Investigation of the &quot;Arena Zagreb&quot; multifunctional hall´s roof

Mladenko Rak

Investigation of the "Arena Zagreb" multifunctional hall´s roof

Mladenko Rak

2009

descriptionSee full PDF

Abstract

The structure consists of curved steel girders suspended over L=104 m span by means of cables system and of suspenders. The investigation of girders was conducted so that hangers were suspended from the main girder’ s medium nodes. The platforms for loads amounting from the total of 440 kN were fastened to the suspenders. The investigation was conducted throughout 5 phases. The vertical deflections, deformations (in each main cables), dynamic investigation were conducted. Results were compared for experimental and numerical investigation.

Figures (532)

The frequency of segment formation is usually higher than 10 kHz, which is above the level of conventional response of classical sensors. By releasing the shear energy of the lamella, a far more dominant peak source is observed in the signal of vibrations in case of production of saw-tooth chips. This is best illustrated by test data, Figure 2. 4. References Fig. 2: Parameters of median and frequency changes depending on the chip structure.

Metallography can be routinely used to study porosity in AlSi castings. Fig. 1: Casting pores in AISi7Mg by metallography

Fig. 2: Largest pore sizes in Al-Si alloy by metallography Measurements of largest pore equivalent sizes found in a_ significant number of metallographic fields of view can be discussed in the light of the Gumbel’s distribution, [2]. An example of largest pore size data sets (i.e. Maximum Feret Diameter) entered into the Gumbel plot is shown in Fig. 2. Since each data set appreciably fulfills the linearity condition, the Gumbel statistical distribution can be used for fitting and extrapolation. The slope of the regression line is an indicator of the scatter in the data. Inspection of Fig. 2 shows that set A (modified with Na and cast in steel mold) has the largest scatter among the three materials while set B (modified with Sr and cast in steel mold) and C (modified with Sr and sand cast) have similar and reduced scatter.

representing pixel by pixel the absorption coefficient of the material crossed by X-rays. From these scans, cross-section images (i.e. slices) of the material volume are obtained with a back-projection algorithm. The second step of the procedure is the 3D reconstruction of pores from the series of slices. An original software has been developed to obtain all the pores inside the volume under investigation (i.e. a cylinder 4 mm in diameter and 1 mm in length shown in Fig. 3).

the original pore. Therefore guidelines for estimating representative pore sizes based on coalescence conditions of neighboring pores are needed. Fig. 5: Cross-section of a shrinkage pore

The schematic draw of the testing equipment is shown in Fig. 1.

solution was stored in a furnace and after each and every test the physiologic solution was changed for a new 37°C solution.

The maximal force curve in the function of the travel distance is shown in Figure 2. Fig. 1: Maximal force — travel displacement for the tested sample The maximal force is given by the test results because it was recorded in the function of the travel distance and the cross section is detailed in Table 1. The trend of the curves is similar in all the four case, after loading they have a faster uploading. In two cases the first 5mm travel distance don’t need a considerable force. The prepared samples are not rigid samples and it happens that the length of the sample remains longer after fixation than the 5mm distance between the two grips.

Fig. 1: Tower with positions of measuring cross- sections, second level enumeration of strain gauges. Corresponding author: danko.radica@fesb.hr

Fig. 2: Characteristic time interval for wind 3ora on second sensor level

Fig. 4: RLCC of Bora (load case 1) rrom recorded data 1t 1S ODV1OUS that tower response on wind load is predominantly in first vibration mode what is agreement with [3]. and the respective frequency is 0.97 Hz.

Fig. 1.3 shows a more realistic design of the sleeve and fig. 1.4 represents a combination model 2 and model 3.

Fig. 1: Mesh model of the delaminated plate in the figure 2, the numerical and experi- mental results for the case of delamination placed in the symmetry axis is presented. In the figure 3, the test rig is presented.

Fig. 3: Variation Test rig Fig. 2: Variation of maximum transversal displacement versus inplane load. Case of the delamination placed between layers 8 and 9

‘ig. 1: Views of real specimen with installed FBG before and after the pressure test. Composite rotation-symmetrical elements (e.g. pressure vessels, pipes, shafts, mast, etc.) are most often fabricated from long fibers by winding technology. That method is completely automated and widespread [1-2]. The composite layers are packed in an ord programmed way in compliance with oadings. The braiding method for c ered and foreseen omposite elements manufacturing is new and a echnique in comparison with classica from 5° to 86°. Furthermore that makes possible embedding additional create so called “smart materia embedded other layers (i.e. with ternative winding method [2] and allows realization of reinforcement layer at an any angle in a range echnique elements between composite layers. It is possible to ” s” with specific magnetic properties) as well as sensors (ie. optical fiber based ones), which locally measure strain and temperature distribution composite material. Ss inside

Fig. 2: Exemplary force — strain and force — displacement curves for specimen with reinforcement angle 30° [5]. Fig. 2: Exemplary force — strain and force — displacement curves for specimen with reinforcement angle 30° [5]. During the test a compressive force (pressure), displacement of pistons as well as deformation (strain) in circumferential direction were measured. Because of large deformation on the outer surface of specimens, even few percents, Fiber Bragg Gratings (FBG) for strain measurements were used. It let to perform strain measurement very precisely in a wide range [3,4] which was not possible to obtain by classical method (electric resistance wire strain gauge).

4. References 3. Summary Fig. 3: Average values of circumferential strains for three different tube types: 30°, 45° and 60° in a function of applied load [5]. It was a result of a matrix failure in composite. So called delamination between two phases (glass fiber and epoxy resin) occurred. It was also o moment when a specimen started to whiten (see fig. 1). Maximal registered strain by FBG sensors reached 2.5% (25000 ts).

Fig. 4: Comparison of circumferential strain for tube specimen 30° [5].

The attacked locality becomes the anode and the surrounding surface becomes the cathode, which is considerably larger then the anode. Cations M"" hydrolyse and pH in the pit is rapidly decreasing (Fig. 1). The fall of pH inside the pit supports its farther growth. The last phase of the pit's development is its repassivation, which is achieved by the decrease of potential on the value of the

Fig. 2 S-N curves of AISI 316L and AISI 316Ti steels Fatigue investigations were carried out under fully reversed axial loading in the region from N = 10° to N = 10° cycles using high- frequency resonant ultrasonic machine. Fatigue tests were conducted under constant amplitude loading at a load ratio R = -1, frequency f ~ 20 kHz and ambient temperature, using smooth specimens with polished surface.

Fig. 3 Fatigue fractures and corrosion pits Surface crack initiation due to the formation of extrusion and intrusion during fatigue test was typical for the both steels (AISI 316L and AISI 316Ti) without corrosion damage. On the other hand, after corrosion damage, both steels were characteristic by fatigue crack initiation from the corrosion pits on the surface (Fig. 3). In this case the fatigue crack initiation was highly shortened due to the occurrence of corrosion pits and the total number of cycles was very close to the number of cycles of fatigue crack propagation.

Fig. 1: Crack in concrete, microscopic observation. Many contact methods and contactless methods are available, based on mechanical, optical and electric measurement for crack width changes in time: deformeters, mechanical indicators, enlarging magnifiers, microscopes, inductive potentiometer sensors, string tensometers, and optical sensors with optical fibres. 3. Crack width

Fig. 2: Crack width evaluation by proper software. 4. Conclusions

Fig. 1. Schedule of the Quasi-Steady Method. The quasi-steady method assumes that the characteristics of a propeller in each phase of unsteady motion are approximated by those in uniform flow at the same phase. Thus, if the non- uniform flow field is known around a propeller from experiments, the flow speed and the angle of attack at each phase are easily calculated according to the diagram represented in fig. 1.

Taking into account the overall dimensions of a concrete 55,000 tdw bulk carrier having a four- bladed propeller with a diameter D = 6.4 m, and using the flow-chart shown in fig. 3, one can obtain the corresponding results for the computed propeller torque and thrust amplitudes and phases.

