The Factors That Affect the Expansion of the Tape for It to Avoid Side Effects in the Production of Composites in Online LATP Technology

: During LATP (laser automated tape placement), the compaction roller contacts the prepreg and affects the pressure distribution directly. Moreover, the design parameters of the roller are optimized with the aim of improving pressure uniformity. This paper examines the impact of the contact line and surface that depends on the compaction force, the design of the roller, the angle of inclination and the angle of inclination of the strip. These factors signiﬁcantly affect the expansion of the tape, and it is important to determine it to avoid side effects in the production of composites (formation of gaps or overlaps). Their presence increases the percentage of pores of the ﬁnal material and thus reduces the mechanical properties. The results show that the pressure uniformity can be improved signiﬁcantly by design optimization of the roller, which indicates that higher bond quality between layers is achieved. The lower the speed and higher the compact force in this technology give better intimate contact with a lower percentage of voids and good mechanical characteristics.


Introduction
Thermoplastic tape placement is a special case of an out of autoclave process; it has a high potential for the aerospace industry due to its high degree of automation. Unidirectional (UD) continuous fiber-reinforced tapes with high fiber content are used as semi-finished material. Within the laser thermoplastic tape placement process, tapes are melted by a heat source, such as a laser beam, and bonded to laminates by compaction force induced by a consolidation roller. However, the achieved consolidation quality, especially the interlaminar void content, does not meet the standards of the aerospace industry yet. This paper aims to investigate the effect of the compaction force on the consolidation state of the incoming tape and its consequences on the governing mechanisms during consolidation. For this purpose, experimental research has been performed to analyze the state of the tape during the LATP process. In recent years, more research has been conducted to understand the parameters that affect the heating, compaction and cooling phases of the LATP process [1][2][3][4].
The development of intimate contact between the layers, which is a prerequisite for bonding, consists of equalizing the unevenness of the tape and the laminate. Firstly, bond strength in regions where intimate contact is achieved, then interdiffusion of polymer chains occurs due to accidental thermal movement. The interdiffusion process is generally explained by the mobility of the polymer chains. The polymer matrix consists of intertwined chains that have limited movement. Their mobility, and thus the degree of diffusion, increases with increasing temperature [4][5][6]. Secondly, inter-laminar void content in the final structure is a direct result of intimate contact development. Interlaminar voids reduce

Materials and Methods
The thermoplastic composite materials evaluated in this study are: • UD prepreg material LM-PAEK Toray TC1225 Cetex ® /T700 • UD prepreg material PEKK/CF Ten Cate TC1320/AS4 All characteristics of the UD prepregs are summarized in Table 1. The tapes with 25.4 mm width were selected for the LATP process investigated in this work.

LATP Process (Laser-Assisted ATP Manufacturing)
The samples for the experimental tests were made at IACR of a six-axis robotic machine manufactured by Mikrosam. The parts of the machine are shown in Figure 1: Consolidation roller with outer diameter of 60 mm 2.
Spool for UD tape 3.
Laser Heat source 3 kW Laser line LDF series diode laser system (optic lens 33 × 43 mm and focal distance 200 mm).

4.
Mandrel/Tool 3. Laser Heat source 3 kW Laser line LDF series diode laser system (optic lens 33 × 43 mm and focal distance 200 mm). The material used in this study was TC1320-1 CF/PEKK supplied by Ten Cate and LM-PAEK TC1225 Cetex ® /T700 in 1 in. slit tapes. The laying of the samples takes place on a flat aluminum table with several variable process parameters, as given in Table 1. The minimum and maximum values of compact pressure, lay-up speed, inclination angle and laying angle were taken as variable parameters. The experiments were performed with different combinations of the previously listed variable process parameters in LATP technology. In order for the amorphous phase of the polymer PEEK to remain in the glass state and not cause further changes in the microstructure after compaction, the mass at which the sample is placed is heated to 175 °C to allow the compacted strip to cool below its glass transition temperature (160 °C). Samples for mechanical tests were cut from the mid-section of the flat pallets so that the acceleration and deceleration phases of the placement head did not affect the microstructure.

Mandrel/Tool
A thermal camera was used to monitor the surface temperature of the strip during the laying process.

