G. BAR[I] et al.: ACCURACY OF REPLICATION OF SURFACE TEXTURE WITH A CONTACT PROFILE METHOD 41–46 ACCURACY OF REPLICATION OF SURFACE TEXTURE WITH A CONTACT PROFILE METHOD NATAN^NOST PONOVITVE POVR[INSKE TEKSTURE S KONTAKTNO PROFILNO METODO Gorana Bar{i} * , Bernard Nuji}, Vedran [imunovi} Faculty of mechanical engineering and naval architecture, University of Zagreb, Ivana Lu~i}a 1, 10000 Zagreb, Croatia Prejem rokopisa – received: 2023-03-07; sprejem za objavo – accepted for publication: 2023-12-07 doi:10.17222/mit.2023.823 Surface texture is of great importance in specifying the function of a surface. The surface finish of products influences the fric- tion performance, wear resistance, bearing, sliding, lubricating properties, functionality, etc. Replicas of surface textures are widely used in many industries for evaluating the condition of textures when no measurable surfaces are available for measure- ments or in the case of non-dismantled objects of immense dimensions. The replication’s reliability and accuracy are significant concerns in actual applications. Although there are several parameters, Ra and Rz are still most often defined in technical prod- uct documentation as the indicators of whether a surface texture is within certain tolerance limits. In real production conditions, tactile stylus profilometers are still the most common instruments for measuring surface texture. For all the reasons mentioned above, the accuracy of the Technovit ® 3040 replica was analysed. The quality of the Technovit ® 3040 replica was researched on 15 samples processed with six conventional machining methods. Average roughness parameters Ra and Rz were measured on original surfaces and then compared to replica surfaces. A qualitative comparison of the profile geometries of original surfaces and their replicas was discussed. Keywords: surface texture, stylus profilometers, replica, accuracy, Technovit ® 3040 Tekstura ali morfologija povr{ine je zelo pomembna za dolo~itev njene funkcije. Fini{ oziroma gladkost povr{ine vpliva na trenje, odpornost proti obrabi, nosilnost, drsenje, mazanje, funkcionalnost itd. Replike razli~nih povr{in se pogosto uporabljajo v mnogih industrijah za ovrednotenje stanja teksture, ko so povr{ine ki jih `elimo ovrednotiti ali izmeriti te`ko dosegljive, ali v primeru odstranitve zelo velikih objektov. Zanesljivost repliciranja in njena natan~nost sta zelo pomembni za dejanske aplikacije. Na razpolago imamo nekaj parametrov hrapavosti povr{ine kot sta naprimer Ra in Rz s katerimi najbolj pogosto opi{emo povr{ino v produktni dokumentaciji za ovrednotenje povr{ine, ali je ta oziroma ni znotraj zahtevanih toleranc. V realnih pogojih produkcije (proizvodnje) so profilometri s tipalom, pisalom oziroma tanko iglo, ki drsi po povr{ini (angl.: tactile stylus profilometers) najbolj pogosti in{trumenti za merjenje teksture povr{ine. Avtorji zaradi vseh razlogov opisanih zgoraj v tem ~lanku opisujejo analizo natan~nosti replik izdelanih z maso Technovit ® 3040, ki je hitro su{e~a smola na osnovi metil metakrilata. Avtorji so kakovost replik izdelanih s Technovitom ® 3040 ocenjevali na 15-tih vzorcih. Replike so odvzeli s povr{in, ki so bile izdelane s {estimi klasi~nimi postopki mehanske obdelave (stru`enje, ploskovno in bo~no bru{enje, lepanje in rezkanje). Na originalnih povr{inah in na replikah teh povr{in so dolo~ili parametra hrapavosti Ra in Rz ter jih nato primerjali med seboj. V ~lanku nato razpravljajo in kvalitativno primerjajo profilne geometrije originalnih povr{in z njihovimi replikami. Klju~ne besede: povr{inska tekstura, profilometri s pisalom, replika oz. kopija, primerjava, natan~nost, Technovit ® 3040 1 INTRODUCTION Based on the results obtained from a replica, we can determine the condition of surface roughness of the orig- inal surface. Therefore, it is essential to reliably deter- mine the quality of the achieved reproduction. There are different replica materials currently available on the mar- ket. They differ with respect to the material, the applica- tion method and the type of the measuring instrument to be applied. F. Baruffi et al. investigated the reproduction quality of the commercial silicone replica material RepliSet, manufactured by Struers GmbH, on two specifically de- signed micro-milled samples made from mould steel. 