Fig. 3. Flow-chart for the Calculation of the Propeller Harmonic Excitations. 4. References 3. Experimental investigations The prediction capacity of the theoretical models was check on physical model. Experiments have been carried-out at the Romanian Ship Research and Design Institute ICEPRONAV Galatz-Romania on dummy model (scale 1:30) in order to determine in the towing tank the nominal wake, the loading coefficients and the efficiency (fig. 4 and 5). The torque and thrust fluctuations were determinate in a cavitation tunnel with a propeller dynamometer in the measurement section 850x850 mm. The data from this device may be digitally read and automatically recorded by a

Fig. 4. Measurement Devices: 1-Dynamometer; 2-Am- plifier; 3-Printer; 4-Swich; 5-Osciloscop; 6-Dynamic Analyzer; 7-Recorder.

The torsion angle of the ship hull model cross section is determined by taking into account the displacements of the points placed on the outline (fig. 1), according to the rotation of the rigid body (thin-walled beam hypothesis).

7. References Fig. 3: Variation of the relative normal stress in the midship open section

Fig. 5: Variation of the relative torsion angle along the ship model The variation of the relative torsion angle (¢,/M,.) along the ship model obtained so from FEM analysis, thin-walled beam theory and from experimental measurements is presented in figure 5. As it is seen in figure 5, the maximum value of the parameter (¢,/M,) in the midship (open section type) is almost 2 times more than the maximum torsion angle in the closed area.

The force amplitude was decreased in consecutive tests from 11 KN to 5.3 KN as to obtain stress amplitudes from about 160 MPa to 75 MPa as shown in Tab. 1.

Tab. 1: Fatigue data obtained for ZK60 alloy.

Some SEM analyses revealed that this material contains mangan and_ titanium, although it should be 98 % magnesium and 2 % zirconium and silicium. As seen in Fig. 2 voids are inevitable. Fig. 2: SEM analyses of ZK60.

The corresponding S-N curve is represented in Fig. 1. Although not too many data are available up to now, we may suggest that a reasonable value for the fatigue limit could be 100 MPa.

Fig. 3: Histeresis loops obtained during a test at an amplitude of 150 MPa. However, another test at an amplitude of 75 MPa didn’t lead to the failure of the specimen. It is interesting to monitor the histeresis loops during each test. Just as an example, in Fig. 3 the first 100 cycles, the middle ones, and the last 100 cycles are shown.

Fig. 1: Specimens type, dimensions and injection gates; position of micro-CT examined samples. The effect of injection point location was analyzed both in light of fibre orientation and of the elastic properties computed by a micro- mechanical model based on the Cell Method, introduced in [4].

The experimental results of the fatigue tests are shown in fig. 2. Fig. 2: S-N curves for specimens with lateral notches.

Fig. 4: Preferred fibre orientation with respect to specimen y axis (above), and Index of Anisotropy IA (below), at different z positions (mm), core layer. Fig. 3: Lateral and longitudinal injection, point A, core layer: t-CT reconstruction.

Fig. 5: Apparent elastic modulus (MPa) along specimen x axis, at different z positions (mm), core layer.

Fig. 1: Indentation curves of a-B,C thin film measured at room and elevated temperature.

Fig. 2: Indentation hardness and reduced modulus of a-B,C film versus temperature.

There are often applied machines in the min- ing of the lignite coal about the long operating period frequently reaching 50 years [2]. There- fore the analyses of the state of these machines are being conducted in order to determine the alternative of their exploitation or replacing it by new. One of examinations conducted at this occasion there are measurements of oscillation and dynamic influences during mining [1]. These examinations are aimed at determining real dynamic loading and the scope of changes of stresses which can be a base for determining the fatigue life of elements of the load-carrying system [3]. Linking experimental methods with numerical methods it is possible on the basis of measurements in selected points to appoint the state of stresses and deformations in the entire load-carrying system. Fig. 1: View of RS-560 chain excavator

Because of great loadings of the central part of the excavator sensors were arranged, as it is showed on fig. 2. The measurements were con- ducted during the mining in different locations of the bucket chain boom. The instance course of changes of accelerations registered in the middle part in measuring point 6 was shown on fig. 3. On the basis of these courses a amplitude spectrum of accelerations was determined in the form like on fig. 4.

Fig. 4: Widmo amplitudowe przyspieszen. Fig. 3: Course of acceleration changes in point 6.

The influence of measured dynamic loads on the load-carrying structure of the excavator was determined with the help of computer simula- tions. A FEM shell-beam model was built [4, 5] in I-DEAS NX software shown on fig. 5. Fig. 5: FEM model of load-carrying structure of excavator

Fig. 6: Huber-Misses stresses distribution comming from dynamic loadings

apex), and in two deck structure cross sections, in total in 16 measuring points (Fig. 2).

Fig. 8: PSD record (1* and 3" natural frequency) Table 1 shows the comparison of the first 10 experimentally and numerically determined natural frequencies. Figure 8 shows the record

During the load testing displacements were measured in two lines, above all the supports and in the centres of deck structure span. Figure 7 graphically presents experimentally and numerically determined displacements for the 7” loading phase in which maximal displacements were measured.

Fig 7. Comparison of experimental and numerical displacements for the 7" loading phase 4. Conclusions

A wide range of systems has been developed to control the position of running metal bands. This paper deals with the mostly used guiding system, which controls the position of the metal band changing the skewing angle B of a tail pulley. system is shown in figure 1. by This

Figure 2: Mechanical model A of the metal band II) Bending theory of second order with consid- eration of the shear deformation - friction contact between band and pulleys

Figure 5: Representative measurement results The results were verified by measurements as well. Two strain gauges were applied on the metal band with a dis The results had and the glue on he band had to in figure 5 for ance of 11.5 mm to the band edge. o be multiplied by an increase factor o get the effective stresses in the band edge. Because of the small thickness of the band (0.3 mm) he influence of the thickness of the strain gauges the effective strain on the surface of be considered. The results are shown a skewing angle of 0.211° and a prestressing of 100 N/mm. Summary of the results

Figure 4: Representative results of FEM The finite elements method was used to verify the calculated results. The stresses for the band between the pulleys are illustrated in figure 4 for a skewing angle of 0.211° and a prestressing of 100 N/mm?. Since a fixed contact was chosen between the band and the pulleys the bending stress, which is caused by the curvature of the band, is not included in the results. The effective stresses for the points B1 and B2 follows as:

Figure 7: Results for point B1 and B2 contact with the tail pulley (point Al and A2). In figure 7 the results are stated for the points B1 and B2. where the band is in contact with the pulley. Two conditions have to be fulfilled to prevent the band from damage and to ensure a safe band run:

Fig. 1: Isometric view of testing facility. It is equipped by twenty-one hydraulic actuators. A real time control system guarantee synchronous application of each kind of load history on bogie frame. Here, a set of ad hoc fittings and twenty-one actuators guarantee load application, on each direction, on wheels, with the superposition of twist [2]. The same apply

Tab. 1: Typical calibration and uncertainty parameters. Prior to effectuate the required tests on the bogie, a method to evaluate calibration curves of measurement chain and uncertainty has been established. A typical flow chart is shown in tab. 1, where statistical data are calculated from strain gauges data and used to evaluate some commonly accepted parameters. Typically, more than seventy strain gauges are applied on a bogie frame.

Fig. 1: The RSPS test- geometrical model with boundary conditions All the operator’s protective structures must meet the requirements of standard PN-92/G- 59001 “Self- propelled mining machines — Rock Slide protective structures for operator — Requirements and Tests”, that describes the required tes s as a dynamic loading of a protective structure by the impact on the upper part of the structure of the research bob that falls on the s tructure with the potential energy E regard to impact point (presented on figure 1).

Fig. 2: The RSPS test- « displacement contours for the maximal deflection. The results of the analyses in the form of displacement contours are shown in Figure 2.

The measurement results show the good conformity with the results from numerical simulations. The permanent plastic displacement and concerned with _ that deflection in the impact area amount to 42mm. In the numerical analysis the permanent deflection amounts to 17mm. Fig. 5: The graph of vertical displacements for chosen points of the operator cabin

Fig. 4: Experiment - view of the cabin with the loading bob In the range of experimental test (verification test) the destructive strength test was conducted on the prototype of a protective structure. The impact energy of the research bob was 60000J according to the RSPS standard. The view of the test is presented on figure 3 together with the graph of the defection of the selected point measured by the Polytec Laser vibrometer PS V-400-B.

