Prediction of Failure Due to Fatigue of Wire Arc Additive Manufacturing-Manufactured Product

Abstract

Currently, the focus of production is shifting towards the use of innovative manufacturing techniques and away from traditional methods. Additive manufacturing technologies hold great promise for creating industrial products. The industry aims to enhance the reliability of individual components and structural elements, as well as the ability to accurately anticipate component failure, particularly due to fatigue. This paper explores the possibility of predicting component failure in parts produced using the WAAM (wire arc additive manufacturing) method by employing fractal dimension analysis. Additionally, the impact of manufacturing imperfections and various heat treatment processes on the fatigue resistance of 30CrMnSi steel has been investigated. Fatigue testing of samples and actual components fabricated via the WAAM process was conducted in this study. The destruction of the examined specimens and products was predicted by evaluating the fractal dimensions of micrographs acquired at different stages of fatigue testing. It has been established that technological defects are more dangerous in terms of fatigue failure than microstructural ones. The correctly selected mode of heat treatment for metal after electric arc welding allows for a more homogeneous microstructure with a near-complete absence of microstructural defects. A comparison of the fractal dimension method with other damage assessment methods shows that it has high accuracy in predicting part failure and is less labor-intensive than other methods.

1. Introduction

Currently, instead of traditional technologies for producing blanks, various 3D printing techniques are being increasingly used. Some of the most common methods include SLM (selective laser melting), WAAM (wire arc additive manufacturing), and laser powder bed fusion (LENS/DMD) [1,2,3]. In the process of additive manufacturing, blanks can be successfully produced from alloys based on titanium [4], aluminum [5], steel [6], and nickel [7], which is an advantage of this technology. Other advantages of additive manufacturing include the following:

• The possibility of fully automating the production of product blanks;
• Reducing the resource intensity of production, which is particularly pronounced in the production of products made from expensive materials such as titanium and nickel alloys;
• Reduction in the overall cost of production, providing the opportunity to produce single and small-scale products that would not be economically viable using traditional techniques [8,9].

The reliability of individual parts and structural elements, particularly those operating at low temperatures, is determined by various material parameters [10]. It is known that most components of mechanisms and structures operate for a long time under cyclic loading conditions [11,12,13,14]. Statistics show that approximately 90% of failures and accidents are related to the fatigue nature of loading [13,14].

Fatigue is a complex multifactorial phenomenon. In analyzing fatigue strength, factors such as frequency [15,16,17], asymmetry of the load cycle [16,18,19], stress concentration, and others are added to those that determine the fracture process under static loading. Additionally, the role of surface conditions [20] and temperature increases significantly [21,22,23,24].

The issues of the physical nature of fatigue have been extensively studied [25,26]. This includes the development of the mechanism for the formation and growth of fatigue cracks [27], the detailed examination of fatigue structures and patterns of their changes, the study of hyperstress, fatigue at cryogenic and elevated temperatures, and many others.

During the course of our literature review, we found that the characteristics of fatigue failure in materials produced by additive manufacturing are relatively rare in the literature and have received little attention [28,29].

In recent years, approaches to nonlinear dynamics and fractal representation developed by I.E. Krasikova, V.I. Trefilov, M. Zaiser, and others have been widely used in metallurgy [30,31,32]. These methods allow for a quantitative assessment of structural degradation processes based on a specific criterion: the fractal dimension.

The fractal dimension (D) is a quantitative indicator of an image. Based on the fractal dimension index of the structure and material fractures, various material characteristics can be identified. In particular, fatigue characteristics [33], hardness and strength [32,34], elongation and contraction [35], toughness [36], and a number of other material characteristics can be determined [37,38,39].

At the moment, there are active programs being developed to explore the fractal dimension of images. When studying metal structures, many researchers define the fractal dimension of an image of a structure as the fractal dimension of a grain boundary curve [40].

Thus, an analysis of the literature and the current state of research in the field showed that there is a lack of experimental data on the physical and mechanical properties, fatigue strength, and durability of 3D-printed structural materials produced by electric arc surfacing. Additionally, there are no recommendations for methods to improve the products (e.g., heat treatment) using WAAM technology in order to achieve the necessary structural parameters and properties of the finished product. Furthermore, there are no established models of damage accumulation or fatigue failure mechanisms for these materials.

