape forming is the interlinear slip. Mechanical tests were

also performed and showed that wrinkles with an L/D less than 10 create a large

reduction of tensile strength as compared to non-wrinkled samples. Wrinkles with L/D

≥15 show comparable tensile strength to non-wrinkled samples.

2
2. INTRODUCTION

The use of advanced carbon fiber reinforced polymer composite materials has

expanded widely in the past decade specifically in the aerospace industry. The high

stiffness-to-weight ratio, fatigue resistance, and damage tolerance make carbon fiber

composites suitable for either military or large civil aircraft. One of the most important

issues in changing from conventional metallic alloys to composite materials is to

ensure there are no compromises in structural integrity. Boeing 787 and Airbus A350,

for example, contain such carbon fiber composites in approximately 50% of their

structural weight. Manufacturing of composite parts for aircraft structures has

traditionally used either monolithic or sandwich structures, assembled with mechanical

fasteners to a c-channel, rib or stringers made from composite, aluminum or titanium.

The assembly with mechanical fasteners adds both cost and weight to the structure.

One way to reduce the complexity and weight is to integrate a number of parts into one,

complex, co-cured composite part.The manufacturing of such composite parts has

traditionally been done by hand lay- up. Due to the increase in the number of

composite parts that are being used in the build- up of aircraft components (and also

because of the large scale of these parts), hand lay-up can be both time consuming and

costly. This has led to an increased focus on automation processes such as Advanced

Fiber Placement (AFP) or Automatic Tape Lay (ATL) in combination with multi-layer

forming.

The process by which a composite part is manufactured can be divided into three

phases: lay-up, forming, and cure-assembly. The forming phase of complex geometries

3
Can be extremely difficult and can result in defects (such as wrinkles) during the process.

Wrinkles can impact the structural integrity of a part. It has been shown that out-of plane

defects often occur during forming of a recess area within a part.

The forming process can be achieved through a variety of forming processes such as

vacuum forming, mechanical forming, or robotic forming. An example of vacuum

forming is Hot Drape Forming (HDF). Where a prepare lay-up is placed on a mold,

heated to a temperature below the curing temperature and formed down to the mold with

a rubber diaphragm using vacuum [4]. Figure 1 shows a schematic of the process.

Figure 1: Hot drape forming process

It has been shown that the forming process generally depends on multiple material

specifics and inherent mechanisms, such as locking angle, interplay shear, ply bending,

intra-ply axial loading, and compaction/consolidation. Hence, close attention has to be

given to understand the overall material behavior. In addition, the use of computer

modeling is necessary to define the optimal set of process parameters to avoid the

production of inferior parts. Through models and small scale experiments, many authors

have identified processing parameters that affect wrinkle development in automated

composite processing. Material based parameters such as interplay shearing. Material

friction

4
or tack, locking angle, and ply thickness all have been reported to affect wrinkle

initiation in a variety of processes. Many of these parameters (e.g., tack and interplay

shearing) are influenced by the change in resin viscosity as a function of temperature.

Other parameters, such as the locking angle, can influence how the in-plane deformation

affects the shapes to which a material can adopt. High locking angles typical of PW

fabrics have less conformability than do the lower locking angles of 8HS.

Process parameters that have been identified as influencing defect initiation are: form

temperature, form rate, form pressure, and initial contact point of the laminate. Even the

geometry of the part (such as the tool radius, joggles, ply drops, and laminate thickness)

can affect wrinkle development. C-channel geometries are optimal demonstration

articles because they possess all of the aforementioned geometries as shown in Figure 2.

In a series of unidirectional tape laminates were hot drape formed into a c-channel

geometry incorporating a joggle which is the change of web height along the length of

the c-channel. The joggle is outlined in Figure 2. It was shown that localized

compression stresses exist in the joggle region.

Figure 2: C-Channel geometry (adapted)

5
This paper further advances the work presented in and addresses how the drivability of

weaves can reduce wrinkle development in c-channels. It also presents a correlation

between resin viscosities, ply count. and out-of-plane wrinkle magnitude. Two thermoset

material systems with three different weave typesuni-directional, plain weave, and 8H

fabrics are investigated herein.In addition, mechanical testing are performed on samples

taken from the cured c-channel part in order to quantify the impact of wrinkle on tensile

strength reduction of the part.