1 The analysis was based on comparing the values of areal parameters Sa, Svk, Spk, and Sk, measured using a confo- cal microscope. A qualitative comparison between the original surface and their replicas showed that a replica could accurately reproduce the appearance of the mi- cro-milled surface texture. A numerical comparison re- vealed that the measurements on replicas had consis- tently overestimated the average roughness of the original surface. However, the analysis of 34 cases out of 36 showed that the replication media RepliSet is suitable for characterising micro-milled surfaces when using a non-contact measuring instrument. P. Szawara and R. Ostrowski analysed the suitability of RepliSet F5, manufactured by Struers GmbH, to repli- cate the surface textures of a shaft, tool and grinding wheel. 2 The measurements were carried out using a focal variation microscope and a tactile stylus profilometer. Analysis was focused on the comparison of roughness parameters Ra and Rz. The authors concluded that RepliSet F5 is not suitable for evaluating the wheel po- rosity of the grinding wheel used in the research. Due to its high viscosity and fluidity, it penetrates too deep into Materiali in tehnologije / Materials and technology 58 (2024) 1, 41–46 41 UDK 551.4.03:53.088.3 ISSN 1580-2949 Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 58(1)41(2024) *Corresponding author's e-mail: gorana.barsic@fsb.hr (Gorana Bar{i}) the pores and affects the results. However, comparing roughness parameters Ra and Rz obtained on original surfaces and replicas of four shafts showed the RepliSet F5 suitability for replicating the original surface accu- rately. Y. C. Liu et al. investigated three replica materials – Repliset, Technovit and Press-O-Film on four ground surfaces with average roughness values ranging from 0.2 μm to 1.6 μm. 3 Measurements were performed using a tactile stylus profilometer and focus optical micro- scope. Since RepliSet and Press-O-Film replicas are soft, they were measured using a non-contact method, i.e., a focus optical microscope. Technovit replicas are hard and suitable for measurements using a contact method – a tactile stylus profilometer. The obtained results showed that on the Technovit replica, the Ra deviation was within 5 % throughout the analysis measurement range. RepliSet and Press-O-Film replicas exhibited high accu- racy in replicating surfaces with Ra ranging from 0.4 μm to 1.6 μm; however, they were not suitable for a smooth surface with Ra at the level of 0.2 μm. A literature review revealed that very few published papers analyse the quality and accuracy of replicas in surface roughness metrology, especially replicas that can be measured with contact methods. It is well known that different machining processes result in different, but for each process characteristic, surface topographies. The typical topography of an origi- nal surface can influence the quality of surface reproduc- tion. Therefore, in this study, an analysis was carried out on surfaces processed with six different conventional machining processes. In real production conditions, tactile stylus profilo- meters are the most widely used instruments for evaluat- ing the surface roughness and roughness parameters Ra and Rz are commonly used to measure the surface tex- ture. Therefore, in this study, measurements of roughness parameters Ra and Rz were carried out using a tactile stylus profilometer. Most replicas have properties that can only be mea- sured with non-contact methods. One of the rare replica materials, which can be measured, according to its man- ufacturer Kulzer GmbH, with a contact method, is Technovit ® 3040, a two-component resin based on methyl methacrylate. 4 Therefore, the above product was applied to 15 samples produced with six conventional machining processes as the material for making replicas. 