Integrated Assessment of Coating-Steel Substrate Adhesion—Comparison of Mechanical and Ultrasonic Methods

Featured Application

The proposed method for assessing the adhesion of adhesive coatings can be used at the stage of the vehicle body repair process. Ultrasonic tests performed during post-accident repairs and their results can be related to the results of the studies included in this article. This will allow for estimation of the mechanical adhesion of coatings based on ultrasonic tests and the value of the reflection coefficient |r|.

Abstract

Adhesive coatings, including car paints and car putties, have found application in the construction of motor vehicles. This article contains the results of mechanical and ultrasonic tests of the adhesion of the car putty coating to a steel substrate. The main objective of this article is to determine the correlation between mechanical adhesion and the ultrasonic measure of adhesion—the reflection coefficient |r|. The results indicate that with the increase in the value of the coefficient |r|, the mechanical adhesion of the coating to the substrate decreases. The highest average mechanical adhesion of the coating to the substrate was obtained for sandblasted samples and was 2.74 MPa, corresponding to a coefficient reflection value of 0.71. The test results have an important application aspect and may be useful at the stage of assessing the adhesion of adhesive coatings in a non-destructive manner.

1. Introduction

The production as well as the repair process of motor vehicles after road accidents and collisions includes a number of stages, during which various technologies for joining materials are used, such as bonding [1], spot welding [2], welding using various methods [3,4,5], and self-piercing riveting [6]. An equally important stage of body renovation is the reconstruction of adhesive coatings. This has both a decorative purpose (improving the visual effect) and allows to protect the car body against corrosion phenomena [7,8]. During the production stages of adhesive coatings, various defects may occur, affecting their operational durability [9]. These defects include uneven thickness [10], excessive brittleness of coatings [11], contamination in the coating [12], as well as delamination of individual layers [13] and poor adhesion to the steel substrate [14]. The last of the defects is of significant importance from the point of view of the previously mentioned decorative and protective effect of the car body. Adhesion is defined as the cohesion, bonding, or adhesion of the coating to the substrate [15]. It depends mainly on the method of preparing the substrate surface, as well as factors that act during the application of the coating [16]. In the case of the mechanical theory of adhesion, formed at the stage of preparing the steel substrate, the phenomenon of anchoring the material on macro-irregularities, as well as on micro- and sub-micro-irregularities of the surface, occurs primarily, which results in obtaining the appropriate strength of the connection.

The assessment of the adhesion of coatings to the substrate can be carried out using two different groups of methods: destructive [17] and non-destructive [18]. Methods from the first group cause damage to the coating at the place of testing, which is why they are rarely used. Most scientists and people performing coating adhesion tests require further use of the coating after the adhesion test. Therefore, non-destructive testing methods are desirable, which can provide information on the condition of the connection without interfering with its structure [19]. Among the known non-destructive testing methods used to assess the adhesion of coatings to the substrate, the following can be distinguished: thermographic [20], eddy current [21], radiographic [22], and ultrasonic [23] methods.

The thermographic method consists of analyzing the heat flow through the contact point of the coating with the substrate (or other adhesive connection) during its heating. Most often, heating is performed from the side of the steel sheet using a short thermal pulse, and from above, using a thermal imaging camera, the temperature distribution is monitored and the heat is observed to move through the tested connection. This allows for the detection of “cold zones” areas where the adhesion of the coating to the substrate is weak or completely absent. This technique is particularly effective in the assessment of thin layers on large surfaces, enabling quick and non-invasive detection of defects in the form of weakened adhesion of the coating to the substrate [24].

The radiographic method in the examination of adhesive joints, including the coating with the substrate, consists primarily in the X-raying of the tested area using ionizing radiation, and then recording and analyzing the obtained images, which reveal the presence of possible discontinuities or defects. Various types of radiation are used for X-raying, such as X-rays, gamma radiation, or alpha radiation. This technique allows primarily for the identification of porosity and cracks occurring in the connection between the coating and the substrate [25].