This paper discusses two important aspects of additive manufacturing: ensuring the defect-free production of materials and predicting their fatigue failure. The practical significance lies in the ability to predict the specified characteristics of materials produced by additive electric arc surfacing using a fractal analysis method. This method allows for an accurate and fast prediction of the properties of these materials, thanks to a large amount of experimental data.

The main goal of this study is to investigate the possibility of predicting part failure using fractal dimensions in WAAM-produced parts. Additionally, the influence of various heat treatment methods on the fatigue strength of 30CrMnSi steel will be explored.

2. Materials and Methods

2.1. Experimental Setup and Materials

In this study, 30CrMnSi steel wire was used to surface the samples. The chemical composition was determined using optical emission spectrometry on the FOUNDRY-MASTER UVR device. The detection limit for control elements is 0.001% of the mass fraction. RMS error is not more than 3%.

The chemical composition of the wire can be found in

Table 1. Chemical composition of initial welding wire, wt. %.

Material	C	Si	Mn	Ni	S	P	Cr	Mo
30CrMnSi	0.269	1	0.93	0.099	0.015	0.020	0.961	0.005

.

In [41,42,43,44,45], the technology of 3D printing using electric arc welding is described. Our research on the process of 3D printing workpieces using electric arc welding was conducted on an experimental setup designed and created specifically for this purpose (Figure 1) [24]. The method for 3D printing on CNC machines used in the setup is protected by a patent: RU 2696121 C1 [46].

To obtain samples, walls with a thickness equal to one surfacing roller pass were printed. To assess the anisotropy of the metal properties, samples were cut from the resulting wall for tensile and fatigue testing, both longitudinally and transversely relative to the surface.

The production of samples was carried out in accordance with the following modes:

• I= 150 A, U = 25 V (samples No. 1);
• I = 110 A, U = 17 V (samples No. 2).

Hardening and tempering and air-hardening were used as heat treatments for the 30CrMnSi steel. Quenching was carried out by heating the steel to 950 °C and then cooling it in water. Tempering was applied at 550 °C for two hours on all samples. Air-hardening was also applied at 950 °C, but this time the steel was cooled in air. The quenching and air-hardening temperatures are higher than the standard ($A_{c3}+30\dots 50 °C$) because it is necessary to overheat the metal to prevent the formation of structural defects during surfacing.

2.2. Fatigue Testing

Samples for fatigue testing were cut along and across the surface direction from printed and heat-treated blank materials using electroerosion cutting. These samples had a thickness of 3 mm and a working area of 60 × 15 mm² (Figure 2). After cutting, the working parts of some samples were ground and polished to a roughness of Ra 0.2, and etching was then performed to assess the structural evolution and crack growth features in the working areas during testing.

Fatigue tests were conducted on a specialized testing rig. The setup is illustrated in Figure 3. Samples were subjected to a cantilever bending load (cycle asymmetry coefficient R = −1). The frequency of the elastic–plastic cycling was set using a frequency inverter to 8.3 Hz (500 cycles per minute). The stress amplitude in the cycle was controlled by adjusting calculations and a clock-type gauge. During the test, the following parameters were recorded: the number of cycles (N) and the maximum stress amplitude (σ_max) in the cycle. Stress monitoring in the test area was also performed using a strain gauge with thermal compensation for steel.

The operating time of the sample material (N/N*) is defined as the ratio between the current number of loading cycles (N) and the number of cycles at which complete destruction of the sample material occurs (N*).

2.3. Microstructural Studies

The study of the metal microstructure was conducted on microetches using an Altami MET 1C optical microscope. Magnifications of ×50, ×200, and ×500 were employed in the study. Sample preparation was performed according to the standard method, including etching in the metal. The air-hardened metal microstructure is composed of ferrite and perlite, as shown in Figure 4.

Fatigue samples were polished and etched in the area of highest stress using a similar method. Metallographic analysis of the samples was conducted in the region of the most likely failure of the sample, which was determined both by calculation and modeling in the Autodesk Inventor Nastran 2022 environment using the multi-axial fatigue analysis method (analysis of multi-axial fatigue with the main criterion of equivalent von Mises stress).

2.4. Fractal Dimension

The fractal dimension of the DF image was also used as an informative quantitative indicator of the alloy microstructure. A specialized program was used to determine the fractal dimension.