2 EXPERIMENT Samples were selected from five measurement sur- faces (N2, N4, N6, N8 and N10) from three different groups processed with conventional machining (groups A, B and C) as presented in Figure 1. In group A, the N2, N4 and N6 surfaces were processed by grinding, while N8 and N10 were processed with a side milling procedure. In group B, the N2 and N4 surfaces were pro- cessed by lapping, while N6, N8 and N10 were pro- cessed with a face milling procedure. And in group C, the N2 and N4 surfaces were processed with super- finishing, while N6, N8 and N10 were processed with a turning procedure. According to the instructions by the manufacturer Kulzer Technik, the Technovit ® 3040 replica material was prepared and applied to the selected surfaces. 5 Tech- novit ® 3040 is a two-component methyl methacrylate resin based on a powder and a liquid component. In this experiment the ratio of powder to liquid was 2:1. A plexiglass mould was made, as shown in Figure 2,tofa- cilitate the application of the prepared replica material. After the replica material had been mixed and poured on measurement surfaces, five minutes were taken to cure the material. Replicas were detached easily from the original surfaces. The replicas of the surfaces from Fig- ure 1 are presented in Figure 3. G. BAR[I] et al.: ACCURACY OF REPLICATION OF SURFACE TEXTURE WITH A CONTACT PROFILE METHOD 42 Materiali in tehnologije / Materials and technology 58 (2024) 1, 41–46 Figure 1: Original surfaces Sign ’ was added to distinguish the surface replicas from the original surfaces. Measurements were per- formed in the Croatian National Laboratory for Length (FSB-LPMD) using stylus instrument Perthometer S8P, manufactured by Mahr Perthen. The stylus instrument was traceably calibrated using the Croatian state stan- dards for roughness. During the measurement, a stylus with a 5 μm radius was traversed normally to the surface at a constant speed and measuring force of 1.3 mN. Fil- tering was done using a Gaussian filter with the cut-off value of c and evaluation length ln as given in Table 1. Table 1: Measurement conditions Surfaces c ,mm ln,mm N2, N2’ 0.25 1.25 N4, N4’ 0.8 4.0 N6, N6’ 0.8 4.0 N8, N8’ 0.8 4.0 N10, N10’ 2.5 12.5 Cut-off values and evaluation lengths were selected according to the texture conditions on all surfaces except for N10 and N10’ due to the sample size limit, on which no sufficient trace length could be recorded. However, when measuring the original surfaces and their replicas the same measurement conditions were applied, and reproducibility conditions were fulfilled. 3 RESULTS AND DISCUSSION Using the measurement conditions defined in the pre- vious section, five roughness profiles were taken on each measurement surface, and the arithmetical means of roughness parameters Ra and Rz were calculated and presented in Table 2. The results obtained on the original surfaces are marked as N2 to N10, and the results ob- tained on their replicas are denoted as N2’ to N10.’ Table 2: Arithmetical means of Ra and Rz parameters for original sur- faces and replicas Group A Group B Group C Ra (μm) Rz (μm) Ra (μm) Rz (μm) Ra (μm) Rz (μm) N2 0.0338 0.3036 0.0514 0.4438 0.0430 0.2784 N2’ 0.0458 0.3158 0.0540 0.3782 0.0448 0.2856 N4 0.1622 1.1660 0.1538 1.3638 0.1438 1.5522 N4’ 0.1706 1.2346 0.1914 1.4636 0.1982 1.6516 N6 0.6828 4.5984 0.7642 4.6392 0.6736 3.3676 N6’ 0.7514 4.8994 0.7162 4.3024 0.6628 3.3658 N8 1.3004 7.2014 3.0340 16.4480 3.3038 12.3800 N8’ 1.4986 8.1006 2.8760 16.8300 3.2380 13.6760 N10 14.426 54.634 14.102 64.924 14.530 58.526 N10’ 10.346 49.412 12.572 65.744 13.512 55.548 From the results in Table 2, relative deviations were calculated and presented in Table 3. Table 3: Relative deviations of Ra and Rz parameters for original sur- faces and replicas Group A Group B Group C Ra (%) Rz (%) Ra (%) Rz (%) Ra (%) Rz (%) N2 30.2 3.9 4.9 15.9 4.1 2.6 N4 5.1 5.7 21.8 7.1 31.8 6.2 N6 9.6 6.3 6.5 7.5 1.6 0.1 N8 14.2 11.8 5.4 2.3 2.0 9.9 N10 32.9 10.0 11.5 1.3 7.3 5.2 G. BAR[I] et al.: ACCURACY