Jtam-A4.dvi JOURNAL OF THEORETICAL AND APPLIED MECHANICS 52, 2, pp. 345-358, Warsaw 2014 ANALYSIS OF A CONICAL SLEEVE WITH PIVOT JOINT LOADING OF AXIAL FORCE Andrzej Andrzejuk, Zbigniew Skup, Robert Zalewski Warsaw University of Technology, Institute of Machine Design Fundamentals, Warszawa, Poland e-mail: mang@ipbm.simr.pw.pl; zskup@ipbm.simr.pw.pl; robertzalewski@wp.pl The paper presents theoretical and experimental studies of energy dissipation in a model of a conical sleeve-pivot joint. Energy dissipation between cooperating surfaces of a friction pair including structural friction, elastic and frictional effects between its elements and Lame’s problem are taken into account. A comparative analysis was conducted to compare the theoretical results obtained from numerical simulations and direct experimental data acquired from the MTS testing machine. The analysis of the influence of geometrical and material parameters, external loading on the dissipation of energy is presented too. This paper shows an outline of theoretical considerations, themethod for conducting tests aswell as selected comparative results. Keywords: conical joint, hysteresis loop, experimental testing 1. Introduction Analytical considerations concern the natural energy dissipation problem observed on a tem- porary fastening in a conical sleeve-pivot joint subject to an axial load. In the literature, the authors have not found papers concerning the problem of natural damping of vibrations in a co- nical sleeve-pivot joints (including structural damping). The range of the present paper includes theoretical and experimental investigations of the previouslymentioned assemblies subjected to the axial loading. Sucha type of friction joints canbe commonly found in variousmechanical sys- tems used in daily engineering applications, e.g. machine tools, automobile or aircraft industry. In this paper, authors investigate a fasteningwithCoulomb’s law, elasticity of the joint elements, Saint-Venant’s principle and Lame’s problem taken into account. Beside theoretical tests, also amathematical model, based on the a real experimental research, is proposed and verified. The sleeve-pivot connection shown in Fig. 2a was subjected to an axial load αP (0 ¬ α ¬ α1). In this paper, an assumption has beenmade that α1 =1. Themain aim of the theoretical investi- gation is a detailed analysis of the previously mentioned conical joint taking into consideration structural friction, which appears in a temporary and permanent fastening everywhere, where a step out pressure acts onto cooperating surfaces. Energy dissipation phenomena caused by structural friction is observed in various types of wedge connections in both temporary and per- manent joints. This problem is widely discussed in the literature for more than fifty years. The pioneer work that considered static problems for structural friction with additional simplified assumptions was done by Kalinin et al. (1960). Thereafter, similar problems were investigated in both the domestic and foreign literature. It is worthmentioning that static and dynamic pro- blems observed in the structural friction fieldwere particularly considered inmonographs e.g. by Giergiel (1990), Osiński (1986, 1998), Skup (2010) and papers byGałkowski (1999), Grudziński andKostek (2005), Kaczmarek (2003), Sadowski and Stupkiewicz (2010) or Kosior (2005). The examples of references by other foreign authors includeworks of Ando et al. (1995), Feeny et al. (1998), Lin and Cai (1990), Person (1998), Popp (1998), Sextro (2002). The static analysis of different types of cooperating joints is generally limited to investigations of displacements in an 346 A. Andrzejuk et al. external loading function or determination of the static hysteresis loop (Fig. 1). The structural friction phenomenon is rather a complex problem mainly due to: sophisticated