At least 20 microstructure photographs at various magnifications were taken for statistical and digital analysis in the working area of the specimen at certain loading stages.

The principle of fractal dimension determination is shown in Figure 5.

2.5. Product Manufacturing

A 3D-printed part of the “Front ATV lever” is shown in Figure 6.

The manufacturing process was as follows:

1. Deposition of 30CrMnSi steel on an 8 mm thick substrate, according to the control program and in the selected printing mode;

2. Removal of the substrate by milling;

3. Mechanical processing of the lever (material utilization factor was 0.76);

4. Heat treatment of lever hardening and tempering in previously selected modes;

5. Finishing by milling and drilling.

A special stand was designed and built to conduct fatigue tests on the lever (Figure 7).

The installation consists of the following:

(1) A load-bearing steel frame is made by welding a square-section profile pipe with dimensions of 60 × 60 mm² and a thickness of 3.5 mm, according to GOST 30245 [47]. To adjust the position and height of the pneumatic cylinders, we use ears with mounting holes made of 09G2C steel sheets with a thickness of 6 mm. The frame has a closed spatial shape and uses paws to keep it upright (these do not affect its rigidity).

(2) A DNC series pneumatic cylinder manufactured in accordance with the ISO 15552 standard [48]. The piston diameter is 125 mm, the stroke is 200 mm, and the working pressure ranges from 1 to 10 bars.

(3) Pneumatic distributor with electric control, operating pressure from 1 to 10 bars, and 220-volt voltage.

(4) Obeh ΠP 200 pneumatic distributor control controller. Obeh ΠP 200 is a serial product, and meets the requirements of TR CU 004/2011 [49].

(5) A computer is used to record the readings from the strain gauge. The stand simulates the operation of the lever under real conditions, such as when the vehicle is in motion. A pneumatic cylinder is used to move and lock the free end of the lever at the top position. For the low-cycle test, a spring with a preload of 5000 N and a maximum compressive force of 17,500 N was selected. The force applied to the lever is determined by the compression height of the spring. A programmable relay, Obeh ΠP 200, opens and releases pressure in the pneumatic cylinder using a pneumatic distributor and a timer, as well as recording cycles in the controller’s memory and displaying numbers on the screen.

A single-axis foil strain gauge KFGS 3-350 C1-11 was used to monitor voltages at critical points. The strain gauge was connected according to the half-bridge circuit using the National Instruments NI 9219 ADC (MIR company, Moscow, Russian Federation).

To obtain metal microstructures at different times during operation, critical points were pre-ground, polished, and etched using the procedure described in Section 2.4. At various stages, the test was interrupted and a metallographic analysis was performed.

2.6. Method of Damage Assessment Based on Fractal Analysis of Microstructure Images

The invention relates to the field of scientific research methods for determining the operational time and residual life of a material sample.

A method for evaluating damage is known [50], based on image processing of microstructures and determining the number of objects in the image, as well as classifying the obtained data. However, this method has a relatively low accuracy in determining material damage.

There is also a computational method proposed by Shetulov D.I. [51], which is based on calculating the damage index F. This index is determined as follows:

$$\mathrm{F}={\displaystyle \frac{{\mathrm{n}}_{32}}{{\mathrm{n}}_{31}}}{\displaystyle \frac{{\mathrm{n}}_{34}}{{\mathrm{n}}_{33}}}{\displaystyle \frac{{\mathrm{n}}_{\mathrm{s}\mathrm{b}2}}{{\mathrm{n}}_{\mathrm{s}\mathrm{b}1}}}{\displaystyle \frac{{\mathrm{F}}_{\mathrm{m}}}{{\mathrm{F}}_{\mathrm{s}}}},$$

(1)

where ${\mathrm{n}}_{31}$ is the total number of grains in the microstructure photo;

${\mathrm{n}}_{32}$ is the number of damaged grains;

${\mathrm{n}}_{33}$ is the difference between intact and damaged grains over the entire grain area;

${\mathrm{n}}_{34}$ is the number of grains with wide (more than 6 microns) slip bands;

${\mathrm{n}}_{\mathrm{s}\mathrm{b}2}$ is the total number of slip strips in damaged grains;

${\mathrm{n}}_{\mathrm{s}\mathrm{b}1}$ is the number of winding and intermittent sliding strips;

${\mathrm{F}}_{\mathrm{m}}$ is the actual area of the microstructure;

${\mathrm{F}}_{\mathrm{s}}$ is the area of the working surface of the sample.