Ultrasound is a mechanical wave with a frequency above 20 kHz [26] that is introduced into the area of adhesive joints in order to determine their specific properties. One of such technological properties is the adhesion of the coating to the substrate. The strength of an adhesive joint, also known as adhesion, is related to the normal and tangential stresses transferred by such a joint. The normal stresses occurring in the joint can be measured in a frontal peel test by measuring the force necessary to destroy the adhesive joint related to the surface on which it is located [27]. The determination of this force allows us to set the correlation between non-destructive measures of adhesion and the stresses destroying the adhesive joint. Ultrasonic tests of adhesive joints are carried out using surface waves [15,28], longitudinal [29], transverse [30], Lamb [31] and Stoneley’s wave [32]. Ultrasonic testing of an adhesive joint using perpendicular incidence of an ultrasonic beam of a longitudinal wave causes it to cross two media with different acoustic wave resistances z. After reaching the interface, part of the wave beam passes to the second medium (e.g., coating), and part is reflected and returns to the transmitter–receiver head. Based on the acoustic wave resistances calculated from dependence (1), the reflection coefficient |r| of the longitudinal wave (dependence (2)) can be determined [33].

$$\mathrm{z}=\mathsf{\rho }\times \mathrm{c}$$

(1)

where

z—acoustic wave resistance;

$\mathsf{\rho }$—density of the medium;

c—the speed of the ultrasonic wave in the medium.

$$\left|\mathrm{r}\right|={\displaystyle \frac{{\mathrm{z}}_2-{\mathrm{z}}_1}{{\mathrm{z}}_2+{\mathrm{z}}_1}}$$

(2)

where

|r|—reflection coefficient;

z₁—acoustic wave resistance of the medium 1 kg/m²*s;

z₂—acoustic wave resistance of the medium 2 kg/m²*s.

This reflection coefficient can also be determined by taking into account ultrasonic measurements before and after applying the coating—the so-called total reflection method [34]. The sample before applying the coating is treated as a standard. The ultrasonic wave propagates in the substrate material and is reflected from the boundary surface, which is a free surface. The reflection coefficient takes a value equal to unity. In the case of measurements with a coating applied, an echo of the surface limiting the tested joint is obtained, and the ratio of the boundary echoes obtained during the first and second stages of measurements is designated. Based on the changes in the amplitude of the first pulse of the ultrasonic wave, the adhesion value is calculated in the form of a reflection coefficient.

The main objective of this article is to determine the correlation between mechanical adhesion and the reflection coefficient of the longitudinal wave. Determination of these relationships will allow for non-destructive estimation of the adhesion of the adhesive coating to the steel substrate based on the changes in the |r| coefficient. This will fill the gap in knowledge in the field of non-destructive diagnostic methods used to assess the adhesion of the coating layer to the substrate. The results can be used in bodywork and paintwork workshops for vehicle bodywork during the verification of the quality of the adhesive joint between the steel sheet and car putty coating using the ultrasonic method. The methods and research results proposed in the article will allow for the verification of areas with reduced adhesion. That is important from the point of view of the operation of motor vehicles, especially those that have undergone bodywork and paintwork repairs.

2. Materials and Methods

The correlation tests of coating adhesion to the substrate were carried out according to the scheme shown in Figure 1. At the beginning, 15 disc specimens were selected, made of St3 steel of equal dimensions (diameter 50 mm, height 25 mm). The specimens had an extendable pin in the centre with a partially threaded side surface at the bottom (Figure 2). The threaded surface not only allowed the realization of destructive testing, but also caused the dissipation of possible additional ultrasonic waves that could be generated during testing, by ultrasonic methods. The surfaces of the specimens were then prepared using three different methods in order to vary the surface roughness profile, which affected different adhesion of the coating to the substrate. For this purpose, the following operations were used: sandblasting (series 1), grinding with P240 sandpaper (series 2), and laser beam surface treatment (series 3).

During the sandblasting process of the samples, a ceramic nozzle with a diameter of 4 mm was used, and the working pressure was 5 bar. The surface preparation of the samples in series 2 was performed with an eccentric grinder with a sandpaper disc applied (paper designation P240).

The surface of the last group of samples was prepared with a special AccTek model AKQ-3000 laser (Jinan AccTek Machinery, Jinan, China). Laser cleaning systems show high efficiency in the treatment of a variety of materials, allowing efficient removal of corrosion from metal parts, paint coatings, varnishes, and other contaminants and layers. The laser cleaning procedure, also known as laser ablation, leads to degreasing the machined surface and also nullifies the effects of oxidation and carbon build-up. The maximum power of the laser beam (3000 W) was used. The width of the cleaning path was 60 mm—larger than the width of the disc samples. Four passes of the laser beam were realized on each sample, which cleaned the surface. The samples, after surface preparation, were shown in Figure 3.