mathematical description of the structural friction phenomenon, loads and stresses distribution, nonlinearities of strains and functions describing the structural hysteresis loop, roughness of the contacting surfaces, etc. Fig. 1. Theoretical hysteresis loop of the sleeve-pivot joint 2. Determination of the axial displacement in function of an external load for loading stages of the sleeve-pivot joint A simplified model of a sleeve-pivot joint loaded by an axial force αP is shown in Fig. 2a. Displacements in the loaded sleeve-pivot joint have been examined in four different loading stages. In each stage, an amount of dissipated energywas investigated. The section of the tested object, havingheight ∆xhasbeendepicted inFig. 2.Theoretical considerations havebeen taken separately for the sleeve and pivot. In the next step, the authors established a mathematical formulation for the whole assembly. Fig. 2. Physical model: (a) sleeve-pivot joint, (b) section of the pivot, (c) section of the sleeve, (d) displacement scheme Stage 1 – Loading of the sleeve-pivot joint section (Fig. 2a) (0¬ αP ¬α1P) The external radius r0 of the joint section (Fig. 2b) can be defined as r0 = R−xtanβ (2.1) The force equilibrium equation in the section of the investigated sleeve-pivot joint is ∆σ1xF1 = ∆σ2xF2 → ∆σ2x = r20 r2z0−r 2 0 ∆σ1x (2.2) Analysis of a conical sleeve with pivot joint loading of axial force 347 where: F1,F2 are thefields of the cross-sectional area of thepivot and sleeve section, ∆σ1x,∆σ2x – increase of normal stresses in the considered sleeve and pivot. Considerations for the pivot section (Fig. 2b) The equation of the equilibrium from the projection of forces in the x-axial direction can be described as −σ1xπr20 −µpcosβ2πr0 ∆x cosβ −psinβ2πr0 ∆x cosβ +(σ1x+∆σ1x)πr 2 0 =0 (2.3) where µ is the friction coefficient. From transformed equation (2.3) and formula (2.2), the pressure per unit value p for the contact surface joint is given by p = ∆σ1x ∆x r0 2(tanβ +µ) (2.4) The equation of the equilibriumof forces operating in the pivot section in the radial direction (y-axis, Fig. 2b) is (−σ1r +µpsinβ −pcosβ)2πr0 ∆x cosβ =0 (2.5) Substituting formula (2.4) into equation (2.5), the radial stresses are σ1r = −r0cosβ(1−µtanβ) 2(µ+tanβ) ∆σ1x ∆x (2.6) Taking into account the symmetry of the system in the axial direction of x, y, and z (Fig. 2bc), and in accordancewithHooke’s law (Dyląg et al., 1996), the relative radial strain ε1r was determined σ1z = σ1y = σ1r ε1z = ε1y = ε1r ε1z = 1 E1 [σ1z −ν1(σ1x+σ1y)] (2.7) Taking into consideration dependences (2.7), equation (2.7)3 takes the form ε1r = σ1r(1−ν1) E1 − ν1σ1x E1 (2.8) where ν1, E1 are Poisson’s ratio and Young’s modulus of the pivot section. As a result of the radial deformation of the pivot section, its radius changes. Its absolute radial displacement can be described after substituting formula (2.6) into equation (2.8) by the formula ∆r01 = ε1rr0 = −r20 cosβ(1−µtanβ)(1−ν1) 2E1(µ+tanβ) ∆σ1x ∆x − ν1r0σ1x E1 (2.9) Considerations for the sleeve section (Fig. 2c) The equation of the equilibrium from the projection of forces in the x-axial direction takes the form −σ2xπ(r2z0−r 2 0)+µpcosβ2πr0 ∆x cosβ +psinβ2πr0 ∆x cosβ +(σ2x+∆σ2x)π(r 2 z0−r 2 0)= 0 (2.10) 348 A. Andrzejuk et al. By transforming, reducing and substituting dependence (2.2), a formula describing pressure per unit at the contact surface joint can be defined as in Eq. (2.4). The equation of the equilibrium from the projection of forces in the y-radial direction (Fig. 2b) takes the form (σ2r −µpsinβ +pcosβ)2πr0 ∆x cosβ =0 (2.11) hence, it appears σ2r =−pcosβ(1−µtanβ)= pa (2.12) The distribution of stresses σ2r and displacements up (Fig. 3) in the radial direction can be determined by using Lame’s problem formulas (Dyląg et al., 1997), thus σ2r = E2 1−ν22 ( C1(1+ν2)− C2 r2 (1−ν2) ) up = C1r+ C2 r (2.13) Fig. 