OF REPLICATION OF SURFACE TEXTURE WITH A CONTACT PROFILE METHOD Materiali in tehnologije / Materials and technology 58 (2024) 1, 41–46 43 Figure 2: Plexiglass mould Figure 3: Surface replicas The results from Table 3 indicate minor relative devi- ations of the values of parameter Rz compared to relative deviations of the values of parameter Ra. Apart from Rz obtained on the N2 surface of Group B (lapping) and N8 surface from Group A (side milling), relative devia- tions of Rz parameter are equal to or below 10 %. Two surfaces from GroupA–N2surface (grinding) and N10 surface (side milling) – and one surface from GroupC–N4surface (superfinishing) – had Ra val- ues above 30 %. Figures 4 and 5 give graphical presen- tations of relative deviations of the measured roughness parameters. The relative deviations of parameters Ra and Rz indi- cate that the accuracy of the replicas is not dependent on the type of machining process, nor the level of surface finish defined by the N grade ranging from N2 to N10. To determine whether the uniformity of the texture of a replica was preserved compared to the original surface, a comparison of the ranges of Ra and Rz values obtained on five roughness profiles on each sample was made. The obtained results are presented in Table 4. Table 4: Ranges of Ra and Rz parameters on original surfaces and rep- licas Ranges (μm) N grade Group A Group B Group C Origi- nal Replica Origi- nal Replica Origi- nal Replica R (Ra) N2 0.001 0.006 0.004 0.005 0.007 0.005 R (Ra) N4 0.015 0.018 0.015 0.012 0.014 0.006 R (Ra) N6 0.101 0.055 0.255 0.023 0.030 0.016 R (Ra) N8 0.339 0.151 0.521 0.340 0.120 0.150 R (Ra) N10 2.29 1.4 5.08 1.16 0.48 0.14 R (Rz) N2 0.066 0.105 0.068 0.065 0.220 0.042 R (Rz) N4 0.128 0.239 0.358 0.094 0.181 0.046 R (Rz) N6 0.270 0.972 1.320 0.159 0.154 0.172 R (Rz) N8 2.058 1.788 2.350 3.740 0.430 4.900 R (Rz) N10 4.61 3.99 17.93 2.51 1.45 0.66 G. BAR[I] et al.: ACCURACY OF REPLICATION OF SURFACE TEXTURE WITH A CONTACT PROFILE METHOD 44 Materiali in tehnologije / Materials and technology 58 (2024) 1, 41–46 Figure 4: Relative deviations of Ra parameter Figure 5: Relative deviations of Rz parameter As expected, by increasing the N grade, range values are also increasing. However, the values of the range, for the Ra parameter and Rz parameter, do not show any trend in terms of increasing or decreasing their values achieved on the replica surfaces as compared to the orig- inal surfaces. Both parameters are reasonably close for the original surfaces and their replicas. The geometries of roughness profiles in Table 5 also support this conclu- sion. From the roughness profiles presented in Table 5,a good reproduction of characteristic geometries of rough- ness profiles for different machining processes was de- termined. This corroborates that the applied replica ma- terial did not violate the surface textures. 4 CONCLUSIONS Surface texture is of great importance in specifying the function of a surface. The surface finish of a product obtained with a manufacturing process influences the friction performance, wear resistance, bearing, sliding, lubricating properties, functionality, etc. Therefore, sur- face properties are checked to be within certain rough- ness limits in the manufacturing industry. The surfaces outside these limits would result in failures in the perfor- mance of the product. When surfaces are challenging to access with mea- suring instruments or when parts are difficult to be dis- mantled for measurement, the only way to determine the condition of a surface texture is to apply a replica. In that case, it is necessary to have quantitative information about the accuracy of the reproduction. G. BAR[I] et al.: ACCURACY OF REPLICATION OF SURFACE TEXTURE WITH A CONTACT PROFILE METHOD Materiali in tehnologije / Materials and technology 58 (2024) 1, 41–46 45 Table 5: Roughness profiles on original surfaces and replicas Group A Group B Group C N2 N2’ N4 N4’ N6 N6’ N8 N8’ N10 N10’ A literature review found very little published re- search on the accuracy of replicas made of commercially