Fig. 6: . Displacement in the selected point — laser vibrometer measurement

Fig. 1: Two modes of deformation due to fatigue damage: a) cyclic plasticity and b) ratcheting. Two modes of deformation mechanism due o deve he firs cycling plastici band parame opment of a fatigue damage process can be distinguished depending on a kind of metal and its state defined by manufacture process. In one deformations which occur due to loading are governed by cyclic y tule as a result of micro slip and slip decohesion. The fatigue indicator er can be defined in the case based on changes of the strain amplitude for a constant stress amplitude of a cyclic loading as shown on Fig. la.

The high cycle fatigue (HCF) tests were performed using hourglass specimens shown on Fig.2. Specimens were cut out from a railways wheel made of the P54T alloyed steel with chemical composition as given in Tab.1. Specimens were machining in such a way that all machine traces were along the specimen axis. Fig. 2: Specimen on the left side and special griping system used in the HCF tests.

An exemplary and typical for the tested material behavior is shown on Fig. 3 for specimen under stress amplitude equal to 400 MPa. Fig. 3: The stress - strain curves of a selected cycles for specimen P1 tested under stress amplitude equal to 400 MPa. A stress — strain curve in a sequential cycles has the same elastic form and shifts in the direction of tension for increasing number of cycles. Development of damage due to cycle loading leads to ratcheting and in consequence to fracture of specimen. The tested material exhibit increase value of a mean strain ¢,, with growing number of cycles in loading under constant stress amplitude and zero value of a mean stress. The inelastic strain ¢;, equal to a width of the hysteresis loop has a constant value up to the last period of the lifetime when increases rapidly just before fracture of specimen. The ratcheting in a form of a growing value of a mean strain is a main mechanism of damage development in high cycle fatigue for the tested P54T steel. And the mean value of strain in one cycle can be consider as a fatigue indicator parameter and The additional information on _ strain response of material in sequential cycles of a constant stress amplitude enable us to define: a type of a damage mechanism, a fatigue indicator parameter, a development of damage, and the most important, a range of the applicability of the linear accumulation fatigue rule if exist at all. The information enable us to distinguish subsets of experimental points of the S/N curve which corresponds to different distribution of initial inclusions in the measuring part of specimen.

Fig. 4: The axial strain as a function of the normed fatigue on the left and evolution of the damage parameter, on the right. can be used to define a damage parameter D as shown on fig. 4. 3. Conclusions

The evaluation starts with a_ visual investigation. The investigations are shown in fig.1 Surface cracks, and degradation are important indicators of beam condition. Fig. 1: Crack depth determination by thin plate penetration

Fig. 2: Crack depth determination by acoustics (stress wave sensors)

Fig. 3: Velocity determination in a lamella

3. Results In straight and curved = glued laminated timbers (glulam) residual stresses can be significant due to change of climatic conditions. The stress caused by the manufacturing process, the load and the climatic conditions add together. The resulting stress may cause failure. We can find them in practice when the lamellas break or crack parallel to the fibre direction. Our research team uses a stress evaluation method developed at the University of West Hungary [1]. If we know the geometry of the glulam structure, the properties of the lamellas -the thickness, Young’s modulus, the climatic and the manufacturing residual stress can be determined. We also need the initial and final temperature and _ the moisture content values of the lamellas to calculate the climate residual stresses.

Fig. 7: Residual stresses by the manufacturing process

Fig. 8: Residual stresses due to change of climate conditions 4. Conclusion

Fig. 1: Image correlation principle [2] The Q400 measuring system is very easy to handle. Before any measurement the system calibration is necessary to be performed. The Q-400 system has, for a _ successful measurement, a _ calibration procedure incorporated in the measurement and analysis software ISTRA-4D. By the calibration procedure are determined the intrinsic and extrinsic parameters of the cameras. A test plate with a chess model on it is moved in front of the cameras.

Fig. 2: Experimental set-up The tensile test was conducted according to the EN 10002-1:1990 European Standard for tensile testing of metallic materials at ambient temperature [4]. Five specimens having the same dimensions were tested in the same conditions to check if the measuring results can be reproduced. Instead of using a classical extensometer during the tensile test many images of the specimen were acquired by both cameras of the optical measurement system. The values of the corresponding force at the acquisition time were recorded.

Fig. 3: Engineering stress-strain curve of aluminium Thus for tested commercial aluminium the material constants were Young modulus about 58 N/mm”, Poisson ratio 0,35. The yield stress was about 92 N/mm’ and the ultimate stress about 98 N/mm”.

Fig.1: Cross section, load The investigation of girders was conducted so that hangers (Fig. 1&2) were suspended from the main girder’s medium nodes. The platforms for loads amounting from the total of 440 kN (40 KN per node) were fastened to the suspenders.

The investigation was conducted throughout 5 phases (5 girders in a set), and unloading was performed after the completion of each phase. The vertical deflections in each loading phase were measured at 5 characteristic points (Fig. 2) along each girder. Fig.2: The plan of roof, loads, deflections

5. References The diagram (Fig. 6) shows recordings of relative deformations of cables and suspenders. Figure 7 presents deformations of main steel girder no.3. Fig.6: Records of relative deformations of cables and suspenders

Figure 5 graphically presents experimental and numerical deflections determined in the 3™ phase at one of the characteristic girders no.3. Fig.5: Comparison of experimental and numerical deflections of the girders no.3

certain time series signals were registered along with the functions of power spectral density (PSD) determining relevant dynamic parameters. At the same time, during static investigation, the monitoring installed on two characteristic main roof structure girders was activated (Fig. 4). Monitoring includes continuous measuring of the structure's deformations. Fig.4: Points for measuring relative deformations during monitoring

Fig.7: Recordings of main girder’s relative deformations (no.3) during monitoring

Fig.1: A look at the pedestrian bridge A suspension bridge with 209,5m span connects left and right banks of the Drava river (Fig. 1). The main bearing system consists of cables anchored on either bank over 30,0m high pylons. The deck is suspended by the suspenders from the cables with 4,0m distances. The cables are connected over suspenders and concrete slabs with tendons running beyond the deck and anchored in the pylon blocks. The usable width is 5,0m, and the overall width 6,0m. 2. Investigation and measurement

Figures 2 and 3 show the comparison of the forces results in cables in downstream and upstream direction. ordinal number of suspenders

[ab.2: Comparison of experimental and natura frequencies for the years 2009 and 1993 4. Conclusions

Fig. 3: Forces comparison in suspenders (downstream in 2009, downstream in 1993) Forces in suspenders fell on average by 8% compared to the situation in 1993. With regard to structural analysis the measured forces in suspenders are on average higher by 6%. Largest deviations of measured forces with regard to forces in suspenders from structural analysis are in upstream suspenders no. 59-62. The connection element to the main cables of upstream suspenders 60 and 61 is damaged. The lowest force was measured in the above mentioned suspenders (11,09 and 16,91 kN), and the highest force was measured in adjacent suspenders no. 59 and 62 (69,94 and 67,35 kN). Teaser A weAdante tha wearnedinae At RAGA omarion!

Tab.1: Comparison of experimental and numerical forces in main cables Forces results in cables and their comparison with numerical values from the structural analysis are shown in Table 1. Table 2 shows experimental results of natural frequencies for dynamic investigation of the bridge’s structure.

SPT tests were performed in special fixture developed for these tests for testing in servo- hydraulic testing machine. Sample deformation was measured by extensometer attached to the testing fixture. All tests were performed at ambient temperature. Samples dimensions are following: diameter 8mm and thickness 0,5mm. Tests records are shown in Fig. 1. Fig. 1: Comparison of SPT records with FEM based results.

3. Small Punch Test Corresponding author: jan.dzugan@comtesfht.cz

Tab. 2: Comparison of tensile properties obtained by SPT and tensile test

The lowest compression strength values were calculated for the regenerates produced in the sheep in (reference) group R. In comparison with the reference group, the strength values determined in the sheep belonging to group A (dFM=Imm) are higher by 23% (regions: CB, PB). Analogical in group B (IFM=2mm) the difference, in comparison with the reference group (R), is even larger and amounts to 40%.