3. Axially symmetrical stresses and displacements in the sleeve section The integration constants C1 and C2 occurring in (2.13)1 weredeterminedwith the following boundary conditions σ2r = { −pa = pcosβ(1−µtanβ) for r = r0 0 for r = rz0 (2.14) Therefore C1 = (1−ν2)r20 cosβp(1−µtanβ) E2(r 2 0 −r 2 z0) C2 = (1+ν2)r 2 0r 2 z0cosβp(1−µtanβ) E2(r 2 0 −r 2 z0) (2.15) Substituting dependences (2.15) in (2.13)1, a formula describing the radial stresses in the sleeve section was obtained in the following form σ2r = r20 cosβp(1−µtanβ) r20 −r 2 z0 ( 1− r2z0 r2 ) (2.16) Next, the radialdisplacement up (Fig. 2c)wasdetermined fromequation (2.13)2, additionally taking into consideration integration constants (2.15) up = r20 cosβp(1−µtanβ) E2(r 2 0 −r 2 z0) ( (1−ν2)r+(1+ν2) r2z0 r ) (2.17) Analysis of a conical sleeve with pivot joint loading of axial force 349 Finally, for r = r0 from (2.4), formula (2.17) after transformations can also be written in another form up ∣ ∣ ∣ r=r0 = −r40 cosβ(1−µtanβ) 2E2(r 2 0 −r 2 z0)(tanβ +µ) ( (1−ν2)+ r2z0 r20 (1+ν2) )∆σ1x ∆x (2.18) The displacement ∆ between the cooperating surface elements of the joint (Fig. 2), by making use of formulas (2.9) and (2.18), is given by ∆ = up ∣ ∣ ∣ r=r0 −∆r01 = ∆σ1x ∆x z1+σ1x ν1r0 E1 (2.19) where z1 = r20 cosβ 2 1−µtanβ tanβ +µ [ −r20 E2(r 2 0 −r 2 z0) ( 1−ν2+ r2z0 r20 (1+ν2) ) + 1−ν1 E1 ] (2.20) Amutual axial displacement of the pivot and sleeve joint sections (Fig. 2) for the first stage of the loading can be described as follows u = ∆ tanβ = ∆σ1x ∆x z1 tanβ +σ1x ν1r0 E1 tanβ (2.21) thus formula (2.21) throughmaking use of formula (2.20) takes the form u1 = η3(R−xtanβ)2σ′1x+η4(R−xtanβ)σ1x (2.22) where η1 = χ E2 ( 1−ν2+ r2z0 r2 (1+ν2) ) + 1−ν1 E1 η2 = cosβ(1−µtanβ) 2(tanβ +µ)tanβ η3 = η1η2 η4 = ν1 E1 tanβ χ = 1 (rz0 r )2 −1 (2.23) therefore, the axial strain displacement derivative (formula (2.22))with respect to thedistance x gives: ε1x = du1 dx = η3(R−xtanβ)2σ′′1x+(R−xtanβ)(η4−2η3 tanβ)σ ′ 1x−η4 tanβσ1x (2.24) The relationship between the stresses and axial strains according to Hooke’s law is ε1x = 1 E1 [σ1x−ν1(σ1z +σ1y)] (2.25) Taking into account the symmetry of the system, formula (2.7)1,2 takes the form ε1x = σ1x E1 −2σ1r ν1 E1 (2.26) After substituting equations (2.1) and (2.6) in (2.26) and transformisg, we get ε1x = σ1x E1 +η5(R−xtanβ)σ′1x η5 = ν1cosβ(1−µtanβ) E1(tanβ +µ) (2.27) Comparing formula (2.24) to (2.27), a homogeneous quadratic differential equation with variable coefficients has been obtained η3(R−xtanβ)2σ′′1x+η6(R−xtanβ)σ ′ 1x−η7σ1x =0 η6 = η4−2η3 tanβ −η5 η7 = η4 tanβ + 1 E1 (2.28) 350 A. Andrzejuk et al. Substituting σ1x =(R−xtanβ)λ into (2.28), the characteristic equation is given by λ2−B4λ−C12 =0 B4 =1+ η6 η3 tanβ C12 = η7 η3 tan2β (2.29) the characteristic equation discriminant is ∆41 = B 2 4 +4C12 > 0 λ9,10 = B4∓ √ ∆41 2 (2.30) therefore, the general solution to the differential equation may be written in the form σ1x = C13(R−xtanβ)λ9 +C14(R−xtanβ)λ10 (2.31) The integration constants C13 and C14 were designated for the next boundary conditions, thus σ1x =      0 for x =0 αP πr2 for x = H (2.32) therefore C13 =−C14Rλ10−λ9 C14 = αP πr2(rλ10 −Rλ10−λ9rλ9) r = R−H tanβ (2.33) Finally σ1x = C14[(R−xtanβ)λ10 − (R−xtanβ)λ9Rλ10−λ9] (2.34) The stress derivative is given by σ′1x = C14 tanβ[λ9R λ10−λ9(R−xtanβ)λ9−1−λ10(R−xtanβ)λ10−1] (2.35) Considering x = H, we get σ′1x(x = H)=C14 tanβ(λ9R λ10−λ9rλ9−1−λ10rλ10−1) (2.36) Substitutingobtained solutions (2.33), (2.32), (2.34) and (2.36)) in equation (2.22) depending on the operating external force αP, the formula of displacement of the extreme cross-section (x = H) takes the form u1(x = H)= αP πr ( η4+ η3 tanβ(λ9R λ10−λ9rλ9 −λ10rλ10) rλ10 −Rλ10−λ9rλ9 ) (2.37) Formula (2.37) describes the loading process of the investigated sleeve-pivot joint depicted in Fig. 1, interval 1. Stage 2 – Unloading