available replica materials. A review of commercially available replica materials found that the proportion of replicas that can be measured with contact methods is significantly lower than those declared to be measurable only with non-contact methods. On the other hand, in real production conditions, contact profilometers are still the most common devices for measuring surface texture. It should not be neglected that different machining processes result in characteristic surface textures. The question can be raised to what extent an applied replica accurately reproduces a surface processed, for example, by turning or grinding. Although there are several parameters, Ra and Rz are still most often defined in technical product documenta- tion as the indicators of whether a surface texture is within certain tolerance limits. For all the above reasons, an analysis of the accuracy of the Technovit ® 3040 replica, which the manufacturer Kulzer Technik declares suitable for contact measure- ments, was conducted. The measurements were carried out using a Perthometer S8P calibrated contact profilometer. Roughness parameters Ra and Rz were measured on 15 samples processed with six conventional machining processes and with N2, N4, N6, N8 and N10 grade values. On every sample and its replica, five roughness profiles were taken, and arithmetical means of the Ra and Rz parameters were calculated. From arith- metical means, relative deviations of the Ra and Rz val- ues were calculated. Minor relative deviations of the Rz parameter values compared to relative deviations of the Ra parameter val- ues were determined. Apart from Rz obtained on the N2 surface of Group B (lapping) and N8 surface from Group A (side milling), relative deviations of Rz parame- ters were equal to or below 10 %. Only two surfaces from GroupA–N2surface (grinding) and N10 surface (side milling) – and one surface from GroupC–N4sur - face (superfinishing) – had Ra values above 30 %. A comparison of the ranges of Ra and Rz values ob- tained on five roughness profiles on each sample was made to determine whether the uniformity of the texture of a replica was preserved compared to the original sur- face. The obtained ranges for the original surfaces and their replicas are reasonably close. The results do not show any trend in increasing or decreasing the values achieved on a replica surface compared to the original surface. It is important to emphasise that relative deviations of the Ra and Rz parameters indicate that the accuracy of a replica is not dependent on the type of machining pro- cess, nor the level of surface finish defined by the N grade ranging from N2 to N10. Also, comparing the roughness profiles obtained on originals surfaces and their replicas showed a good reproduction of the charac- teristic geometries of roughness profiles obtained with different machining processes. 5 REFERENCES 1 F. Baruffi, P. Parenti, F. Cacciatore, M. Annoni, G. Tosello, On the Application of Replica Molding Technology for the Indirect Mea- surement of Surface and Geometry of Micromilled Compo- nents, Micromachines, 8 (2017), 195, doi:10.3390/mi8060195 2 P. Szawara, R. Ostrowski, The replication method to analyze rough- ness of shaft, tool, and grinding wheel microgeometry, Advances in Manufacturing Science and Technology, 44 (2020) 4, 129–133, doi:10.2478/amst-2019-0022 3 Y. Liu, C. Ling, A. Malcolm, Z. Dong, Accuracy of replication for non-destructive surface finish measurement, Singapore International NDT Conference & Exhibition, 3–4 November 2011, e-Journal of Nondestructive Testing, 17 (2012) 3, https://www.ndt.net/?id=11530 4 https://www.kulzer-technik.com/en_kt/kt/metallographie/produkt- bereiche/abdruckverfahren_1/technovit_3040.aspx, 26.10.2022 5 https://www.kulzer-technik.com/media/webmedia_local/kulzer_tech- nique/media_4/metallo/downloads_5/66094618_00TechnovitKom- biincl.4021.pdf, 26.10.2022 G. BAR[I] et al.: ACCURACY OF REPLICATION OF SURFACE TEXTURE WITH A CONTACT PROFILE METHOD 46 Materiali in tehnologije / Materials and technology 58 (2024) 1, 41–46