Jtam.dvi JOURNAL OF THEORETICAL AND APPLIED MECHANICS 46, 4, pp. 933-947, Warsaw 2008 AN EXPERIMENTAL ELECTROMAGNETIC INDUCTION DEVICE FOR A MAGNETORHEOLOGICAL DAMPER Bogdan Sapiński AGH University of Science and Technology, Department of Process Control, Cracow, Poland e-mail: deep@agh.edu.pl The work presents an experimental electromagnetic induction device consisting of permanentmagnets and a coilwhich produces electric ener- gy for an attachedmagnetorheological (MR) damper. The study covers design considerations and calculations of magnetic fields, description of the engineered device, and results of experimental tests on the dynamic testing machine. Key words: electromagnetic induction, magnetorheological damper, vibration 1. Introduction A conventional MR damper-based vibration control system is a feedback sys- tem (Fig.1). Sucha system includes: a sensor (to detect vibration), a controller (to manipulate the signal obtained from the sensor according to the assumed control law) and an MR damper (to control the mechanical response of the structure) which is activated by a power controller. In some applications, in- stallation and maintenance of such system can be complex (e.g. large-scale civil structures) and, therefore, new solutions are sought. The drawbacks of the feedback system can be eliminated by introducing a system with an electromagnetic induction device attached to an MR damper (Cho et al. 2005). A schematic diagram of the system is shown in Fig.2. This system involves coupling of vibration of the structurewith the force generated by the MR damper. The electromagnetic induction device produces electric energy (induced current), according toFaraday’s law,which is used to activate theMRdamper. Such a systemdoes not need any sensor, controller andpower controller, and can adjust itself to structural vibrations. 934 B. Sapiński Fig. 1. ConventionalMR damper-based vibration control system Fig. 2. MR damper-based vibration control systemwith an electromagnetic induction device The study summarizes the background design objectives and testing of an experimental electromagnetic induction device which consists of permanent magnets andacoil.The testswere conductedusingadynamic testingmachine. 2. Operation principle The electromagnetic induction device uses the kinetic energy of reciprocating motion of theMRdamperwhich is converted to electric energy. These conver- tion proceeds according to Faraday’s law which expresses the phenomenon found experimentally. An experimental electromagnetic induction device... 935 Themathematical form of Faraday’s law is e=− dΨ(t) dt =−N d dt ( ∫ S B ·dS ) (2.1) where Ψ is magnetic flux linkage through the area of coil cross section S, N is the number of turns, B is themagnetic flux density and e is the induced electromotive force (in volts). Theminus sign is amanifestation of Lenz’s lawwhich states that the sense (clockwise or counterclockwise) of the induced electromotive force created by a change in themagnetic fluxmust be such as to oppose the change producing it. Recalling Eq. (2.1), it is apparent that the time change can work in two different ways. The first is through time changes in the magnetic flux density B(t), and the second is through changes in the shape and orientation of the coil area S. Combination of both can also occur. Note that by defining the magnetic flux linkage as: dΨ =NdΦ=NB ·dS=NBdS cosϕ (2.2) (2) where ϕ is the angle between B and dS. In the case of time changes in the magnetic flux density B(t), Eq. (2.1) can be rewritten in the form e=−N dΦ dt =−NS dB dt (2.3) Thus, Faraday’s law states that the induced electromotive force in the coil equals the negative of the time rate of the magnetic flux change through the coil. In other words, relative motion between the coil and permanentmagnets of the considered device causes a change of the magnetic flux that induces an electromotive force in the coil. This force can also be regulated by the number of turns of winding in the coil or the intensity of permanent magnets. 