The engine via gear unit rotates with the pressing disc 1, to which three pressing cones 2 are pressed with pre-stress. Pre-stress is induced by fastening of two nuts on the supporting axis 3, under which a transducer 4 was placed during measurement. Three bending moments

Fig. 2: Course of axial force at manual fastening.

Fig. 3: Course of bending moments during fastening.

Fig. 5: Course of bending moments at measured places 5. The Figure 4 shows the courses of the axial forces measured by the transducer 4 at normal full output of the machine and at its overload, when the disc holes get clogged. It is obvious that the resulting force at overload is approximately double. The course of the bending moments on the supporting axis at the measured place 5 are shown in the Figure 5. Three strain-gauge half- bridges are along the perimeter of the supporting axis glued in the planes, mutually

The granulator is during exploitation sometimes overloaded, when the holes in the pressing disc get clogged. Some courses of the measured values will be shown — at normal full output of the granulator and at its overload.

Fig. 6: Course of bending moment at measured place 6. At the measured place 6 a strain-gauge half- bridge was also glued, enabling evaluation of the courses of bending moments on the axis of one of the pressing cones 2. The course of the evaluated bending moment at the granulator overload is plotted in the Figure 6. Their values also fluctuate heavily.

Fig. 1: The effect of the temperature and electric field on the specific viscosity of ER fluid calculated. Fig. 1 shows the effect of the electric field on the specific viscosity of ER fluid at 20 and 50 °C. The concentration of the ER fluid was 40 w%. The electrode distance was 3 mm. By increasing the electric field intensity, the specific viscosity of the ER fluid increases significantly as seen in the Fig. 1. At 20 °C the flow of the ER fluid was stopped at 0.833 kV/mm field intensity. At 50 °C the field intensity that was needed to stop he flow was 0.7 kV/mm. As increasing the emperature causes the viscosity of the ER fl fluid to decrease, this could be the explanation of the smaller field intensity that is needed to stop the flow.

Fig. 2: The effect of the electric field on the specific viscosity of ER fluid at different electrode distances. The temperature and the electrode distances are

Polymers usually can be characterized using more relaxation constants. The number of the elements to be considered is optional. In this paper a three element Maxwell model is used. The first block J; models the inertia of the crankshaft, the second block J; models the inertia of the housing, and the J; is the inertia of the inertia ring. The stiffness and the damping parameters of the damper oil were determined by measurement. In Fig. 3, the mechanical model of a single Maxwell element can be seen. The two coordinates that govern the motion are x; and x». The equation governing the motion of the dashpot is

Figure 1: Reduction of all microstructure and S/N-curve parameters to 3 dimensions — Factor Analysis

Figure 2: Response surface of the calculation error as a function of the loadings F, and F,

Another influence that has to be considered s the operation temperature. Figure 3 shows the Jependency of the relative fatigue stress on this yperation temperature. Figure 3: Relative endurance stress as a function of the operation temperature

Table 1: Comparison of distinct parameters to the experimental fatigue stress

‘igure 4: Relative estimated stress versus relative experimental stress

Fig. 1: The fatigue test bar Beside the reference S-N curve, taken for air, three more S-N curves were obtained. They are referred to (1) acidic solution: 50g NaCl + 5g CH;COOH + 945g H,O with pH = 2.7 — 3.0; (11) alkaline solution: 1N water solution Na,CO; and NaHCO, with the ratio 1:1 according to AGA [2] and pH = 9.3; and (iii) water from water main with pH = 6.5. For fatigue tests flat bars (Fig.1) were used.

The S-N curve for water exhibits a small slope. According to expectation this curve lies above the S-N curves for both solutions but below that for air. While for high cyclic loads he fatigue life for water is much shorter than for air, at low cyclic loads their difference decreases and for the life around 10° cycles the ives are very close to each other. At the life 2 x 10° cycles the corresponding stress (fatigue imit) for water is paradoxically higher than for air. The cause of this behaviour lies in specific properties of water; the course of corrosive actions is predominantly given by admixtures which water always contains.

Fig. 3: Variations of the fatigue strength reduction factors on the life

Fig. 1. The force-displacement curve during the loading of the specimen Corresponding author: bgalfi@yahoo.com, janoska@clicknet.ro

Fig. 3. The force-displacement curve during the loading and unloading of the specimen unloading phases. In figure 3 is shown a smoothed version of them. On can observe that during the unloading of the specimen, the effect/the influence of the pre-compressed cells, which have residual deformations, make that the unloading curve is located over the loading one (unusual thing at metallic specimens).

Fig.1: Testing rig of crankshafts at fatigue rotating bending Bending fatigue testing of crankshafts was performed in a fatigue testing rig (Fig. 1).

Tab 2. Failure probability for nonitrided crankshafts for Sy=250 N/mm?

Fig. 2: S-N curve with overall scatter region for nonitrided crankshafts

Fig. 3: S-N curve with overall scatter region for nitrided crankshafts

For every damage level, RVAC is calculated from equation (1). All numerical results are obtained from routines developed in the Matlab package, and shown on Fig.1. It can be seen that RVAC gives information about the relative extend of the damage. Fig. 1: RVAC value for four natural frequencies of beam and five damage levels

is repeated for the four level of damage: d=1 is undamage beam, d=2 coresponds to crack of 1 mm depth, d=3 is crack of 2 mm depth, d=4 is crack of 4 mm depth. The first four natural frequencies for different damage levels are given in Tab.2. Fig.2. shows overlaid FRFs measured at node 8 for four damage levels, and Fig.3. shows RVACs_ for experimentaly obtained data.

Tab. 2: Natural frequencies for different damage levels, experimentally obtained

Fig. 3: RVAC value for four natural frequencies of beam and four damage levels Fig. 2: Overlaid FRFs measured at node 8 for four damage level

Fig. 1 - SEM micrograph showing Mat, Roving and Volumat layers.

Fig. 2. Reconstruction of a 2D cross-section of a cracked specimen.

Fig. 1: Test set-up for surface DMA tests. The experimental set-up of the ring-on-disc (RoD) test set-up on the dynamic testing machine is shown in fig. 1. The same specimen geometry and the surface quality was used as previously in classical friction and wear tests. During these movement of the torsiona normal load and rotationa and a represen plotted in fig. 3. not provide su However, fficient in stick-slip behaviour of the Hence, a nove method implemented on an elec machine. The rotational drive has either a monotonic or an oscillating movement during ests a continuous rotational drive with varying speed was realized ative result of these tests is this type of test does formation about the elastomer specimen. was developed and trodynamical testing

Fig. 3: Results of the friction tests using RoD test set-up. Fig. 2: Test fixture along with the specimen and counterpart for surface DMA tests.

Fig. 4: Monotonic torque-rotation curves at various normal loads and for various loading rates. Torque-rotation (M-@) curves at various normal loads at low loading rate under monotonic loading are shown in fig. 4a for an unfilled TPU specimen.

Fig. 5: Cyclic torque-rotation hysteretic curves at a normal load of 1.5 MPa.

The horizontal measures of the simulation model were 50m x 16m. An additional requirement was that a one meter difference in height can vertically be measured at every point of the co- ordinate system. Fig. 1: Total view of the model of the Danube to the east of Vienna

Several triangulating laser measuring probes (triangulation methods) by Festo are used for length measuring systems [1]. Fig. 2: Measuring bridge with the measuring olatform and five mounted laser triangulation sensors

Fig. 1: Telescopic cover - Direction of wiper friction and resistance forces (F, and F,) Machine tool manufacturers begin to focus on optimizing peripherals such as telescopic covers [1]. Passive resistance of covers and its sealing wipers in particular becomes more significant factor of drive power dimensioning. 2. Sealing wiper

Fig. 2: Comparison of experimental and modelled stiffness characteristics Physical behaviour of sealing wiper ha been examined by a mathematical model and ; has been verified by an experiment [2]. It stiffness characteristic is bilinear respecting tw contact zones of the wiper (compliant wiper li and stiffer sealing plane).

Fig. 6: Normal force curve along wiper length; all wipers in contact with segments This model also shows whether the wiper is in contact with the segment below to assure sealing functionality. If the normal force of any segment reaches zero, it means that there is no preload of wiper. Zero preload represents gap between wiper and the segment. This can happen in case of large cover and it must be prevented by cover modification [3].