3. Experiment

3.1 Materials

Two resin systems were tested in this experiment, Cycom 970 (Solvay) and Cycom

5320-1 (Solvay). Both resins are an epoxy; system 970 is a 350 °F cure autoclave system

while 5320-1 is an out-of-autoclave material 250 °F cure with 350 °F post cure. Two

different fiber twists were evaluated to show the effect of interply shear. Both standard

twist fiber (with 5 twists per inch) and never-twisted fibers were used in the tests and the

results compared. The manufacturer reports that the interlaminar shearing ability of

never twisted fibers is greatly superior to that of the standard twist material. Several

different weave patterns from T650 fibers were investigated as outlined in Table 1.

Table 1: Material Types

Resin System Fabric Type

Unidirectional Tape (UD)
Plain Weave standard
Cycom 970 twist fiber (PW-ST)
Plain Weave Never Twist
Fiber (PW-NT)
Unidirectional Tape
Cycom 5320-1 Plain Weave
8 Harness (8H)

6
Figure 3 shows surface images of the several green prepregs. The prepregs are fully

impregnated. The weaves (b & c) show that the resin primarily consolidates at the tow

junctions.

a b

Figure 3 - Material systems and forms used in this study a) UD b) PW c) 8H

Samples tested were within the manufacture’s recommended working life of 10 days and

20 days for 970 and 5320-1, respectively.

3.2 Methodology

3.2.1 Tool Geometry

Material formability was tested using one tool geometry outlined in Figure 4.

7
 Figure 4: Test tool geometry in x and y directions

The tool had two symmetric joggles and radii of 0.25 inch. Table 2 provides additional

information about the joggle region of the c-channel’s tool.

Table 2: Joggle geometry

Flange Length 3 inches

Joggle Run 9 inches
Joggle Instep 0.3 inch

The double joggle and joggle instep can create a challenging feature to drape form.

During forming, the laminate will follow a geodesic curve. Bending occurs on the tool

side ply, and interlinear shearing occurs as the remaining plies conform to the tool

geometry. The double joggle feature aggressively reduces the cross-sectional perimeter

of the c-channel making it an ideal test specimen. As described in this reduction creates

tensile and compressive stresses in the joggle region which can induce out-of- plane

wrinkles. However, with sufficient interlinear shearing, the plies are able to sufficiently

slide and reduce wrinkling.

3.2.2 Design of Experiment

For each material type, a quasi-isotropic layup was used, as presented in Table 3. For the

plain weave material, n was varied to accommodate variable ply counts.

8
Table 3: Layup for fabric types

Material Type Composite Layup

Unidirectional Tape [+45/0/-45/90/+45/0/-45/90/+45]s
Plain Weave [(45/0)n/45]s
8 Harness [(45/0)4/45]s

A series of experimental tests were conducted, as depicted in Table 4. This experiment

varied 4 variables: ply temperature, ply count, fabric type, and fiber twist. The

replicates performed for each c-channel, as well as the resin viscosity from the

manufacture’s data sheet are provided in Table 4.

Table 4: Test matrix

Test Resin
Resin Fabric Ply
Temperature Viscosity Replicates
System Weave Count
(F) (poise)
970 Uni 18 140 1500 2
970 PW-ST 10 140 1500 1
970 PW-ST 18 120 3000 2
970 PW-ST 18 140 1500 1
970 PW-ST 30 72 22000 1
970 PW-ST 30 100 6500 1
970 PW-ST 30 120 3000 1
970 PW-ST 30 140 1500 2
970 PW-ST 30 160 650 1
970 PW-ST 30 180 350 1
970 PW-NT 30 72 22000 1
970 PW-NT 30 100 6500 2
970 PW-NT 30 120 3000 2
970 PW-NT 30 140 1500 1
970 PW-NT 30 160 650 1
5320-1 Uni 18 125 90 2
5320-1 PW 18 125 90 1
5320-1 8H 18 125 90 2

9
3.2.3 Experimental Method

The experimental setup as well as the tool geometry for this study, are shown in

Figure 6. Flat composite laminates were fabricated based upon the layups provided in