3. Design considerations and calculations of magnetic fields In the design procedure, we assumed that the electromagnetic induction devi- ce: • is attached to theMR damper of RD-1005-3 of Lord Co. (http://www.lord.com), • incorporates moveable permanent magnets and an immoveable coil. 936 B. Sapiński Main parameters of the RD-1005-3 damper whose structure is shown in Fig.3 are: stroke ±25mm, input voltage 12VDC, input current; continuous < 1.0Aand intermittent < 2.0A, coil resistance 5Ω (at 25◦C), force (peak to peak); 2224N (51 ·10−3m/s, 1A) and 667N (200 ·10−3m/s, 0A), response time < 25ms (time to reach 90%ofmax. level during a 0A to 1A (dependent on the amplifier and power supply). Fig. 3. Structure of the RD-1005-3 damper Force-velocity loops of this damper obtained for a sine displacement exci- tation of the piston by amplitude 10 ·10−3m and frequency 1Hz, and input currents 0.0, 0.2, 0.4A are shown in Fig.4 (Sapiński, 2006). The plots reveal that a higher current yields a greater force response while the influence of piston velocity is less significant. Fig. 4. Force-velocity loops of the RD-1005-3 damper The designed electromagnetic induction device incorporatesmagnets who- se arrangement and orientation is explained in Fig.5 (Sapiński, 2008). The magnets are magnetized along their height, which implies that one circular surface acts as the N pole, and the opposite becomes the S pole. Themagnets are arranged in two sets and are held on a bolt made of non-ferromagnetic material. The number of magnets in each set is four. They are arrayed such An experimental electromagnetic induction device... 937 that in each magnet the surface acting as the N pole should interface the S-surface of the nextmagnet. Themagnet sets aremounted on the bolt at the distance Hg such that the surface of one magnet being the N(S) pole should be placed opposite the same surface in the other set. Fig. 5. Sketch of magnets arrangement on the bolt The device employs sintered ring-shaped neodymium magnets (http://www.magnesy.pl). The magnetic flux density from the magnets surface in the distance 0.7 ·10−3m is 0.30T. The demagnetization characte- ristic of these magnets shown in Fig.6 (II quadrant) is given by B=µ0H+J (3.1) where H is themagnetic field strength, J –magnetization and µ0 ismagnetic permeability of vacuum. For the magnets arrangement shown in Fig.5 we computed the magnetic field distributions, assuming that: the magnet height is hm = 5mm (Hm = 20mm), magnet diameter is Φm =30mm, bolt diameter is Φb =10mm and the distance between magnet sets is Hg =20mm. The calculation procedure used the FLUX 2D program (Cedrat, 2000). Figure 7a presents the applied 2Dmeshmodel of the device with the coor- dinate system (r,z). Themodel yields 3819 cells and 7700 nodes. The derived distribution of magnetic field lines is shown in Fig.7b. Note that in the assu- med orientation of magnets, most of the magnetic field lines is concentrated in the gap between the magnets sets. The largest concentration of these lines occurs on the outer edges of the magnets. 938 B. Sapiński Fig. 6. Demagnetization characteristics of magnets Fig. 7. (a) Meshmodel, (b) magnetic field lines Next we computed the magnetic flux density in the device, knowing that the gap between the magnet sets was filled either by air or by a ferroma- gnetic material of cylindrical shape (see the dot line in Fig.5). The magnetic permeability of this material is µ = 104µ0 (µ0 = 4π10 −7H/m is magnetic permeability of vacuum). The derived components of themagnetic flux densi- ty are shown in Figs. 8 and 9. The plots depicted by the continuous line deal with the gap filled by air and by the dot line – by a ferromagnetic material. Figure 8 shows the radial component of the magnetic flux density along the radial axis r. It appears that the largest value of this component, found near the side surface of the magnet, approaches 0.17T for air in the gap and 0.31T for the ferromagnetic material. Figure 9 presents the radial component of the magnetic flux density along the longitudinal axis z. It is apparent that An experimental electromagnetic induction device... 939 Fig. 8. Radial component of the magnetic flux density along the radial axis Fig. 9. Radial component of the magnetic flux density along the longitudinal axis this component varies slightly between two sets, in the range (0.14,0.17)T for air in the gap and takes nearly a constant value of 0.20T for the ferromagnetic material. 4. Description of the device Theelectromagnetic inductiondevice engineeredutilizing the above calculated data is shown in Fig.10. The construction of the device enables us to regulate the number ofmagnets and the height of the gap between themagnet sets and to introduce coils with various electric parameters. The height of the gap set at Hg =20mm(which can befilled either by air or by ferromagneticmaterial) results from the amplitude of displacement excitations of the piston assumed in experiments. Care must be taken to ensure that the magnets held on the 940 B. Sapiński bolt can freely move inside the tube. Besides, the width of the slit formed as a result of the difference between the inner diameter of the tube and the magnets diameter should be as little as possible. Fig. 10. Engineered electromagnetic induction device The device was equipped with two coils wound with a copper wires with diameter of 0.25mm (coil 1) and 1.25mm (coil 2) on a carcass. The coils height was 22mmand the internal diameter was 34mm.Coil 1 had 912 turns and resistance 39.6Ω while coil 2 had 45 turns and resistance 0.1Ω. Care must be taken to ensure that the coil windings should be close to themagnets in order to remain the largest proportion of the windings in the zone of the maximum flux density. 