Fig. 7: Frictional force during cover extension Figure 7 shows 15 % difference of cover sealant frictional force during cover extension. In case of two covers usually mounted onto movable axis of a machine tool, this difference is lower (5 %). This difference is caused by change of relative distance of wipers that provide support of compliant segments.

Fig. 1 Cross section of bonding part Table 1 Content of conductive filler printed wire boards with the conductive adhesive resin. The material is FR-4. The copper foil is coated on the board in the diagonal direction of the board. The change of electrical conductivity during loading tests is measured through this copper foil. The conductive adhesive resin is put on the center point circle with diameter 0.8mm of the copper foil. Then the two boards are adhered by overlapping each other. The specimen is cured for one hour at 130 centigrade under pressure 5MPa. The rod with the diameter of 0.28mm is put between two boards to keep the constant distance of the two boards. Figure 1 shows the cross section of the adhesively bonded part.

Fig. 3 Combined strength for mixed filler content Fig. 2 Tensile strength for different filler contents

Fig. 1: Positioning system of portal type

Fig. 5 shows the proposed screen layout or the user interface. When the scanner is started this mode is the first to be displayed. It provides a button menu for selection of all menus, buttons for selecting manual or automatic referencing (in sequence Z,X,Y,B), start and stop buttons for referencing, status visualization of the servo-system and the laser head, visualization of the current coordinates and buttons for manual positioning along all axes (the latter are needed in case the system is in a location unsuitable for starting the Referencing mode).

The 50 MPa equipment designed to stretch aluminum plates is made of two jaws in which the ends of the plates are fixed. Each jaw has eight lamellae, drilled and cross-cut at the lower side. The aluminum plate is fixed in the lamellae through a system of clamping dies made of two flattening beams and two wedges (Fig. 1). Fig. 1: The fixing system of the aluminum plate For a correct evaluation of the performances and the technological parameters of the equipment in service, an experimental study was undertaken for the model of a lamella, using the photoelastic technique. The obtained results were compared with numerical results, yielding from Finite Element analyses.

Fig. 2: The considered positions of the wedges thickness of 9.5 mm. The aluminum plate was considered wedged using the wedges and flattening beams. That is why the whole assembly plate — wedges was modeled as a single wedge made also of a 9.5 mm thick Plexiglas plate. The photoelastic measurements were undertaken for three different positions of the wedges, found often in the real process of stretching (Fig. 5).

The isoclines and the isochromatic fringes (Fig. 3) were recorded corresponding to the three considered positions of the wedges. Fig. 3: The isochromatic fringe field (Case 1)

With the data obtained from the above calculations, the variation curves of the principal stresses on the contour of the hole were plotted (Fig. 4).

Fig. 5: Comparative plot of the principal stresses (Case 1).

Wood exhibits a hierarchical structure and is composed of fibers which can be represented as hollow tubes oriented in stem direction (see Fig. 1(a)). The fibers are built up of a cell wall material consisting of stiff cellulose microfibrils which are embedded in a soft polymer matrix. The S2 layer of the cell wall occupies approximately 80 % of the cell wall and mainly contributes to the load bearing capacity of wood. Within the S2 layer, the stiff microfibrils are inclined to the longitudinal cell axis by the so. called mircofibril angle, giving an approximately transversely isotropic material. Fig. 1: (a) Wood fiber consisting of different cell wall layers and (b) conical indentation in S2 layer of wood cell wall: orientation of the material axis and indentation direction. The mechanical properties of the S2 layer can be measured by tension tests on single wood fibers or by means of nanoindentation. As reported by several authors (see, e.g., [1]), the measured indentation modulus in longitudinal direction is considerably smaller than the expected Young’s modulus in the same direction. This can be explained by the small

Combining the micromechanical model with the model for the indentation modulus allows to study the influence of MFA and moisture content on the indentation modulus of wood cell walls. Figure 2(b) shows the decrease of the predicted indentation modulus with increasing MFA for a typical chemical composition of spruce and varying moisture content. Fig. 3: Predicted variation of indentation modulus as a function of microfibril angle and moisture content.

Fig. 2: Comparison of predicted and measured [6] indentation modulus.

Flexible polyurethane foams are commonly used materials in automotive applications, especially for internal cockpit parts and seats. During accidental impact passenger’s head has to be stopped by the headrest (Fig.1). As it is showed in Fig.2, padding material of a headrest has a_ significant influence on a_ head deceleration [3]. Mechanical properties of two material systems (polyurethane foam) with a certain range of densities and components proportions have been studied. Dynamic impact tests have been carried out. Results are presented below. Analytical constitutive model of compression has been proposed, which assumes that density, components proportion and strain rate are separable functions [1, 4]. The model has been verified by experimental results of static tests [4]. Impact tests have been conducted due to two objectives: to investigate impact behaviour of examined material and to validate numerical solution based on the method of lump mass [2].

Fig. 2: Head deceleration for three different foam materials of a headrest [3]

Impact tests were carried out on the specia experimental rig shown in Fig.7. Impactor falls from a certain height and hits a specimen. Velocity of impactor is measured by optica device just before contact with a specimen. Deceleration of the impactor during foam deformation is measured by accelerometer and

Fig. 4: Sample no.1 - Various impact velocities. recorded by a computer. Results of impact tests are presented at Fig. 4,5,6.

fig. 5: Samples no.1 — 4 (various components proportions) — Impact velocity ~1,2 m/s. Fig. 6: Samples no. 1 — 4 (various components proportions) — Impact velocity ~3,7 m/s 4. References

Fig. 7: Impact test device. 1 - foam specimen, 2 - impactor, 3 - impactor guides, 4 - speed measure device, 5 - accelerometer.

Fig. 1. Selected measuring locations: measuring spot I Lamellae of bucket wheel boom; measuring spot 2 — Lower carrier of counter weight structure; measuring spot 3 — Upper carrier of counter weight structure

Force accession measurement was performed at described points at extremely high ambient temperature, in August 2007. Placement of strain gauge at lamellae was done when excavator was under load. First measurement, performed during excavation, provided two signals related to technological operation of excavator. Force accession (stress) diagram in time domain, shown at Fig. 2, for all three measuring spots, represents first signal of stress during block excavation and super- structure slewing to the left. Fig. 2. Stress diagram during excavation, in time domain — first signal (excavation, slewing to the left) Highest stress accession, of 24.6 N/mm’, was recorded at lower carrier of counter weight. Stress accession at boom's lamellae is max. 17.8 N/mm”, while at the stress accession at upper carrier of the counter weight was max. 12.5 N/mm’. Stress at lamellae is constant throughout the excavation, while the stress at the counter weight carriers varies — it is larger within the lower carrier at the beginning of the process and as the process progress it builds up

Force accession (stress) diagram in time domain is shown at Fig. 3, for second signal of stress during block excavation and super- structure slewing to the right. Fig. 3. Stress diagram during excavation, in time domain — second signal (excavation, slewing to the right)

Force (stress) accession, for all three measuring spots during transportation of the excavator, but not in excavation, is shown at Fig. 4.

transmitting (T) and receiving (R) transducers along with stepper motors (M) and _ their controllers. The program (in LabView) allows control of kinematics of the reflectometer and automatic data acquisition. The program for signal analysis (in Matlab) performs averaging, spectral analysis, calculation of reflection coefficient and its diffraction correction. The diffraction correction is based on angular spectrum method. The reference signal for calculation of reflection coefficient is taken for face to face position of transducers. The experimental system works both in air (using Ultran transducers) and in water (with Panametrics immersion transducers).

Fig. 2. Reflection coefficient for (a) water/water saturated porous sintered glass, 0.5 MHz, (b) air/air saturated porous glass system, 0.2 MHz. Results of two tests are shown, glass with 260 um average diameter of grains.

Fig. 3. Calculated reflection coefficients for (a) water/water saturated porous sintered glass, 0.5 MHz, (b) air/air saturated porous glass system, 0.2 MHz. In the second case the predictions of single and two phase models are shown.

The specimens were made from one composite with 45% volume fraction of Terfenol-D powder, which was divided in two, and afterward cross section perpendicular and rectangular to polarization lines were made. The specimens are presented in figure 1.