Table 3. Each laminate was placed on the tool (Figure 6B) and heated. Laminate

temperature was determined by averaging three thermocouples placed in the laminate

flange. Once the laminate reached the desired temperature, vacuum was pulled and held

until the charge cooled to room temperature. Green state wrinkles were characterized

based upon both tool location and size, as depicted in Figure 7. The maximum height,

wrinkle wavelength, and out-of-plane wrinkle depth were measured. The ratio of

wavelength/depth (L/D) represents the out-of- plane wrinkle magnitude and is used in

analysis. A low L/D (i.e., LD < 10) represents a large green state wrinkle, which heavily

influences mechanical properties. A large L/D (i.e., L/D > 100) is a minimal green state

wrinkle which is likely to be mitigated during cure. Formed wrinkles were classified into

three groups: Span, Chord, and Off-Angle. Span wrinkles occur along the length axis of

the c-channel while chord wrinkles occur perpendicular to the length. Off-angle wrinkles

are a combination of chord and span wrinkles. These wrinkles are at some angle greater

than 0° but less than 90° referenced from the long axis. The three wrinkle types are

further depicted in Figure 5.

10
 Figure 5: Wrinkle Types

Figure 6: Hot drape forming setup: (A) Hot drape forming apparatus - (B) Composite laminate on tool

11
 Figure 7: Wrinkle size measurements

Several 970 PW-ST c-channels were cured using the manufacture’s recommended cure

cycle shown in Figure 8. Micrograph and tensile test specimens were taken from the

cured wrinkle zones in order to evaluate the effect of the wrinkle on the tensile structural

behavior. To fully capture the effect of cured wrinkles, a non-standard test method was

used to test various size samples. These samples were cut at various lengths because the

cured wrinkle sizes varied from sample to sample.

12
Table 5 shows the details of the specimens used for the mechanical testing. Unnotched tension tests (UNT)
were considered for this work.

Figure 8: Cycom 970 cure cycle

13
Table 5: Specimen dimensional detail

Nominal Dimensions Wrinkle Details

Test
Coupon Name Length Width Thickness Length Width Depth
Type
(in) (in) (in) (in) (in) (in)
UNT A 3.2 0.80 0.26 1.88 0.24 0.04
UNT B 3.8 0.79 0.25 3.8 0.2 0.03
UNT C 3.8 0.80 0.25 3.8 0.22 0.04
UNT D 3.8 0.80 0.25 1.87 0.15 0.01
UNT E 3.31 0.80 0.25
UNT F 3.13 0.80 0.25
No Wrinkles Observed
UNT G 3.31 0.80 0.25
UNT H 2.90 0.80 0.25

The specimens were placed into a hydraulic wedge grip, and load was applied until

failure. Note that the specimens were sanded near the edges in order have an even

surface for grip attachment. The main purpose of the tensile tests was to evaluate the

load reduction as function of wrinkle size and to get a preliminary understanding of the

relation between wrinkle size and tensile behavior and was not intended to find the

ultimate tensile load of the composite. All samples were tested at a constant

displacement of 0.05 in/min at room temperature.

4. Results and Discussion

4.1 Forming Result

Chord wrinkles formed on the web and the flange of the PW fabric c-channels shown in

Figure 9, and off-angle wrinkles only on the flange of the unidirectional tape material as

shown in Figure 10. No span wrinkles were observed in any c-channel. Off-angle

14
wrinkles were formed on the unidirectional tape c-channels, shown in Figure 10, and the

room temperature PW c-channels. These wrinkles are caused by the zero oriented fibers

bridging over the joggle recess due to an inability for plies to slip. The wrinkles were

eliminated when the resin viscosity was increased in the PW trials. The plain weave

fabrics exhibited chord wrinkles at the base of the flange shown in Figure 9. These

wrinkles were not observed, however, in the 8 harness weave. Figure 11 shows the result

of forming the 8H weave c-channel with no wrinkles observed post forming.