Pressure Measurement
The actual pressure distribution under the compaction roller is unknown in the LATP process, so during placement, the compaction roller applies the pressure while rotating. However, the effect of roller rotation on the normal pressure is negligible, and the dynamic case can be simplified into a static one [7]. To measure the contact area under the roller so that the pressure can be calculated, knowing the predetermined force, pressuresensitive films are used, as demonstrated in Figure 2. During the experiments, a force of 200 to 1000 N was used, with a step of 200 N. Then, the dimensions of the stains on the pressurized films were measured, and the contact surface was calculated. The contact surface was calculated for each force in order to determine the pressure.  The material used in this study was TC1320-1 CF/PEKK supplied by Ten Cate and LM-PAEK TC1225 Cetex ® /T700 in 1 in. slit tapes. The laying of the samples takes place on a flat aluminum table with several variable process parameters, as given in Table 1. The minimum and maximum values of compact pressure, lay-up speed, inclination angle and laying angle were taken as variable parameters. The experiments were performed with different combinations of the previously listed variable process parameters in LATP technology. In order for the amorphous phase of the polymer PEEK to remain in the glass state and not cause further changes in the microstructure after compaction, the mass at which the sample is placed is heated to 175 • C to allow the compacted strip to cool below its glass transition temperature (160 • C). Samples for mechanical tests were cut from the mid-section of the flat pallets so that the acceleration and deceleration phases of the placement head did not affect the microstructure.
A thermal camera was used to monitor the surface temperature of the strip during the laying process.

Pressure Measurement
The actual pressure distribution under the compaction roller is unknown in the LATP process, so during placement, the compaction roller applies the pressure while rotating. However, the effect of roller rotation on the normal pressure is negligible, and the dynamic case can be simplified into a static one [7]. To measure the contact area under the roller so that the pressure can be calculated, knowing the predetermined force, pressure-sensitive films are used, as demonstrated in Figure 2. During the experiments, a force of 200 to 1000 N was used, with a step of 200 N. Then, the dimensions of the stains on the pressurized films were measured, and the contact surface was calculated. The contact surface was calculated for each force in order to determine the pressure. The material used in this study was TC1320-1 CF/PEKK supplied by Ten C LM-PAEK TC1225 Cetex ® /T700 in 1 in. slit tapes. The laying of the samples takes a flat aluminum table with several variable process parameters, as given in Tab minimum and maximum values of compact pressure, lay-up speed, inclination a laying angle were taken as variable parameters. The experiments were perform different combinations of the previously listed variable process parameters in LA nology. In order for the amorphous phase of the polymer PEEK to remain in the g and not cause further changes in the microstructure after compaction, the mass the sample is placed is heated to 175 °C to allow the compacted strip to cool below transition temperature (160 °C). Samples for mechanical tests were cut from the tion of the flat pallets so that the acceleration and deceleration phases of the pl head did not affect the microstructure.
A thermal camera was used to monitor the surface temperature of the strip the laying process.

Pressure Measurement
The actual pressure distribution under the compaction roller is unknown in t process, so during placement, the compaction roller applies the pressure while However, the effect of roller rotation on the normal pressure is negligible, and namic case can be simplified into a static one [7]. To measure the contact area u roller so that the pressure can be calculated, knowing the predetermined force, p sensitive films are used, as demonstrated in Figure 2. During the experiments, a 200 to 1000 N was used, with a step of 200 N. Then, the dimensions of the stain pressurized films were measured, and the contact surface was calculated. The con face was calculated for each force in order to determine the pressure.  The main factors that affect the intimate contact of this process are the compact force, the temperature of the laser, the speed of laying, the tilt angle of the head, angle of the laser, the thickness of the roller, the roughness of the surface of the tape, etc. For this purpose, the following parameters were taken as variables: -Compact force (400-600 N) -Thickness of the siliconized layer of the roller (P1 thickness of silicon 5 mm and P2 thickness of silicon 15 mm) -Tilt angle of the head (10-13 deg) -Angle of lay up (0, 45, −45, 90 deg) Other parameters were taken as constants. The contact line La formed during compaction of the base or contact surface depends on the compaction force, the type of roller (design of the thickness of the siliconized layer), the angle of inclination of the machine head, the diameter of the roller and the angle at it which takes the tape. In order to measure the contact area, paper is used, and the dimensions are measured. The obtained values that are measured are given in a table, graphs and diagrams are obtained from the obtained incision. Figure 3 shows the results of the analysis of the change in the shape of the rollers as a function of compaction force; the measurements are shown, including a linear fit for each of the four variables. Roller 1 (P1) is stiffer than roller 2 (P2), and this can be observed in the relation between the compaction force and the shape of the roller. Figure 3b shows that La increases with increasing compaction force. This effect is more apparent for roller 2 with a 15 mm thickness.