Each group of samples was prepared using a different processing method. This was performed in order to create different conditions for the car putty application. Roughness is a multi-parameter surface feature used to determine its geometric condition, which was verified at randomly selected points of whole samples (15 measurements—one on each sample). The tests were performed on a Taylor Hobson Surtronic 3+ profilographometer (Sutronic, Wegalaan, The Netherlands). The basic parameters of the surface roughness profile were studied, such as:

R_a—average surface roughness for the length of the measurement;

R_z—an average of the five largest total deviations recorded for length.

This allowed for verification of the influence of the values of the basic parameters of the surface roughness profile on the mechanical adhesion of the coating to the substrate and the propagation of the ultrasonic wave in the adhesive bonding area.

Ultrasonic testing was carried out in two separate stages:

• Stage 1—measurement of the amplitude of the first pulse (bottom echo—h₁) on the specimen—specifically, the specimen pin (Figure 4a);
• Stage 2—measurement of the amplitude of the first pulse h, from the boundary of the coating-substrate interface (Figure 4b).

In both the first and second stages, the ultrasonic head was applied to the pin from the side where the coating was not applied. In the first stage, the bottom echo was obtained due to the reflection of the longitudinal ultrasonic wave from the surface to which the putty coating was then applied (the boundary of the steel-air media). The second stage involved measurements for the joint. In this case, the propagating ultrasonic wave was partially reflected from the boundary of the coating-substrate interface and partially propagated into the coating and was attenuated in it. In both the first and second stages, 3 ultrasonic measurements were made for each sample (it was estimated from the previous 20 measurements for one sample). On the surface to which the head was applied, 5 mL of coupling gel was applied at each measurement, and the head itself was loaded with a piece of steel (around 200 g). This resulted in an even and constant pressure of the head on the sample with each ultrasonic measurement. The tests were carried out on a USM 35XS ultrasonic flaw detector and a DS 12 HB 1–6 (6 MHz) ultrasonic head (both made by Krautkramer, Cologne, Germany). The diameter of the ultrasonic transducer (12 mm) was minimally smaller than the diameter of the pin (15 mm), which allowed testing in a large part of the area of this sample. Changes in the amplitude of the ultrasonic wave pulse (stages 1 and 2) allowed the determination of the ultrasonic wave pulse amplification difference according to the following relationship:

$$∆\mathrm{W}=20\times \log {\displaystyle \frac{{\mathrm{h}}_1}{\mathrm{h}}}$$

(3)

where

ΔW—amplification of the ultrasound wave pulse;

h₁—the amplitude of the first (highest) pulse during the first stage of the measurement;

h—the amplitude of the first pulse (the highest) during the measurement of the second stage—after the application of the coating.

Then, taking into account Equation (4) and the ultrasonic longitudinal wave pulse amplification difference, the reflection coefficient |r|, which is a non-destructive measure of coating adhesion to the substrate, was determined.

$$\left|\mathrm{r}\right|={10}^{-{\displaystyle \frac{∆\mathrm{W}}{20}}}$$

(4)

where

|r|—reflection coefficient;

ΔW—amplification of the ultrasound wave pulse.

Between the two stages of ultrasonic testing, a car putty coating was applied on the sample surface. The Novol 200 hybrid car putty coating (Novol, Komorniki, Poland) was used. This was a two-component polyester car putty with a wide range of applications, which combines the properties of glass microfibers and synthetic microsphere fillers. This combination gives it an extremely low level of shrinkage, making it ideal for filling extensive damage. The car putty coating was applied directly to the degreased surface (after prior surface treatment). To facilitate application, as well as to maintain a uniform layer thickness (1 mm) on the specimens, the sleeve shown in Figure 5a was used. This maintained reproducibility, as well as speeding up the process and allowing for application of the putty simultaneously to the entire group of specimens in a time not exceeding the setting time, which is 4 to 6 min at 20 °C. Samples with the putty coating applied are shown in Figure 5b.

The next stage of testing included a pull-off test on a testing machine using a special holder (Figure 6). During the test, the jaw feed rate of the testing machine was used at 25 mm/min and a single destructive test lasted up to 30 s. In each measurement, the maximum force required to detach the coating from the substrate was recorded. Thus, using the surface area of the pin, the destructive stress of the coating-substrate interface was calculated, which was taken as the mechanical adhesion of the coating to the substrate.