Figure 9: Typical web and flange wrinkles of PW fabrics

Figure 10: Typical web and Flange wrinkles of tape c-channels

15
 Figure 11: Post forming of c-channels for 8H material

The PW flange chord wrinkles formed at the inset of the joggle as shown in Figure

12. The change in area at the joggle creates regions of compression in the laminate,

which induces the ply bucking. Chord web wrinkles also were observed in several PW c-

channels. It is challenging to identify the root cause of these wrinkles because several

factors influence the formation of these wrinkles, as was observed through the testing and

outlined herein. Excess sagging from the transfer of the charge onto the tool was

observed to correlate with the web wrinkle locations. Additionally, material defects e.g.,

individual ply wrinkles observed toward the end of the material roll Figure 13 also can

induce web wrinkles even when these wrinkle are worked out during layup. Additionally,

web wrinkles occurred in all room temperature tests indicating that the process has an

influence independent of temperature. Because the web wrinkles are influenced by so

many external factors (and because a sufficient process was not developed to negate

these factors), the web wrinkles are only reported in the results and not further discussed.

16
 Figure 12: Wrinkle generation of 970 resin PW tests

Figure 13: Pre-existing chord wrinkles in individual plies

Averaged experimental results are provided in Table 6. It is important to note that, in

cases of the cycom 970 30 ply samples tested at either 160 or 180 °F) no wrinkles

occurred in the c-channels. Due to the tolerance of the wrinkle depth measurements,

samples with no wrinkles were reported as having an average L/D of 150

Table 6: Experimental Results

17
 Resin Fabric Ply Temperature Wrinkle Wrinkle Average Average Average Average
System Weave Count (F) Location Type H L D L/D

970 Uni 18 140 Flange Chord 4.322 0.328 0.032 12.142

970 Uni 18 140 Web Chord 1.300 0.275 0.046 11.514
970 PW-ST 10 140 Flange Chord 0.717 0.417 -0.001 358.333
970 PW-ST 18 120 Flange Chord 0.975 0.475 0.041 12.333
970 PW-ST 18 120 Web Chord 2.650 0.400 0.053 15.556
970 PW-ST 18 140 Flange Chord 1.200 0.510 0.033 15.455
970 PW-ST 30 72 Flange Chord 2.333 0.633 0.101 7.963
970 PW-ST 30 72 Flange Off-Angle 2.17 0.43 0.03 15.17
970 PW-ST 30 72 Web Chord 2.10 0.36 0.02 17.26
970 PW-ST 30 100 Flange Chord 0.976 0.536 0.059 12.507
970 PW-ST 30 100 Web Chord 3.155 0.425 0.014 38.125
970 PW-ST 30 120 Flange Chord 0.980 0.380 0.018 36.714
970 PW-ST 30 120 Web Chord 3.600 0.450 0.018 26.667
970 PW-ST 30 140 Flange Chord 0.923 0.409 0.037 54.952
970 PW-ST 30 140 Web Chord 2.180 0.460 0.030 16.850
970 PW-ST 30 160 - - - - - 150
970 PW-ST 30 180 - - - - - 150
970 PW-NT 30 72 Flange Chord 1.68 0.30 0.018 21.25
970 PW-NT 30 72 Flange Off-Angle 1.75 0.37 0.042 8.92
970 PW-NT 30 72 Web Chord 2.84 0.39 0.04 10.12
970 PW-NT 30 100 Flange Chord 1.3 0.21 0.031 78.39
970 PW-NT 30 120 - - - - - 150
970 PW-NT 30 140 - - - - - 150
970 PW-NT 30 160 - - - - - 150
5320-1 Uni 18 125 Flange Chord 1.725 0.184 0.411 0.593
5320-1 Uni 18 125 Flange Off-Angle 4.775 0.278 0.503 0.555
5320-1 Uni 18 125 Web Chord 3.409 0.278 0.478 0.587
5320-1 PW 18 125 Flange Chord 1.810 0.323 0.063 9.326
5320-1 8H 18 125 - - - - - 150

From these data, four comparisons can be made. Among the Cycom 970 PW-ST tests

both the effect of resin viscosity (temperature) and the number of plies can be shown.

The effect of ply slip can be observed by comparison of the Cycom 970 PW-ST

18
with the PW-NT. The effect of material weave pattern can be observed in the Cycom

5320-1 samples.