ILSS and Void
In order to see how the compact force acts on the mechanical characteristics while paying attention to the width of the strip depending on the angle of laying (which was measured in the previous part) in order to avoid gaps and overlap, composite plates were produced with two compaction forces (200 and 1000) and two different passing speeds (3 and 6 m/min) for both materials. The other variables are constant (angle of the head, 13 degrees; design (0, ±45, 90), number of layers, 16; laser temperature (for PEKK 420 • C, for PAEK 350 • C); mass temperature, 175 • C). Production of the produced plates is given with a picture, Figure 4. From the obtained plates using the two standards ASTM D2734 and ASTM D2344, pores and ILSS are calculated for both materials. The value of contact line (La) for different rollers, lead angle and angle of lay-up are given in Table 2.  The void content can be obtained by measuring the density of the fibers and the resin separately. Together with their weight fractions of the composite, this would yield a theoretical density, which can be compared to the actual density of the composite. The density should theoretically only contain the fibers and the resin with no void inclusions. In actuality, voids are presumably present. Since voids add volume but not mass, the presence of voids will decrease density. Therefore, the difference between the theoretical and measured density of the composite is the void content in volume fraction. The ILSS ASTM D2344 test method determines the short-beam strength of highmodulus fiber-reinforced composite materials. The specimen is a short beam with a thick laminate from 2 mm up to 6 mm according to standard dimensions with a length of the span (L)-is thickness × 4 mm. The specimen is loaded in three-point bending with a speed of testing of 1 mm/min.

The Factors Affect the Expansion of the Tape
During LATP, uncontrolled compaction pressure can cause defects (interstitial cavity, cavity, overlap. Therefore, this must be studied. With the help of the compaction roller, a compact surface is obtained under the roller. The compact force and the surface coming from the roller determines the pressure. In this article, this surface that occurs under the roller is determined by an experiment with pressure-sensitive films. The results show that defining the width of the tape after the compaction force (that differ from differing factors, Figure 3) will make the correct generation of the flat pallet in order to avoid defects (gaps and overlap) during the LATP process.
The contact length, La, increases with increasing compaction force, and the rate is the same for both rollers. This means that pressure is applied for a longer duration during the process and more time is available for consolidation. The average pressure applied does not, however, increase at the same rate as the compaction force due to the increase in the contact length. The pressure can be increased with roller 1, and the contact length is shorter than for roller 2 at the same compaction force. Furthermore, it should be noted that the position of the nip-point is not constant and moves with a change in contact length. As the compaction force increases, the compaction roller deforms more and more, so that the area under the rollers increases, which is a limitation for achieving high pressure. The average compaction pressure has a non-linear relationship with the applied compaction force, as shown in Figure 5. The results in Figure 5 provide useful information for further discussion.   The influence of compact pressure factors is also shown in Figure 7. From the results (Figures 6 and 7) obtained in the analysis of the impact of the contact force, the angle of inclination, the angle of the head and the design of the thickness of the compaction roller on the contact line (contact area), it was noticed that it must be controlled and determined for different materials before the LATP process. This is important to determine the exact spread of the strip that is laid after the compaction in order to have the correct gaps between courses. If this is not specified, gaps and overlaps will occur, which will affect the mechanical properties of the final material.