As can be seen from the above dependencies, damage assessment is based on considering the number of damaged and undamaged grains, as well as the number and area of slip strips. However, this method has the disadvantage of being complex to calculate the damage. To improve the accuracy of damage estimation, it is suggested to use an additional quantitative parameter to evaluate the microstructure, such as the fractal dimension of the image.

The technical result is achieved through the use of the fractal dimension indicator for the microstructure image, as well as image processing of the microstructures. This allows us to determine quantitative parameters, which are then analyzed using neural network modeling. This process helps assess the damage to a structural alloy and estimate its remaining life.

The image processing of the microstructure and the determination of quantitative parameters were carried out in the following steps:

(1) The obtained images of the microstructures were filtered and binarized;

(2) Using a computer program, the concentration of slip bands and defects in the material was calculated according to the following formula:

$$\mathrm{n}={\displaystyle \frac{{\mathrm{N}}_{\mathrm{d}\mathrm{e}\mathrm{f}}}{{\mathrm{F}}_{\mathrm{m}}}}({\mathrm{m}\mathrm{m}}^{-2}),$$

(2)

where ${\mathrm{N}}_{\mathrm{d}\mathrm{e}\mathrm{f}}$ is the number of slip bands and defects in the analyzed microstructure image;

${\mathrm{F}}_{\mathrm{m}}$ is the area of the analyzed image (in mm²).

(3) Additionally, after pretreatment, the relative area of microstructural defects and slip bands formed during fatigue loading was determined according to the following relationship:

$${\mathrm{F}}_{\mathrm{r}\mathrm{e}\mathrm{g}}={\displaystyle \frac{{\mathrm{F}}_{\mathrm{d}\mathrm{e}\mathrm{f}}}{{\mathrm{F}}_{\mathrm{m}}}},$$

(3)

where F_def is the area of slip bands and defects in the analyzed microstructure image in mm².

(4) To calculate the fractal dimension of the image of the DF microstructure, a rectangular grid was placed over the processed image with a cell size of e (the range of cell sizes was chosen based on the average grain size of the metal (d) from 0.01d to d). After that, the number of cells (N) containing the boundary or fragment of the slip strip, as well as the defect, was calculated. The fractal dimension was then determined using the equation of the regression line, which is given by the following:

$$\log \left(\mathrm{N}\right)={\mathrm{D}}_{\mathrm{F}}·\mathrm{l}\mathrm{o}\mathrm{g}(1/\mathrm{e})+\mathrm{C},$$

(4)

where $\mathrm{C}$ is the regression line coefficient.

(5) The image of the material’s microstructure in its initial state was analyzed to determine the initial quantitative microstructural parameters F_original, n_original, and D_original. The analysis was performed by averaging over at least 10 images.

(6) Using the software, the relative change in the fractal dimension of the ΔD_F image was estimated in individual areas. At the same time, in areas where the largest increase in fractal dimension was observed, it was possible to predict the appearance of microcracks and macrofractures in the future.

The spread of experimental data was estimated based on statistical processing of the calculated results of several images (approximately 10–16) in the area with maximum stress, which is the zone of a dangerous section of a part or sample.

Based on this developed diagnostic method, we have also created a computer program, which has been registered with the appropriate authorities (Certificate No. 2023684547, dated 16 November 2023) [52].

3. Results

3.1. Metallography

During the metallographic analysis of samples made from 30CrMnSi steel, we obtained microstructures of the metal deposited in both modes (Figure 8).

The microstructure of sample No. 1, with mode I of 150 A and U of 25 V, is composed of ferrite and sorbite. This indicates the occurrence of quenching and tempering processes during the surfacing of subsequent metal layers. Based on the height of the sample under study, structural heterogeneity can be observed, with clear differences in the size of ferritic colonies (Figure 9).

An abnormal ferrite–perlite structure was observed in the metal microstructure of sample No. 2 (mode I = 110 A, U = 17 V). Due to severe overheating during the surfacing process and accelerated cooling, ferrite was released in the form of closed networks along the boundaries of the previous austenitic grains. The Widmanstetten structure can be seen in the metal, which is most clearly visible at high magnifications (Figure 10). Perlite is presented as both highly dispersed plates and partially spheroidized colonies.