4.1.1 Effect of Resin Viscosity and Fiber Type

Study of the 970 system samples with a ply count of 30, indicates that the effect

of the resin viscosity on wrinkle development can be evaluated. Figure 14 shows the

interaction of resin viscosity to the average flange wrinkle magnitude, L/D. It is

observed that, as the viscosity decreases, the magnitude of the wrinkle also decreases

(increasing L/D). The data shows that there is a “sweet spot” for forming in order to

reduce the magnitude of wrinkles. For PW-ST, that viscosity occurs between 3000 and

6500 poise.

Consideration of the differences in wrinkles due to fiber twist reveals that the

never twisted fiber has superior forming properties over standard twist fibers. This is

likely a result of the reduced interlaminar shearing within the NT laminate compared to

the ST laminate. It is interesting to note that, at room temperature, the resulting wrinkles

between the ST and NT tests are similar. This indicates that other material properties

influence wrinkle development at low temperatures, while interlaminar shearing

dominates at higher temperatures. It is theorized that the tack of the prepreg may

dominate at room temperature allowing for the equivalent wrinkle sizes.

19
Figure 14: Effect of 970 resin viscosity on wrinkle L/D

4.1.2 Effect of ply count

The number of plies affects the forming as well as the wrinkling. The 970 system

PW-ST fabric at 140 °F shows that as the ply count increases, the magnitude of the

wrinkles also increases. The 10 ply charge possessed very small wrinkles, averaging

0.001 inch in depth, which would dissipate under debulk. Therefore, these cracks are

inconsequential. The 18 ply charge possessed one wrinkle in the flange with an L/D of

15.4. While this wrinkle would not dissipate during debulk it was less pronounced than

the 11 wrinkles in the 30 ply charges with L/D’s ranging from 5 to 200. As ply counts

increase, the total necessary force to deform a laminate also increases. The 30 ply

laminate had insufficient vacuum force to induce interply shear, which transferred the

load into the laminate and created compressive stresses leading to wrinkle development.

These results further support the work performed by Hallander, et al. [18], which showed

that compression stresses exist in the joggle region.

20
4.1.3 Effect of weave type

The 5320-1 system c-channels were all formed at 125 °F and 18 plies. The UD

c-channel exhibited wrinkles on the web and the flanges. The flange wrinkles included

both off-angle and chord wrinkles. The chord wrinkles occurred at the inset of the

joggle, while the off-angle wrinkles spanned the joggle region. The existence of off-

angle wrinkles indicates that the double joggle geometry was too aggressive to allow for

the use of unidirectional tape. The 0 oriented fibers bridged the joggle recesses and were

unable to shear causing the wrinkles. Fabric weaves exhibit a pinned joint enabling small

angle deformation, reducing the zero bridging, and successfully eliminating the off-angle

wrinkles.

The chord flange wrinkles occurred at the onset of the joggle. The magnitude of

these wrinkles degreases with increasing deformability of the harness type, as shown in

Figure 15. The UD tape resulted in the largest wrinkles. The PW fabric reduced the size

of the wrinkles as compared to the UD tape; but it did not eliminate them completely.

The 8H fabric reduced the wrinkles completely in the flange.

Figure 15: Effect of 5320 Weave type on L/D

21
 These results show that in-plane shear of the material greatly affects the out-of-

plane wrinkles. The increased locking angles of the 8 harness satin material allowed the

toes to deform by a larger amount, allowing it to conform to the c-channel geometry.

These results show that out-of-plane wrinkles are influenced by the weave type in

addition to being affected by the interply shear.

4.2 Forming Summary

Through these experiments, it was identified that the resin viscosity affects the

size of out-of-plane wrinkles. Both lower viscosities and increased forming temperature

reduce the size of the wrinkles. Fiber twist also has an effect. The never twisted material

is able to slip more easily than does the standard twist material. As a consequence, it can

produce wrinkle free parts at much lower temperatures than can the standard twist

material. It was shown that increasing the ply count increases the complexity of the part,

which increases the number of wrinkles observed. Furthermore, weave type influences

the out-of-plane wrinkles with the highly deformable 8HS producing no wrinkles.

5. Mechanical Testing

As a part of this investigation, a 970 system PW c-channel was cured; and

samples were taken in order to perform tensile testing for evaluation of the effect of

wrinkle on the tensile behavior. Web wrinkles from the room temperature 970 system c-

channel were subjected to a non-standard tensile test with coupons of various sizes, as

shown in Figure 16.