‘ig. 1. Comparison of magnetostriction for solid Terfenol-D and composite material. investigated composite specimen reached 700 ppm in comparison to Terfenol-D 1500 ppm at H=180kA/m (figure 2).

Fig. 1. Sensor position regard to the specimen during investigations of Villari effect. Fig. 1. Comparison of different applied forces on changes in magnetic response of the specimen.

Fig. 1. Comparison of different position of sensor on changes in magnetic response of the specimen at the same value of applied force F=550[N].

Figure 5 shows changes in magnetic response of the investigated specimen depended on different position of sensor along the specimen as it is showed in figure 3.

magnetic response of the specimen rise together with increasing of the force, but shape of the loop did not change.

to enable registration Of mechanical signals (stress o(t) and strain e(t)) and magnetic (the magnetic field strength H(t)). For measurement of the magnetic field strength the experimenters used a unique setup developed at the Institute of Material Science and Applied Mechanics by one of the co-authors of the present work. In this setup four Philips KMZ _ 10B magnetoresistors were used as sensors. They were placed in line alongside the specimen’s section to be measured. Fig. 1. Measuring stand: 1 — hydraulic pulsator MTS 810, 2 — setup for magnetic field measurement, 3 — extensometer, 4 — magnetoresistors, 5 — specimen.

Figure 2 presents the spectrum of the stress signal for the deformation threshold ¢, = 0.0035 mm/mm. As we can see the initial stress value equaled o, = 250 MPa and in the course of the fatigue process increased up to the maximum value o, = 420 MPa. Increasing of stress during fatigue process testify to the material strengthening. This strengthening is directly connected with the increase in the a’ martensite content in the matrix of y austenitic.The stress increased up to the moment when first local fractures of the specimen occurred. It was assumed that the specimen fracture took place at the decrease in force value of 1%. Fig. 2. Spectrum of strain signal.

Fig. 1. Examples of the use of the magnetic field stream disturbance for detection of changes on the ferromagnetics surface: laser engraving A), crack on the disc B) magnetopolygraphy C).

Fig. 2. Window and main features of Maglab software.

Fig. 2: Flank damane of case-hardened gear (A) and gear with WC layer (B) manufactured practices (reference gearing), verified the accuracy of the methodology of design and strength check of gears. Experimentally was achieved the calculation of gearing lifetime. A characteristic phenomenon’ was - significant variance of values of life among each gearing.

Fig. 3: Numerical model of crack propagation

Fig. 4: Metallographic section of tooth Geometry of dimples and pitting cracks at different stages were studied on metallographic sections and cuts made in the most damaged teeth. Cuts were conducted in order to go through centre of macro-dimple. Acquired images were processed using of programme LuciaNet.

0%, 12%, 26%. The applied acoustic emission (AE) apparatus was a Defectophone (KFKI Hungary) with two logarithmic amplifiers and two wideband piezoelectric-transducers type SE1000-H. The examined frequency domain was 20-250 kHz. The applied threshold was 22 uV. The examination of 20-100 kHz range seemed to be essential due to the preliminary analysis, which showed ca. 30% of the total events are in this range (Kannar 2004). The coupling material was silicon grease. The characteristic fracture Results

the full volume of the stressed specimen, which progress randomly towards one another (Bariska 1985). Additionally, the advancement of micro- damage processes is a function of the particular specimen's biological and anatomical structure.

Fig.3. Tension fracture surface of spruce at 0% MC after tensile test

This indirect technology is simple and exact. Even so only a few similar measuring has been found in the bibliography. This method usually was referred like theoretical possibility. To make the test we need for a prime mover to material search and for a machine to determine the deflection. We can use a new instrument to determine the deflection. It is _ the videoextensometer, which give the deflection through optically and computationally way. The experimental arrangement of the simple tension tests are shown in fig. 1. Fig. 1: Picture of videoextensometer during measure 2. Experimental Results

Fig. 2: Specific strains by simple tension Tab. 1. The experimental results of the simple tension

Fig. 1. Different mmC with total chip area A,=10mm? Fig. 2. Different mmC with total chip area A,=50mm?

of homogenization annealing at temperature T = 415 °C, holding time at this temperature for 3 hours and consequential slowly cooling in the furnace. The goal of the heat treatment was removing the original dendritic structure and achieve of polyhedral grains. In Fig. 1 two damping curves marked as the 1" and the 2" measurement are compared. The measurements were performed only in one test bar. It can be seen that the damping does not considerably depend on the temperature up to 200°C for both measurements. However significant damping maximum was detected at 150 °C with other considerable smaller damping peaks in his temperature interval. Concretely the damping value at the temperature 150 °C was 1.75x10* at the 1" measurement and 1.8x10“ for the 2"" measurement. As can be seen from both damping curves the damping further increases from 200 °C with increasing of the emperature. It can be further seen from Fig. 1 hat the 2" measurement has damping phone (i.e. zone without phase transformation) moved o higher value than for the 1" measurement. Higher damping value for 2" measurement can be probably explained by partial annealing of est bar during 1" damping measurement Fig. 1 Damping curves for non-heat treated AZ31

Fig. 2 Damping curves for heat treated AZ31 alloy. The damping curves for annealed alloy are shown in Fig.2. A couple of damping measurements after heat treatment was performed in same test bar as in the case of original state of Mg alloy (without heat treatment) and is marked as the 3" and 4" measurement. These results are compared ogether with damping curves for original state (® and 2" measuremen ). It can be seen that the heat treatment considerably influenced the character of damping curves, namely by occurrence of two damping maximums at the emperatures 173°C and 230°C. The damping value was 2.8x10~ for the 3” measurement, and 3.1x10* for the 4" measurement at 173 °C and approximately 4.5x10“ for both measurements at 230 °C. From Fig. 2 can be seen that maximum damping values measured for heat treated alloy are almost three times higher than that for non- heat treated alloy. That is why the maximum peaks for original state of alloy nearly disappeared. Measured damping maximums can be explained by relaxation processes took place in material for both original and heat treated state of AZ31 alloy. The relaxation processes can be connected with dislocation motion and their interaction with vacancies, impurities, secondary phases or grain boundaries [6]. For lower temperatures (to 120°C) the internal damping process is probably a result of dislocation motion in basal planes whereas at higher temperatures grain boundary sliding plays dominant role.

The response in the form of ratio of absolute value of acceleration at top plate and base plate of the model over the ratio of excitation frequency to its first natural frequency both for TLCGD installed but without water and tuned TLCGD are shown in Figs. 2 and 3.

Fig. 1: Experimental configuration of laboratory model; green colour of water is due to dye for better visibility of its surface strokes.

Tab. 1: Average + SD of normalized deviation of individuals’ spatial-temporal parameters In conclusion, the simplified gait analysis method can be used for describing he spatial and temporal parameters of a number of gait cycles. T parameters of each subject investigated can calculated. parameters ena The bles the mode normalized ing of dynamic perception, because it is independent from t value of parameters due to normalization. O tests show thai the size of he parameter independent from lateral dominance at healt subjects. The size of the normalized deviation of parameters d he average and standard deviation of Hy epends on age, on the intensity

Fig. 1. MTS Synergie 100 loading system and videoextensometer system.

Tab. 2. Values of the stretch ration A2 and (3 for abdominal aortic aneurysms and normal abdominal aorta.

Fig. 2. Stress-strain relationship for abdominal aortic aneurysm cut in circumferential direction (AAAc). Obtained results indicate that incompressibility and isotropy is a reasonable assumption in that AAA tissues under uniaxial tensile tests and can be straightforward applicable to compute the constitutive modeling. They were described fairly precisely by an essential equation derived from used model ( ). The model applied is a phenomenological mode and may be called quasi-structural because of its conformity to the Holzapfel’s and Weizsacker’s postulate [8], which talks about the additive spli of the isochoric strain-energy function into parts connected with the different deformation of a structure; it’s mean connected with elastin and collagen fibres (Fig. 3). It’s very importan mathematical describe behavior of AAAs’ walls under deformation to take into consideration influence of complex structure.