The final stage of the research involved comparing the results of the pull-off test with the results of ultrasonic testing, which was the reflection coefficient of the longitudinal wave. A correlation was sought between these parameters, depending on the sample preparation and surface roughness profile of the samples, taking into account the method of sample surface treatment.

3. Results

The results of the ultrasonic test (amplitude of the first pulse from stage I and II) depending on the method of sample surface preparation are shown in

Table 1. Ultrasonic test results for samples surface prepared by sandblasting.

	Pulse Amplitude—Stage I				Pulse Amplitude—Stage II
	I	II	III	Average	I	II	III	Average
1	43	51	49	48	38	25	30	31 *
2	38	41	46	42	32	35	39	35
3	40	38	36	38	31	29	35	32
4	53	45	48	49	26	31	28	28
5	42	34	37	38	27	30	31	29
6	42	34	36	37	32	26	28	29
7	55	53	58	55	27	31	22	27
8	58	64	51	58	34	36	37	36
9	42	38	45	42	33	35	29	32
10	51	47	48	49	32	34	31	32
11	53	48	37	46	30	28	34	31
12	54	44	51	50	29	38	34	34
13	37	42	44	41	28	32	30	30
14	43	37	41	40	29	29	31	30
15	46	50	42	46	31	37	35	34
average				45				31
standard deviation				6				3

* Decohesion of coating—sample not considered in analysis of results.

,

Table 2. Ultrasonic test results for samples surface prepared by grinding with P240 sandpaper.

	Pulse Amplitude—Stage I				Pulse Amplitude—Stage II
	I	II	III	Average	I	II	III	Average
1	24	26	31	27	21	18	25	21
2	31	34	29	31	27	26	22	25
3	23	25	29	26	18	20	23	20
4	38	43	40	40	30	36	35	34
5	31	32	35	33	28	25	27	27 *
6	33	38	44	38	28	36	33	32
7	33	37	34	35	27	25	31	28
8	26	30	31	29	24	21	20	22
9	26	27	25	26	20	23	22	22
10	31	29	27	29	27	23	25	25
11	12	14	16	14	8	10	11	10
12	32	36	32	33	27	30	28	28
13	26	27	27	27	19	23	21	21
14	44	37	39	40	34	33	30	32
15	28	30	29	29	25	25	23	24
average				30				25
standard deviation				6				6

* Decohesion of coating—sample not considered in analysis of results.

and

Table 3. Ultrasonic test results for samples surface prepared by laser beam treatment.

	Pulse Amplitude—Stage I				Pulse Amplitude—Stage II
	I	II	III	Average	I	II	III	Average
1	64	61	67	64	37	47	45	43
2	59	63	69	64	46	48	45	46
3	62	61	60	61	51	49	49	50
4	67	64	66	66	49	52	55	52
5	61	64	59	61	53	50	49	51
6	63	53	59	58	45	42	44	44
7	63	54	57	58	45	42	43	43
8	63	66	59	63	45	47	48	47
9	66	66	67	66	52	49	47	49
10	64	67	72	68	53	51	54	53
11	64	66	68	66	46	53	49	49
12	57	59	62	59	45	46	48	46
13	63	64	66	64	53	47	51	50
14	63	66	57	62	44	48	48	47
15	62	66	64	64	44	49	55	49
average				63				48
standard deviation				3				3

. For one sample in each stage three measurements were conducted. Meanwhile,

Table 4. The roughness profile parameters value for different surface treatment.

Type of Surface Treatment
Sandblasting																Average	Standard Deviation
Ra	3.11	3.43	3.13	3.41	3.38	2.7	3	2.74	2.61	3.07	2.99	2.77	3.04	2.65	2.59	2.97	0.28
Rz	16	17.1	14.1	16.6	13.8	13.2	14.2	13.7	12.5	15.1	13.6	12.9	17	16.6	12.8	14.61	1.58
P240—sandpaper
Ra	1.12	1.03	0.81	1.23	0.83	1.08	0.43	1.19	0.97	0.66	1.11	0.94	0.99	1.01	1.04	0.96	0.20
Rz	6.46	5.78	5.16	7.4	6.12	6.56	2.71	6.3	6.12	4.23	6.84	5.01	5.74	6.93	6.15	5.83	1.14
Laser beam treatment
Ra	1.11	1.70	1.37	1.20	0.96	1.14	1.20	0.58	1.33	0.70	1.12	1.19	1.09	1.16	1.22	1.14	0.25
Rz	7.30	9.80	8.18	7.28	6.60	7.18	8.73	4.31	7.60	4.31	7.12	7.33	6.99	7.15	7.41	7.15	1.35

summarizes the values of R_a and R_z (the basic parameters of the surface roughness profile of the samples) depending on the way their surfaces are prepared.