of the sleeve-pivot joint section without sliding (α2P ¬ αP ¬ α1P) Unloading of the investigated system consists of decreasing the force value from α1P down to α2P. This stage is represented in Fig. 1 as a straight line A1A2 (interval 2). This process does not produce any changes in the value of displacement. Analysis of a conical sleeve with pivot joint loading of axial force 351 Stage 3 – Unloading of the sleeve-pivot joint section with sliding (α3P ¬ αP ¬ α2P) If we consider the friction forces per unit value equal to −µp acting on the considered slice (the stress is equal to σx), the sliding phenomenon of cooperating surfaces will occur. The loading state of the investigated joint at this stage is depicted in Fig. 4. Similarly to the first stage, formulas are obtained for the unloading process. Fig. 4. The unloading of the object (a) pivot sector, (b) sleeve sector Considerations for the pivot section (Fig. 4a) The equation of equilibrium of forces operating in the joint along the x-axis (Fig. 4a) takes the following form −σ1xπr20 −psinβ2πr0 ∆x cosβ +µpcosβ2πr0 ∆x cosβ +(σ1x+∆σ1x)πr 2 0 =0 (2.38) The equation of the equilibrium of forces in the radial direction on the y-axis (Fig. 4a) is given by (−σ1r2−µpsinβ −pcosβ)2πr0 ∆x cosβ =0 (2.39) Considerations for a sleeve section (Fig. 4b) Considering the equilibrium of force in the horizontal direction (Fig. 4b), we get −σ2xπ(r2z0−r 2 0)−µpcosβ2πr0 ∆x cosβ +psinβ2πr0 ∆x cosβ +(σ2x+∆σ2x)π(r 2 z0−r 2 0)= 0 (2.40) The equation of the equilibriumof forces in the radial direction on the y-axis (Fig. 4b) takes the following form (σ2r +µpsinβ +pcosβ)2πr0 ∆x cosβ =0 (2.41) Similarly to thefirst stage, the axial displacement in the third stage (u3(x = H))maybewritten in the form u3(x = H)= αP πr ( η4+ η9 tanβ(λ11R λ12−λ11rλ11 −λ12rλ12) rλ12 −Rλ12−λ11rλ11 ) (2.42) where η9 = η1η8 η8 = cosβ(1+µtanβ) 2tanβ(tanβ −µ) λ11,12 = B5∓ √ ∆43 2 ∆43 = B 2 5 +4C15 > 0 B5 =1+ η10 η9 tanβ η10 = η4−2η9 tanβ − c1 c1 = ν1cosβ(1+µtanβ) E1(tanβ −µ) C15 = η7 η9 tan 2β 352 A. Andrzejuk et al. Formula (2.42) describes the unloading process of the sleeve-pivot joint, depicted in Fig. 1 (interval 3). Stage 4 – Reloading process of the investigated joint (α3P ¬ αP ¬ α4P) At the initialmoment of the reloading process α3P to α4P, we donot observe themovement of cooperating elements in the investigated joint. The constant displacement is continued until the friction forces µp change the sign (Fig. 1, interval 4). 3. Determining of the energy dissipation for a single loading cycle of the investigated system During a single loading cycle, the amount of dissipated in the system energy can be expressed by ∫ αP(u) du and corresponds to the field of the hysteresis loop in the triangular form OA1A2 or the quadrangle form A1A2A3A4 (α3 > 0) (Fig. 1), depending on the loading method of the investigated system. The amount of energy dissipation for the single loading cycle (Fig. 1) can be given in the following form ψ = P 2 umax(α1−α2)− P 2 umin(α4−α3) (3.1) Assuming α = α1 and independent from α parameters η1-η9, r, R, λ9-λ12, β, H, formula (2.37) takes the form u1(x = H)= u1max = α1P(m17+m18) (3.2) and for α = α2, formula (2.42) is given by u3(x = H)= u3max = α2P(m19+m18) (3.3) where m17 = η3 tanβ(λ9R λ10−λ9rλ9 −λ10rλ10) πr(rλ10 −Rλ10−λ9rλ9) m18 = η4 πr m19 = η9 tanβ(λ11R λ12−λ11rλ11 −λ12rλ12) πr(rλ12 −Rλ12−λ11rλ11) (3.4) Comparing formulas (3.2) and (3.3), we get α2 = α1 m17+m18 m19+m18 (3.5) Assuming α = α4, formula (2.37) takes the form u1(x = H)= u1min = α4P(m17+m18) (3.6) For α = α3, formula (2.42) is given by u3(x = H)= u3min = α3P(m19+m18) (3.7) Similarly – for stage 4, when the friction forces change their direction, α4 value was determined by comparison of formulas (3.6) and (3.7), therefore α4 = α3 m19+m18 m17+m18 (3.8) Finally substituting expressions (3.2), (3.5), (3.7) and (3.8) in formula (3.1). we get ψ = P2 2 (m19−m17) ( α21 m17+m18 m19+m18 −α23 m19+m18 m17+m18 ) (3.9) Analysis of a conical sleeve with pivot joint loading of axial force 353 4. Results of numerical simulations Simulation tests were carried out in the Mathematica 6.1 environment. The numerical calcu- lations were conducted assuming the following values of parameters taken directly from the experimental setup: ν1 = ν2 = 0.29, E1 = E2 = 2.1 · 1011N/m2, µ = 0.15, β = 12◦, 14◦, 16◦, 18◦, R =38 ·10−3m, r =28.03 ·10−3m, H =40 ·10−3m. As a result of numerical calculations, hysteresis loops were obtained for both frictional and elastic models (Figs. 5 and 6) of the cooperating elements. Additionally, Lame’s problem has been taken into account. Fig. 5. Structural hysteresis loops for the frictional sleeve-pivotmodel for various loading values P : 1 – 25kN, 2 – 50kN, 3 – 75kN, 4 – 100kN and β =12◦, v =0.29, µ =0.15 Fig. 6. Hysteresis loops for different values of coning angles of the friction joint; v =0.29, µ =0.15 the data depicted in Figs. 5 and 6 enable comparison of the amount of energy dissipated in the investigated joint for various load values (Fig. 5) and cone angle values (Fig. 6). Analyzing the results, we can observe that higher values of the loading forces result in increasing of the investigated displacement and the amount of energy dissipated in the system (Fig. 5). The conducted numerical tests revealed that the increasing of the cone angle β results in linear decreasing of the structural hysteresis loop area (Fig. 6). Detailed analysis of the data depicted inFig. 7 enableddetermination of theoptimal valueof the friction coefficient (µ =0.13) assuming the highest amount of the dissipated energy criteria. Thedata depicted inFig. 8 reveals characteristics of the hysteresis loop area ψ in function of the loading force P.Numerically acquired characteristics ψ(P) have stronglynonlinear character for all considered cone angle values. Analyzing the data presented in Fig. 8, one can observe that the obtained characteristics ψ(P) are nonlinear for different values of coning angles. 354 A. Andrzejuk et al. Fig. 7. Energy losses in function of the friction coefficient for the unloading process of the tested model; P =100kN, β =12◦, v =0.29 Fig. 8. Energy losses in function of the loading for various values of the cone angle β of the investigated system; v =0.29, µ =0.15 As a conclusion, it is worth mentioning that in the design process of material systems and constructions including friction cone joints, it is possible to use the natural way of vibration damping by using structural friction occurring in the friction cone joint. The obtained results confirmed that by an appropriate selection of geometrical parameters of the tested joint (for example Fig. 6 or Fig. 8) and material properties (for example Fig. 7), the amount of natural energy dissipated in the investigated system can reach its highest possible value. The obtained results have confirmed that conducting such kind of experimental and numeri- cal tests can be very useful. The presented in the paper analytical considerations allow one – at the initial stage of design – to examine vibration damping strategies without involving the real model of the expensive cone joint element. Such an approach seems to be especially reasonable from the economical point of view. 5. The experimental model Themain goal of experimental testing is to choose the most suitable mathematical model that would constitute the best approximation of the real model response, i.e. minimizing the diffe- rences between the areas of experimental and numerical hysteresis loops. In order to perform experimental tests, a real model of the friction pair of the sleeve-pivot joint was designed and manufactured (Fig. 9). The experimentally investigated model was made of steel S2. In the te- sting specimen, a