Fig. 3. Collagen and elastin fibres in aortic abdominal aortic aneurysm walls. 3. References [1] G. Holzapfel, R. Ogden, Biomechanics of soft tissue in cardiovascular systems. SpringerWien NewYork (2003).

$Fig. 2: Results of microhardness and RS for the tempered steel W 300 machined by graphite electrode in finishing mode. The state of residual stresses (RS) was determined by X-ray diffraction (XRD), Barkhausen noise (BN) and layer removal methods (LRM). Phase composition of cut area was evaluated by Rietveld line profile analysis from XRD data [3]. Microstructure and chemical composition of white layer and heat affected zone was investigated using optical microscope and electron microprobe with EDX analysis. The samples were also examined for microhardness HV 0.2.$

Fig. 2: Results of microhardness and RS for the tempered steel W 300 machined by graphite electrode in finishing mode. The state of residual stresses (RS) was determined by X-ray diffraction (XRD), Barkhausen noise (BN) and layer removal methods (LRM). Phase composition of cut area was evaluated by Rietveld line profile analysis from XRD data [3]. Microstructure and chemical composition of white layer and heat affected zone was investigated using optical microscope and electron microprobe with EDX analysis. The samples were also examined for microhardness HV 0.2.

Fig. 1: Loading programme: (a) strain control signals; (b) stress responses. biaxial stress conditions to be achieved in the material tested.

yield point, Fig. 2. The effects are very strong and depend on the amplitude of the torsion cycles. An increase of the cyclic shear strain amplitude leads to the further decrease of the stress-strain characteristic and the ductility improvement. 4. Remarks Fig. 3: Comparison of the initial yield surface (0) and subsequent yield loci determined after mono- tonic tension and torsion cycles for the strain ampli- tudes equal to 0.3% (1). £0.5% (2). 0.7% (3).

The photographs of isochromatic pattern were taken for the models (fig.1) under various levels of loading in elastic-plastic range. The isochromatic fringes taken for the model III (with three holes) for a dark-field polariscope at chosen loading levels are shown on fig.2. Fig.2. Isochromatic fringes at loading levels: a) s = 0.879, b) s = 0.928 levels of loading in elastic-plastic range. The ihe strain components separation on the basis of the isochromatic fringes was held using the analytical method of characteristic. This method makes use of the relations between the

Fig.1. Models of the constructional elements. Corresponding author: B.Kozlowska@mchtr.pw.edu.pl Warsaw University of Technology, Faculty of Mechatronical Engineering, ul. Sw. Andrzeja Boboli 8, 02-525 Warsaw, Poland.

Fig. 3. Model of the material characteristic For strain separation and determination of the stress components in the elastic-plastic zones of the models, the 6-sectional model of material characteristic was accepted (fig.3). 3. Conclusions

Fig. 4. Analyzed area of model with three holes for loading level s = 0.928

Fig. 5. Stress distribution 0, (x, y) for s = 0.928

Fig.6. Stress distribution o, (x, y) for s = 0.928

$scan length). This results in the material fraction in a certain profile height. The Abbott curve is then created by plotting the height steps over the corresponding material fraction (see Fig. 2 and Fig. 5). The approach to determine the tribologically relevant Abbott parameters is shown in Fig. 2.$

scan length). This results in the material fraction in a certain profile height. The Abbott curve is then created by plotting the height steps over the corresponding material fraction (see Fig. 2 and Fig. 5). The approach to determine the tribologically relevant Abbott parameters is shown in Fig. 2.

Fig. 2: Evaluation of the secant with minimal inclination

3. Conclusion The Abbott parameters for surface profiles investigated with stylus instruments and with confocal laser scanning microscopes show a significant difference. To find out the reason for this phenomenon, a method for simulating the movement of the stylus edge over the surface profile was developed (see fig. 5). Through the motion of the stylus over the profile, the edge produces an envelope which was taken to be responsible for the difference between the Abbott parameters. The stylus edge was formed with a radius of 5 um and an opening angle of 90°.

Fig. 6: Stylus edge touching the surface profile From the new surface profile which is given by the envelope the Abbott parameters were calculated again.

Fig. 1: Requirements on reinforcing tire cords. Therefore tire steel-cords are exposed to various chemical and thermal influences (fig. 1) during cyclic loading states by _ tensile- compression in tire loading processes. The aggressive environment (e.g. action of salts in winter) activates the corroding process ym steel-cord surfaces [2]. If cords are with

The experimental study of macro and microstructures of tire e.g. belt plies is also important for computational modelling of tire. Fig. 3: Tensile test of steel belt ply with corrosion.

Fig. 2: Damage adhesive bond “steel-cord surfaces- elastomer drift”. 2. Experimental Results

Fig. 5: Comparison of corrosion degradation and undamaged of steel belt ply.

It is necessary also knowledge about tensile stiffness of belt ply with corrosion for different cord angles and cord form, fig. 3. The fracture character of test specimen with 25° angle and thin wire cord after tensile and corrosion test in a corrosion chamber is on fig. 4 as an example. Fig. 4: Sample of fracture character of test specimen with corrosion and detail corrosion on cord surfaces.

rT his paper deals with parallel overview o numerical calculations and static testing results fo a typical truss as a primary structural element of the steel roof construction in new Varazdin Spor Hall. This Sport Hall is one among other few sport auditoriums that had been designed and buil during 2008. for the purpose of World handbal championship held in Croatia in January 2009. Sport building codes and regulations were not the only and crucial requirements for construction type selection and design for this 5000 audience, but also awarded architectural solution, construction technology, and the most important short time for building erection. Therefore, for this 63 x 84 m plan, simple triangular braced girder was selected as the primary structural element for steel roof construction (Figure 1). So, all long term roof counterfort works, needed surely for more attractive space structures, were avoided. i ’

Numerical analysis of such subsequently softened girders had to be checked by static load testing. Corresponding autor: ivan.paska@coning.hr

Figure 6: Deflections during the test procedure The results of measured deflections on measuring places No. 6, 7 and 8 can be seen in Figure 5 and compared with computational values in Figure 6 and Table 1. The results of measured strains on measuring places No.1 and 2 are shown in Figure

“eae 3. Experimental results and comparison with numerical results of testing 4. Conclusions

Figure 7: Comparison of experimental and numerical deflections Table 1: Comparison of experimental and numerical deflections

From the measured strains stresses are calculated and compared with computational values in Table

Due to the time and budged restriction it was not possible to build testing loading equipment capable to perform a radial loading evaluation. The decision was to rely on numerical model of radially loaded spider fitting and spherical joint attachment. It is not the problem to build such a numerical model based on CAD geometry, but problem is to rely on such a model. Numerical methods, such as FEM, are always giving logical results, but question is how accurate is the numerical model.

have been used to detect the deformations of laded pair of spiders (Fig.1). When comparing the results, it is important to apply the unidirectiona strain gage in the region where principal strain direction can easily be selected. As a strain gage averages deformations on whole region of application, the region of application has to be free from stress concentrations. Therefore a numerica model, based on the CAD geometry, is the firs step needed to locate such a region. For the case of spherical joint attachment, the frictional contact problem has to be considered. Thus a numerical model of spherical joint attachment is modeled and compared with experimental data for axial loading test. The load- displacement data, deformed shape and rupture scenario have been compared. From the numerical model it can be concluded that collapse occurs due to the plastic deformation of cylindrical body and plastic yielding due to the Hertzian stresses few microns under the sphere’s surface (Fig. 3 and 4).

Table 1 Loading capacity, evaluated and declared 5. Concluding remarks

Figure 3 FEM model of the spherical joint wiht stress concetration due to the Hertzian stresses Figure 4 Yealding of the sphereical joint due to the Hertzian stresses and deformation of cylindrical body

Figure 5 The simulation of radial loading

In this concern it must be kept in mind, that in modern measurement techniques (e.g. electrical, optical, fibre-optical- methods) the respective signals like changes in resistance, voltage, intensities and frequencies etc. are measured. These analogue signals are to transform according to the inherent features of the respective technique and the relation between the metered signals and_ the displacements and strain (fig. 1). Fig. 1: Relation metered signals to strain and displacements.

The quantities C, v, 0, are unknown, thus a mixed inverse problem is on hand. A separation of the unknown quantities becomes necessary and will be carried out by means of the sensitivity matrix based method (eq.(4)).