Based on the results of the analysis of the microstructures of samples deposited using different modes (Figure 8, Figure 9 and Figure 10), we can conclude that when the samples are deposited using the mode with a linear energy input of Q = 600 J/mm (mode No. 1), there is a more active recrystallization of the structure. This is due to the increased thermal energy supplied during the additive process. Despite this more favorable structure when surfacing using this mode, there is structural heterogeneity in the height of the sample, which could lead to a reduction in the mechanical properties of the metal.

It should also be noted that there is an increased risk of metal spattering and the formation of porosity and other technological defects when surfacing with this mode (I = 150 A, U = 25 V, Q = 600 J/mm). This could also lead to a decline in the overall properties of the material. The presence of macroscopic defects of a technical nature is clearly visible on the walls after milling (Figure 11).

When examining the deposited walls (Figure 11a), large accumulations of macrodefects are clearly distinguished in the workpiece for the manufacture of sample No. 1. These macrodefects can be characterized as pores with a lack of fusion. In contrast, when surfacing blanks for sample No. 2, macrodefects were practically not detected (Figure 11b).

The accumulation of defects can lead to a decrease in the mechanical properties of the material.

Carrying out heat treatment options such as hardening and tempering and air-hardening makes it possible to achieve a defect-free homogeneous structure throughout the height of the sample. The structure of the metal after heat treatment is shown in Figure 12.

The microstructure of the air-hardened metal is composed of ferrite and pearlite, which is in agreement with the theoretical concepts of metal heat treatment (Figure 12a). Following the hardening and tempering, the metal’s structure consists of troostite–sorbite, also consistent with the theoretical data (Figure 12b).

3.2. Fatigue Tests

To study the effect of structural and technological defects on the fatigue strength of metal, samples were tested without heat treatment in both modes. The fatigue curves for samples No. 1 and 2 are shown in Figure 13 for the low-cycle fatigue range.

According to the test results of the non-heat-treated samples, mode No. 2 was selected as the more favorable option. Further study on the fatigue strength of the material was conducted on samples prepared according to the following parameters: I = 110 A and U = 17 V.

Fatigue curves were generated for 30CrMnSi steel in its original state and after various heat treatments. The results are presented in Figure 14.

3.3. Fractal Dimension

The fractal dimension was determined using sample No. 2. The metal was examined at various operating times, both immediately after surfacing and after heat treatment. A general view of the micrographs used in the study can be seen in Figure 15.

The dependencies of the obtained quantitative parameters of the microstructure during fatigue loading for 30CrMnSi steel are shown in Figure 16.

3.4. Tests of the Lever

During the testing of the ATV lever on the installation described in Section 2.5, we recorded the evolution of the fatigue process. This allowed us to apply fractal dimensionality to predict the destruction of this component. The fractal dimension data obtained from the tests are presented in Figure 17.

The approval of the developed method was conducted by calculating the fractal dimension index using the formula proposed by Shetulov (method 1) and according to the method described in [50] (method 2), as well as the proposed method. The average calculation time and accuracy of forecasting were compared to evaluate the accuracy of the methods. The accuracy was estimated by comparing the number of cycles before sample destruction (N) with the number of cycles where the damage exceeded 0.95 (N_predicted).

The obtained data on the approbation are shown in

Table 2. Comparison of damage assessment methods.

Material	Method 1 (Shetulov D.I.)		Method 2 (Andreeva O.V.)		The Proposed Method
	Average Calculation Time, min	Prediction Accuracy	Average Calculation Time, min	Prediction Accuracy	Average Calculation Time, min	Prediction Accuracy
30CrMnSi	14	86%	0.6	66%	0.2	88%

.

4. Analysis of the Results

4.1. Metallographic Studies

During the study of the microstructure of the deposited samples, it was found that when metal was deposited using both studied modes (I = 150 A, U = 25 V and I = 110 A, U = 17 V), a structural heterogeneity in the height of the deposited wall was formed. This effect is characteristic of the WAAM surfacing method and has been widely described in the literature [42,53], including for surfacing other metals [54,55,56].

The appearance of structural heterogeneity can be explained by differences in cooling temperatures during surfacing. The first layers of metal cool faster than in later stages of workpiece manufacture, due to more active heat dissipation caused by the weight of the substrate.