cone angle β =14◦ was applied. The overall construction model is shown in Fig. 9. It was designed in such a way that its position in the machine gripping jaws during the loading process would not change (Fig. 9 – elements 3 and 4). Tomeasure displacements of the Analysis of a conical sleeve with pivot joint loading of axial force 355 system, extensometers 2 and 3 (Fig. 10) were used with a measurement base of 10mm and a measurement nominal range ±1.2mm (sub-range of the nominal range ±0.24mm), which were included in the standard equipment of the universal strength testingmachineMTS 809 (Fig. 10 – elements 2, 3). Fig. 9. Design of the experimental model of a sleeve-pivot friction joint (decomposed system): 1 – lower pressure plate grip, 2 – upper grip, 3 – pivot, 4 – sleeve Fig. 10. Fastening of the tested model and an overview of its components: 1 – sleeve-pivot, 2 – extensometer type Ext 26, 3 – extensometer type Ext COD; 4, 5 – upper and lower gripping jaws of theMTS testing machine, 6 – steady pin, 7 – programmer Themethod of fastening extensometers 2 and 3 (Fig. 10) to tested elements 3 and 4 (Fig. 9) is illustrated in Fig. 10. In elements 3 and 4 (Fig. 9), a steady pin was mounted in order to ensure a coaxial measurement base for extensometers 2 and 3 (Fig. 10). In order to provide the best conditions for the cooperating joined elements (maximumcontact surface, surface pressure, smoothness of the motion), the conical surfaces were subject to a surface finishing process that consisted of very precise grinding. 6. The results of experimental testing The tests were conducted using the MTS testing machine at the Institute of Machine Design Fundamentals of Warsaw University of Technology. In the measurements, the Test Ware SX special software has been used. The fastening scheme of the investigated system is illustrated in Fig. 10. The single measurement methodology consisted of initial loading the system up to a maximal force value Pmax limited by safety loading threshold of the tested material (elastic range of deformations). Next, the system was unloaded down to the pre-assumed value Pmin and reloaded again. The loading process was fully controlled by the computer to avoid arising the torque while the experiment was conducted. Each experimental test was preceded by the 356 A. Andrzejuk et al. calibration processes of both the extensometer and the control-measurement system. All expe- rimental data, acquired from the laboratory tests was transformed into suitable characteristics plotted in the Mathematica 6.1 software. (Figs. 11 and 12). The same numerical environment was applied to approximate the obtained graphs and to calculate the area of experimental hy- steresis loopswith a numerical integrationmethod. Typical experimental results are depicted in Figs. 11 and 12. The systemwas loaded up to Pmax =25, 50, 75, 100kN, and then unloaded to Pmin =5kN. Fig. 11. Verification of the experimental and numerical data for P max =50kN, number of loading cycles: 4 Fig. 12. Experimental hysteresis loops for the model loaded by various forces P max , number of loading cycles: 4 In Fig. 12, a set of experimentally determined hysteresis loops has been depicted. The pre- viouslymentioned characteristics have been obtained for various loading forces Pmax from25kN to 100kN, and a constant unloading condition Pmin = const = 5kN. The loading program realized on the tensile strength machine consisted of linear increasing of the compression force value up to Pmax value, maintaining the maximal load for 10s, and unloading process down to the threshold value Pmin. Four independent loading cycles have been realized. It has to be mentioned, that the described loading process of the cone joint is very com- plex. It is quite difficult to separate the structural friction phenomenon from the experimental data mainly due to the