In component formulation eq. (3) holds The strains ¢; have been obtained by evaluation of metered signals.

The main elements of the upper structure of the classical rail track constructions are the rails, sleepers and the track screen (the ballast). The connection of the rails and the sleepers is achieved by an appropriate method of fastening. The Fig. 1 shows the longitudinal cross-section of the classical rail track construction, [1]. If the rail track ballast is observed, its main purpose is to elastically and equally transfer the loads caused by the rail vehicles to the plane of the lower structure, to secure the direction and height of the rail track and to dampen the vibrations in the rail track. For the ballast to fulfil the said requirements, it must have sufficient dimensions (width and thickness) and must be constructed of a good quality material. The best material for the ballast is crushed stone produced by eruptive origin. The optimal thickness of the ballast is 25 to 30 cm measured from the lower par of the sleeper (Fig. 1). This paper considers impact of the thickness of the ballast on the vertical stability of the rail rack. The measurements were performed on he rail grid of the Croatian Railways (location: Skrljevo - Meja - Pecine). The said section is in a cut and has longitudinal slope of 22 %o, hickness of the ballast amounts to 15 to 20 cm. During the year 2005 the wooden sleepers were replaced with the concrete sleepers. But, after wo years from the replacement, the new

sleepers sunk due to further crushing of the ballast (Fig. 2). Due to subsequent crushing, the crushed stones lost their sharp edged form, which in tum lead to reduction of mutual wedging of the granules, which finally lead to settlement of the sleepers. Fig. 2: Settlement of the concrete sleepers.

Fig. 4: Deflection of the sleeper (operational train).

transversal shifts or vibrations of the constructions simultaneously in two orthogonal directions. The device is based on a laser transmitter and optic-electronic receiver that are mounted on the object where the measurements are being made. The Fig. 3 show the measuring equipment PSM90 fitted on the rail track.

Fig. 5: Deflection of the sleeper (freight train).

Fig. 6: Deflection of the sleeper (passenger train).

The scheme of the part of stainless steel sealing component for air condition system is shown in fig. 1. Fig. 1: The scheme of the part of measured component (side and top view).

Fig. 2: Image of parts of circular teeth with height measurement. First, height of each tooth was measured according to the reference plane (fig. 2, 3). Average height was computed to be 217 um for the outer circle and 242 um for the inner one with 2 um standard deviation.

Fig. 3: X-Z profile 10 from the previous image.

Fig. 2: Tension test — fibres angle 30° and 45°. Fig. 1: Bending test — fibres angle 30° and 45°. Fig. 1 shows that both computational models give contradictory results. Results obtained by the bimaterial model show always a higher stiffness than results obtained by the unimaterial model. A causation of the difference between results is obviously due to the zero bending stiffness in the unimaterial model as mentioned in introduction.

test. Complete results for all declinations of fibres were published in [5].

The cement samples were prepared by hand and vacuum mixing technique, both with exact timing. The quantitative composition of the used bone cement is presented in Tab. 1.

Fig. 1: Surface temperature courses of the hand- mixed samples with 2mm thickness The measured surface temperature courses are represented in Fig. | - 4. The averaged values of peak temperatures and times required to obtain the peak temperatures are represented in Tab. 2 - 3.

Fig. 2: Surface temperature courses of the hand- mixed samples with 5mm thickness

Tab. 2: Averaged values of peak temperatures and times required to obtain the peak temperatures of differently thick hand-mixed cement samples Fig. 3: Surface temperature courses of the vacuum- mixed samples with 2mm thickness

‘ig. 4: Surface temperature courses of the vacuum- mixed samples with 5mm thickness

Fig. 2: Results of wear measurements. The coefficient of traction was monitored during each test and mainly disc weight loss, disc diameter loss, hardness and_ surface roughness were investigated at the end of every test. After specimen sectioning the surface layer was prepared for metallographic observation in order to gain a conception about maximal defects and shear strain accumulation. Some results of the contact fatigue with fixed slipping can be seen in Fig. 2. At the same time, measurements of residual stresses are provided before and after contact loading. Three methods of experimental stress analysis were used for measurements — hole drilling strain gage method, X-ray diffraction method and magnetoelastic method. The residual stresses before contact loading are relatively small and tensile (~5O0 MPa). The

Fig. 3: Courses of principal stresses in pulley after 2 000 000 cycles by hole drilling method.

Fig. 4: Residual stresses by X-ray diffraction measurement.

The results of residual stresses obtained by means of X-ray diffraction method after 2 000 000 cycles is shown in Fig. 4.

Fig. 1: Positions of Strain Gages and Transducers on Testing Specimen To determine buckling load of axially compressed composite panel here strain gauges are used. Strain gauges were applied on both sides of the layered composite plate in loading direction, Fig. 1.

Finite element model and first buckling mode is shown in Fig. 3. In Figure 4 are given experimental and computation results of buckling and post-buckling behaviour using relation between load P and out of plane displacements (W).

Fig. 3 The First Buckling Mode by FEM Much research has been conducted on buc kling composite plates. A review of the subject by Leissa' quotes more than 90 references. Most of his work is theoretical in nature, and very if any, attention is given to post-buc post-buckling behaviour”*. ittle, kling behaviour. To determine stability behaviour of composite panel here non-linear finite element analysis is used to determine buckling load and

Fig. 1: Phoenix electrical locomotive After such modification, tests are made in purpose to obtain a new certification from a Railway Commission. The tests were carried out according to ERRI B12/RP 17 (ERRI — European Railway Research Institute). 2. Measurement points

3. Tests The acquisition software used was Catman 4.5 (an Hottinger product). The tests were performed in two steps:

5. Conclusions In the figure 3 is presented graphically how the tests were done.

Regarding to Goodman — Smith diagram, al measured stresses were on “a2” curve (figure 3). The results, was smaller than permissible stress, so we can concluded that the design solution used by manufacturer was good.

Fig. 2: Dabar Bridge — structural system. Dabar Bridge, fig. 1, is 330 meters long double bridge (separate longitudinal structures for every traffic direction), over seven spans of 40 + 5 x 50 + 40 meters. It was designed as classical reinforced concrete plane frame system with longitudinal I-girders (fig. 2) connected only with continuous 20 centimetres thick reinforced concrete deck.

Fig. 3: Dabar Bridge — static loading. Static load was performed with six trucks weighting approximately 35 metric tonnes each, see fig. 3, placed in the mid-spans. Vertical displacements were measured by geodetic method of levelling in every span along four lines (A, B, C and D), over the supports (1, 3, 5, 7, etc.) and in the mid-spans (2, 4, 6, 8, etc.), altogether at 12 points on each span.

Fig. 1: Grandstand RC girder cross-section. 2. Single Girder Testing

Strains of concrete were measured by Barry movable extensometer positioned at the supports and in the mid-span, see fig. 3.

Fig. 5: Assembled grandstand structure. Assembled girders were connected in the thirds of the spans creating monolithical grandstand structure, fig 5. Only dynamic testing was performed by jumping of humans.

Maximal designed bending moment i.e. corresponding testing load is presented in fig. 6. Fig. 6: Static loading performance.

Fig. 1: Manner of collecting sample: cutting slice a), cutting sample b), sample c) Corresponding author: adam.mazurkiewicz@utp.edu.pl

Tab. 2: Value of determination coefficient R’ for the relationship of BMD with fractal dimension

Tab. 1: Value of determination coefficient R° for the relationship of BMD with volume of layer between BMD and fractal dimensions. In the tables, maximal values of R’ are presented for the study of the description of power, logarithmic and numerical functions.

To characterize the deformation behaviour of timing belt tooth during the contact with a steel counterpart a novel testing device was developed and implemented on a servohydraulic testing machine and the set-up is shown Fig 2. The belt was preloaded in the test equipment and the belt tooth was loaded by the movement of a steel counterpart pulley. The local strains in the contact area and in the entire tooth were measured by digital image correlation technique. The development of the major strain during the loading stages is shown in Fig. 3 for a specific belt type and tooth profile. The highly

Investigation of the "Arena Zagreb" multifunctional hall´s roof

Sign up for access to the world's latest research

Abstract

Related papers

Related topics

Related papers