ILSS and Void
In order to see how the compact force affects different speeds and different materials, and previously to use the data obtained for the analysis of bandwidth, eight plates with different parameters were produced.
Using the previously mentioned formulas from the standards (ASTM D2734 and ASTM D2344) for the calculation of pores and ILSS, after five repetitions for each part and with an average value and standard deviation less than 10, the calculations are made (these results are shown in Table 3). From the obtained values, diagrams are obtained for the porosity and for the ILSS calculated according to the standard for both materials, shown in Figure 8.    From the obtained diagrams, it can be concluded that at lower speeds for bo rials with increasing compact force, the percentage of pores decreases. From Figu also noticed that the selected parameters give a lower percentage of pores in th with PEKK matrix and with lower speed. The reason for the higher percentage of samples with less compact force is due to the presence of larger pores between th as seen in the pictures below. The sample SL23PL35 (Figure 9), obtained by a higher compaction force and lower speed, showed only certain areas with a poly trix. This means that at lower deceleration speeds and by applying higher com force, a good quality bonding of the layers is also achieved. The lower speed allo time for the polymer to heat up at the coupling point, in which case, even with th cation of lower pressure, good bonding and squeezing of the empty air can occu a higher speed is ap From the obtained diagrams, it can be concluded that at lower speeds for both materials with increasing compact force, the percentage of pores decreases. From Figure 8, it is also noticed that the selected parameters give a lower percentage of pores in the boards with PEKK matrix and with lower speed. The reason for the higher percentage of pores in samples with less compact force is due to the presence of larger pores between the layers, as seen in the pictures below. The sample SL23PL35 (Figure 9), obtained by applying higher compaction force and lower speed, showed only certain areas with a polymer matrix. This means that at lower deceleration speeds and by applying higher compaction force, a good quality bonding of the layers is also achieved. The lower speed allows more time for the polymer to heat up at the coupling point, in which case, even with the application of lower pressure, good bonding and squeezing of the empty air can occur. When a higher speed is applied, as in the sample SL23PL32, and higher pressure, voids appear but not in a large percentage. This means that the higher speed does not allow enough time for good compaction when laying the layer with the previous layers, i.e., with the laminate.   The samples marked SL23PL40 and SL23PL41 are obtained with a low speed of 3 m/min and with different compaction forces 200 and 1000 N. The sample SL23PL40, obtained with a lower compaction force, has a higher percentage of pores, and the pores are observed in the connection of the first layers. The SEM images with larger magnification clearly show that the fibers are well coated with a matrix and that it has achieved good adhesion. In the sample SL23PL41 (Figure 10), obtained with higher compaction force, parts with extruded polymer matrix were noticed, and the pore content was 1.76%. Both samples have a good interface, and a good quality connection between the layers is achieved.
The results showed that the samples with a lower percentage of voids have higher mechanical characteristics ILSS. observed in the connection of the first layers. The SEM images with larger magn clearly show that the fibers are well coated with a matrix and that it has achiev adhesion. In the sample SL23PL41 (Figure 10), obtained with higher compactio parts with extruded polymer matrix were noticed, and the pore content was 1.76 samples have a good interface, and a good quality connection between the achieved.
Laminate The results showed that the samples with a lower percentage of voids hav mechanical characteristics ILSS.

Conclusions
From the analysis of the influence of the parameters of the tape width (con for various factors explained in this paper, it was concluded that the design of t paction roller (the thickness of the siliconized layer) and the angle of laying hav influence. The following conclusions can be drawn from the results: • The results confirmed that each laying angle of the laminate gives a differen surface.

•
From all four types of angles in this paper, the angle of laying in the directio fibers (0) gave the largest surface; thus, giving good intimate contact for go solidation and adhesion on the stencil, and thus gluing layer by layer.

•
For QI laminate with other angles (±45; ±30; 90), we need to increase or decr compaction force (according to Figures 6 and 7) to have a similar compact l face), as well as for the laying of the fibers in the direction (0) in order to have bonding of the layers without gap and overlap. From the analysis of tested samples based on PEKK and PAEK resin, No. 1 be summarized: • The minimum lay-up speed and the maximum compact force for both mater the best results (low porosity less than 2%).

•
The selected parameters for the samples with PEEK resin have better me characteristics than the PAEK samples. The reason for that may be the grea

Conclusions
From the analysis of the influence of the parameters of the tape width (contact line) for various factors explained in this paper, it was concluded that the design of the compaction roller (the thickness of the siliconized layer) and the angle of laying have a great influence. The following conclusions can be drawn from the results:

•
The results confirmed that each laying angle of the laminate gives a different contact surface. • From all four types of angles in this paper, the angle of laying in the direction of the fibers (0) gave the largest surface; thus, giving good intimate contact for good consolidation and adhesion on the stencil, and thus gluing layer by layer.

•
For QI laminate with other angles (±45; ±30; 90), we need to increase or decrease the compaction force (according to Figures 6 and 7) to have a similar compact line (surface), as well as for the laying of the fibers in the direction (0) in order to have an even bonding of the layers without gap and overlap.
From the analysis of tested samples based on PEKK and PAEK resin, No. 1-8 could be summarized:

•
The minimum lay-up speed and the maximum compact force for both materials have the best results (low porosity less than 2%).

•
The selected parameters for the samples with PEEK resin have better mechanical characteristics than the PAEK samples. The reason for that may be the greater flow of the PAEK resin, which can be seen from the SEM images.