The average value of the pulse amplitude considering the different methods of surface preparation and stages of ultrasonic testing (stages I and II) is the highest for samples prepared by laser beam treatment and is 63% and 48% of the screen height of the ultrasonic flaw detector, respectively. The lowest average results of ultrasonic pulse amplitudes are observed for samples prepared by grinding with P240 sandpaper (average 30% and 25% of the screen height). The technological operation of sandblasting gives intermediate results (45% and 31%). The standard deviation is highest for specimens subjected to the grinding operation, and in the second stage of ultrasonic measurements, is 6%, indicating a high variability of results in this set of ultrasonic measurements. Laser beam treatment has the lowest standard deviation, indicating the high repeatability and uniformity of the ultrasonic pulses. For both sandblasting and laser beam surface treatment of the samples, the average pulse amplitude is significantly higher at stage I compared to stage II (14% and 15% difference), which shows significant changes in the pulse amplitude of the ultrasonic wave in stage II of the tests. This may indicate better adhesion of the coating to the substrate—a large part of the ultrasonic wave “flowed” into the coating and was attenuated there (this is evidenced by a significant difference in the amplitude of the ultrasonic wave pulse between stages I and II). With P240 sandpaper grinding, the difference is much smaller (5%), suggesting a less pronounced change between two stages of ultrasonic measurement.

Taking into account the table below and the obtained results of the roughness profile, it should be concluded that sandblasting gives a surface with the highest roughness profile (R_a about 3 µm, R_z about 15 µm), which is characteristic of this method—during surface processing, significant irregularities are formed and the surface is rough. In contrast to sandblasting, grinding with P240 paper produces a much smoother surface (R_a less than 1 µm, R_z about 6 µm). A low roughness indicates a lower risk of surface defects, but at the same time should affect the adhesion of the coating to the substrate—smaller roughness means a smaller contact area between the car putty and the sample and, consequently, can cause a reduced value of mechanical adhesion. Laser beam surface treatment generates roughness profile parameters between the results obtained from the technological operation of sandblasting and grinding with P240 sandpaper (R_a about 1.14 µm, R_z about 7.15 µm). This method can provide a controlled surface structure and repeatability of the technological operation. The standard deviations are relatively small for all methods, indicating the consistency and repeatability of the measurement results. In addition, it should be noted that the Rz ≈ 4 × Ra relationship, which is known from the literature [35], was achieved, confirming the correctness of the measurements and the consistency of the results for typical roughness parameters.

Summarizing all the results of both ultrasonic and surface roughness profile tests, it should be concluded that higher surface roughness (R_a, R_z) is associated with higher ultrasonic wave pulse amplitude (pulse height) in stage I and stage II of ultrasonic measurements. This is the case for sandblasting, where the highest R_a and R_z values (about 3 and 14.6 µm) correspond to an average ultrasonic amplitude of 45 (stage I) and 31 (stage II). Laser beam processing generates roughness that results in the highest ultrasonic wave pulse amplitudes (63 and 48 on average). This means that not only the roughness itself but also the nature of the surface (e.g., regularity, uniformity of microstructures) can affect the reflection of ultrasonic waves and their amplitude. The laser beam can create surfaces with more homogeneous and more favourably oriented microstructures, which improves ultrasound reception despite lower roughness than with sandblasting.

The lowest values of R_a and R_z correspond to the lowest values of the pulse amplitude of the ultrasonic wave. Grinding produces a relatively smooth surface, which may result in weaker penetration of the ultrasonic wave into the adhesive coating.

For each method of sample surface preparation, the amplitude of the wave pulse decreases from stage I to stage II, which is consistent with the mechanism of changing surface or material layer properties between stages.