existence of internal friction, which was neglected in the theoretical considerations. Table 1 shows the comparison of energy losses obtained as a result of numerical simulations and real experimental tests and the percentage difference between them. Analysis of a conical sleeve with pivot joint loading of axial force 357 Table 1. The percentage difference between average values of experimental and numerical hy- steresis loop areas No. Load Theoretical Experimental Difference [KN] model ψ [Nm] model ψ [Nm] [%] 1 25 207.192 166.64 19.57 2 50 827.716 767.95 7.22 3 75 1862.360 1739.08 6.62 4 100 3310.860 3113.04 5.97 7. Conclusions The paper presents amathematical model of the sleeve-pivot cone joint and results of numerical simulations and experimental tests conductedona realmodel.Acomparative analysis allows one to formulate a conclusion that the numerical response of the model reflects the real behaviour of the investigated cone joint. The characteristics depicted in Fig. 11 are comparable – both quantitatively andqualitatively. Thedivergencebetween the theoretical andexperimental results ismainly caused by the simplifying assumptions taken for themathematical model, i.e. constant friction coefficient, negligence of internal friction, precision of manufacturing of the real model, problems related to the fixing of the model in the testing machine (positioning of the model in gripping jaws of the testing machine) and problems encountered duringmeasurements. References 1. Ando Y., Ishikawa Y., Kitahara T., 1995, Friction characteristics and adhesion under low normal load, Journal of Tribology, 117, 569-574 2. DylągZ., JakubowiczA.,Orłoś Z., 1996, 1997,Strength ofMaterials (inPolish), vol. I and II, Publishers Technical of Science,Warsaw 3. Feeny B., Guran A., Hinrichs N., Popp K., 1998, A historical review on dry friction and stick-slip phenomena,Applied Mechanics Review, 51, 321-341 4. Gałkowski Z., 1999, Influence of structural friction of the vibrations sleeve-shaft (in Polish), Science of Books Rzeszów University of Technology, 174, 283-288 5. Giergiel J., 1990, Damping of Mechanical Vibrations (in Polish), Polish Publishers of Science, Warsaw 6. Grudziński K., Kostek R., 2005, Influence of normal micro-vibrations in contact on sliding motion of solid body, Journal of Theoretical and Applied Mechanics, 43, 37-49 7. KaczmarekW., 2003,Analysis of abolted jointwithelastic and frictional effects occuringbetween its elements,Machine Dynamics Problems, 27, 1, 21-40 8. Kalinin N., Lebiediew Ju., Panowko J.G., Lebiediev V.J., Strachow G.I., 1961, Kon- strukcyonnoe dempfirovanie v nepodvizhnykh soedinenyakh, AkademijaNaukŁatwijskojCCR,Riga 9. Kosior A., 2005, Influence of Parameters of Joints with Structural Friction on Elastic and Dam- ping Properties of Mechanical Systems, PublishingHouse of theWarsawUniversity of Technology, Mechanics, Exercise, book 209 10. Lin Y.K., Cai G.Q., 1990, Random Vibration of Hysteresis Systems, Nonlinear Dynamics in Engineering Systems, Springer Verlag, Berlin, Heidelberg, 189-196 11. Person B.N.J., 1998, Sliding Friction, Springer, Berlin 12. Popp K., 1998, A historical review on dry friction and stick-slip phenomena, Applied Mechanics Review, 51, 321-341 358 A. Andrzejuk et al. 13. Osiński Z., 1986,Damping ofMechanical Vibrations (inPolish),PolishPublishingHouse,Warsaw 14. Osiński Z., 1988,Damping of Vibrations, A.A. BALKEMA/Rotterdam/Brookfield 15. Sadowski P., Stupkiewicz S., 2010, Combined effect of friction and macroscopic deformation on asperity flattening,Tribology International, 43, 9, 1735-1741 16. Sextro W., 2002,Dynamical Contact Problems with Friction, Springer, Berlin 17. Skup Z., 2010,Nonlinear Phenomena in the Damping Vibration (in Polish), Publishing House of theWarsawUniversity of Technology, pp. 376 Manuscript received May 14, 2013; accepted for print October 14, 2013