22
 Figure 16 - Test specimens cut from the c-channel:
4 specimens with wrinkles (A-D) and 4 specimens without wrinkles (E-H)

Several cross-sectional micrographs were taken at the specimen wrinkle location

in order to observe the waviness of the plies and the deformation due to the wrinkle

defect, as shown in Figure 17. It is clear from the micrographs that the wrinkles

originated on the tool side of the laminate and carried through the laminate to form a

visible bag side wrinkle.

Figure 17 - Cross section micrographs of the wrinkles

The results of the tensile tests are shown in Figure 18. The results are reported as

a percentage change with respect to a wrinkle free laminate. The results show that a

23
cured L/D less than 10 significantly reduces the tensile performance of a laminate, while

an L/D of 15 has similar tensile properties to a wrinkle free laminate.

1.2

1
Tensile Strength Correction Factor

0.8

0.6

0.4

0.2

0
0 2 4 6 8 10 12 14 16
Wrinkle L/D

Figure 18: Wrinkle strength correction factor

From these results, it should be noted that defects (such as wrinkles) generated during the

forming process can have a significant impact on the structural performance of a part.

Therefore, a clear pathway forward as to how to addressing such defects is necessary.

This can be done either by determination during the design phase as to what size wrinkles

are acceptable within limitations, by improvement of the process to eliminate their

existence prior to final cure, or by a combination of both.

6. Conclusions

This paper discusses the formability of two different resin systems namely Cycom

970 (Solvay) and Cycom 5320-1 (Solvay) with various weaves. The effect of forming

24
temperature, ply count, fiber type, and fabric type on out-of-plane wrinkle development

was investigated. It was observed that magnitude of wrinkles decreases as the resin

viscosity decreases or forming temperature increases. The never twisted material is able

to slip more easily than the standard twist and thus can be used to achieve wrinkle free

parts at much lower temperatures than can be obtained with the standard twist material.

Moreover, increase of the ply count increases the likelihood of wrinkle formation. The

fabric weave type also plays an integral part of the forming. The weaves performed

better than did the unidirectional fabric due to the relatively small angle deformation that

each ply can endure before wrinkling. Mechanical tests were performed on cured

wrinkles, and it was observed that the wrinkles had an impact on the tensile performance

depending on the wrinkle L/D value. Thus, in order for the industry to decrease the

impact of wrinkle development during processing, it must improve manufacturing

techniques by fully characterizing and understanding material systems. Future work will

investigate characterization of the intraply and interply shearing of the various fiber

types and correlate how these parameters affect wrinkle development in hot drape

forming processes.

25
 7. References

[1] Potter K. Beyond the pin-jointed net: Maximising the deformability of aligned

continuous fibre reinforcements. Compos - Part A Appl Sci Manuf 2002;33:677–

86. doi:10.1016/S1359-835X(02)00014-3.

[2] Larberg YR, Åkermo M. On the interply friction of different generations of

carbon/epoxy prepreg systems. Compos Part A Appl Sci Manuf 2011;42:1067–74.

doi:10.1016/j.compositesa.2011.04.010.

[3] Gutowski TG, Dillon G, Chey S, Li H. Laminate wrinkling composites scaling

laws for ideal. Compos Manuf 1995;6:123–34.

[4] Sorrentino L, Bellini C. Potentiality of Hot Drape Forming to produce

complex shape parts in composite material. Int J Adv Manuf Technol 2015.

doi:10.1007/s00170-015-7998-x.

[5] Long a. C. Composites forming technologies 2007:329.

doi:10.1201/9781439824344.

[6] Gereke T, Döbrich O, Hübner M, Diestel O, Krzywinski S, Cherif C. Numerical

draping simulations of textile composite reinforcements. ECCOMAS 2012—

European Congr. Comput. Methods Appl. Sci. Eng. E-b. Full Pap., 2012, p. 8094–

104.

[7] Döbrich O, Gereke T, Cherif C, Krzywinski S. Analysis and finite element

simulation of the draping process of multilayer knit structures and the effects of a

localized fixation. Adv Compos Mater 2013;22:175–89.

doi:10.1080/09243046.2013.791239.

26

View publication stats