In ultrasonic testing, wave reflection depends not only on roughness measures such as R_a and R_z, but also on contact stiffness and surface microstructure characteristics (e.g., homogeneity and roughness orientation). Therefore, it is most likely that higher ultrasonic wave pulse values were obtained for samples whose surface was treated with a laser beam, which are characterized by moderate roughness, than for grinding operations with paper of lower roughness.

4. Result Analysis and Discussion

Completed in two stages, the ultrasonic and pull-off tests allowed the development of data in the form of graphs (Figure 7, Figure 8 and Figure 9), which show the reflection coefficient |r| of the longitudinal wave in relation to the adhesion of the coating to the substrate, defined in MPa.

Considering the data presented in the figures, it should be noted that the results are similar in value. For samples whose surface was sandblasted, the adhesion is in the range of about 1.6 to 4.3 MPa, which corresponds to reflection coefficient values of 0.5–0.8. Higher values of coating adhesion to the substrate correspond to lower |r| coefficient values. Samples ground with P240 paper achieved mechanical adhesion of about 1.4 to 3.3 MPa. At the same time, the values of the reflection coefficient ranged from about 0.7 to 0.9. In the last of the series of samples tested, in which surfaces were treated with a laser beam, the values of coating adhesion to the substrate in the range of 1.5–4.4 MPa were obtained.

Samples after sandblasting have the highest average adhesion strength of 2.74 MPa, indicating better adhesion of the coating to the substrate than samples in other measurement series. The average reflection coefficient value (0.71) is the lowest, which is in line with trends indicated by other researchers [36,37]—the lower the reflection coefficient value, the higher the value of mechanical adhesion of the coating to the substrate.

Treating the surface of the samples with a laser beam resulted in more varied results. The average peel stress is 2.71 MPa, which corresponds to an average coefficient reflection value of 0.76.

The P240 sanding causes a slightly lower adhesion than sandblasting (2.23 MPa on average), which may be due to the lower roughness and less development of the surface under the coating into which the car putty penetrates during application. The reflection coefficient values are higher relative to the sandblasted samples, indicating weaker adhesion of the coating to the substrate (0.81).

Summarizing the obtained test results, it can be concluded that for all tested samples, regardless of the method of surface preparation, as the value of reflection coefficient |r| decreases, the adhesion of the coating to the substrate increases. This relationship is consistent with the work of authors [15,38]. Moreover, for the surface prepared with a laser beam, a large scatter of results and a low correlation coefficient were obtained. Such a result indicates that the studied characteristic (reflection coefficient) does not clearly affect the other one (adhesion) in linear terms. In the case of sandblasting and paper sanding, much higher correlation coefficients were obtained (about 0.7–0.8), showing that there is a good linear relationship between these variables, and changes in one can be predicted to a good approximation from changes in the other. On this basis, using the determined correlations, the mechanical adhesion of the coating to the substrate based on the reflection coefficient |r| can be estimated.

5. Conclusions

Based on the realized correlation studies comparing the reflection coefficient of the longitudinal wave with the mechanical adhesion of the coating to the substrate, the following statements can be made:

• Regardless of the type of surface preparation of the samples (sandblasting, sandpaper grinding, and laser beam treatment), as the mechanical adhesion of the coating to the substrate increases, the value of the reflection coefficient |r| decreases.
• The highest average mechanical adhesion of the coating to the substrate was obtained for sandblasted samples and was 2.74 MPa, corresponding to a coefficient reflection value of 0.71, while the lowest mechanical adhesion value of 2.23 MPa and the highest average reflection coefficient value of 0.81 were obtained for samples ground with P240 sandpaper.
• The selection of a suitable surface preparation method should take into account the need to achieve synergy between surface structure (surface roughness profile) and adhesion strength, so the optimal substrate preparation technique should be tailored to the type of coating, the material of the substrate, and the operating conditions of the adhesive coating-substrate connection.

In the next stage of the study, the authors intend to compare the values of the reflection coefficient of the longitudinal wave determined by different methods with the mechanical adhesion of the coating to the substrate. In addition, it is planned to use other adhesive coatings to confirm the obtained trends in terms of changes in the reflection coefficient |r| with changes in the parameters of the surface roughness profile and mechanical adhesion. This will expand the knowledge of non-destructive testing of coating adhesion to the substrate using the ultrasonic method.