Despite the formation of abnormal and defective structures in both samples, this does not affect the properties of the final product, as heat treatment is applied during the manufacturing process, which allows for the creation of a defect-free structure. This has been confirmed by metallographic analysis of the samples after treatment and by the literature data [5,57,58,59,60].

During the microstructural analysis of the deposited material, it was discovered that a significant number of large technological flaws (pores, unwelded spots, etc.) only formed when the surfacing process was carried out according to the following settings: I = 150 A and U = 25 V.

Despite the fact that sample No. 1 has a more favorable structure in terms of material properties, its fatigue life is lower than that of sample No. 2 by an average of 70%. This difference may be due to the presence of macroscopic pores, non-metallic inclusions, and other technological defects in sample No. 1’s metal. Based on the data presented in Figure 13, we can conclude that these defects have a larger impact on the metal’s fatigue strength than imperfections in its microstructure, and they should be eliminated through both visual inspection and non-destructive testing of the material.

4.2. Fatigue Tests

During fatigue tests of samples without heat treatment and with different surfacing methods, it was found that microstructural defects have less impact on fatigue strength compared to technological defects such as pores, non-metallic inclusions, etc. The fatigue strength of samples taken from a defective workpiece was on average 30% lower than the fatigue strength of similar samples made from metal without these technological defects.

Testing of heat-treated samples revealed that samples produced by 3D printing from 30CrMnSi steel had lower durability than samples obtained from rolled steel by 5–8%. After air-hardening, an increase in the durability of the material was observed by an average of 15–30%. Hardening and tempering also made it possible to increase the durability of the alloy by more than 50%.

As can be seen from the graph (Figure 14), the highest durability corresponds to a sample subjected to hardening and tempering (microstructure of troostite–sorbite), which is consistent with theoretical concepts of metal fatigue [61]. As is known, steel has maximum fatigue resistance after quenching and medium tempering, which leads to the formation of troostite in the structure. The formation of martensite during heat treatment results in a decrease in fatigue resistance, while the lowest fatigue resistance is associated with steels that have a ferrite–pearlite structure. Therefore, the appearance of decarbonized layers in structures that operate under cyclic loading is unacceptable. In steels with a microstructure that includes troostite, fatigue cracks are relatively slow to form and grow.

In terms of fatigue strength for future products, the most beneficial heat treatment is quenching followed by high tempering.

4.3. Fractal Dimension Research

The dependence of the fractal dimension changes during fatigue loading (Figure 16) reflects structural changes in the material and is nearly monotonic. A particularly intense change in the indicator is observed after the appearance of microcracks and major macrofractures in the analyzed area. The criterion for pre-failure can be the relative change in the fractal dimension of the analyzed area (ΔD_F). For 30CrMnSi steel, this value is approximately ΔD_F = −0.1.

It should be noted that, from the perspective of fractal dimension, the most predictable outcome is the destruction of the metal in a hardened and tempered state.

The study of microstructures obtained during the lever test allowed us to determine the nature of changes in the fractal dimension of the finished product. It is worth noting that the overall trend of the relationship between the DF parameter and N/N*, when testing the ATV lever, corresponds to the trend obtained for flat samples.

4.4. Full-Scale Testing of the Part

During field tests, the destruction of the ATV lever occurred at an operating time N/N* greater than 0.95. The number of loading cycles corresponded to the value predicted.

As can be seen from a comparison of the proposed method and known methods for predicting metal fracture (Table 2), the proposed method not only significantly reduces calculation time but also provides the highest level of accuracy in the forecast.

5. Conclusions

1. It has been established that technological defects have a greater impact on the fatigue strength of printed metal than microstructural defects.

2. Despite the fact that the metal structure obtained through 3D printing contains a large number of defects and is heterogeneous in terms of the height of the workpiece, a carefully selected heat treatment process allows for the creation of a defect-free microstructure.

3. High-temperature quenching followed by tempering results in the highest fatigue strength in steel 30HGSA.

4. The fractal dimension technique used in this study enables accurate estimation of the remaining metal lifetime.

5. The forecasting method we have developed and presented not only has the highest accuracy compared to the two other methods in the literature but also significantly reduces calculation time.

6. Predicting product life using the fractal dimension approach is feasible in cases where the surface of the component can be pre-prepared.