5. Experiments and discussion For the purpose of experiments we made a special handle to hold the elec- tromagnetic induction device and the RD-1005-3 damper. The handle design enables motion of its upper part with respect to the lower one in the ran- ge 10mm (for the initial damper deflection). In the guides there are roller bearings which allow two parts to be precisely guided with respect one to another. The handle with the device and damper installed is shown in Fig.11. We tested the device in an experimental setup equipped with a dynamic testing machine of Instron and data acquisition system based on a PC with an I/O board of National Instruments installed and supported by LabView environment runningonMSWindows (http://ni.com).Theschematicdiagram of the experimental setup is shown in Fig.12. An experimental electromagnetic induction device... 941 Fig. 11. System ready for tests Fig. 12. Diagram of the experimental setup We programmed the machine to produce sine displacement excitations of the piston with the amplitude 10mmand frequency 1Hz. This enabled us to achieve the maximum piston velocity associated with the machine efficiency. The data were acquired in 10 cycles (a cycle is denoted as the piston displace- ment up and down). The sampling frequency for each channel was 1kHz per cycle. Main testswere precededby investigation of impacts of the friction force in the handle. That is whywe tested at first the handle in which only the device was held. The obtained results are shown in Fig.13. The plots present time patterns of the friction force measured for a sine excitation of the piston. It appears that the friction force takes the value in between 50 and 70N. These impacts should be taken into account when analysing the influence of induced current on the damper force in the main tests. 942 B. Sapiński Fig. 13. Friction force in the handle for a sine displacement excitation Next we investigated the influence of the input current on the damper force. For that purpose, we activated the damper installed in the handle by a step input current during the programmed excitations of the piston. The achieved results are provided in Fig.14. The plots present time patterns of the damper forcemeasured for the sine excitation of the piston and the step input current from 0 to 0.1A (see grey line). It appeared that the damper force for no current in the coil varied in between −100 to 30N while for 0.1A from −150 to 80N. Fig. 14. Damper force for a sine displacement excitation by a step input current The main tests were conducted assuming that the device was equipped with coil 1 or coil 2 and the gap between two sets of magnets was filled by air or by a ferromagnetic material. The obtained results are shown in Figs. 15 and 16. In the case with no connection between the device and the damper An experimental electromagnetic induction device... 943 Fig. 15. Electromotive force, voltage and damper force for the sine displacement excitation; the device with coil 1, gap filled by air (left) and ferromagnetic material (right) (open circuit) we measured the induced electromotive force and the damper force, and when the device was connected with the damper (closed circuit) we measured the voltage at the coil terminals and the damper force. The plots present time patterns of the measured quantities obtained for a sine displacement excitation for the device with coil 1 (Fig.15) and with coil 2 (Fig.16). The plots depicted by black line apply to the open circuit and by grey line to the closed circuit. The displacement excitations of piston in Figs. 15 and 16 are depicted by continuous lines and the velocity by the dashed line. 944 B. Sapiński Fig. 16. Electromotive force, voltage and damper force for the sine displacement excitation; the device with coil 2, gap filled by air (left) and ferromagnetic material (right) Comparing the results obtained for the open circuit (no current in the coil) with the closed circuit (current in the coil is that induced by the de- vice) reveals slight variability of the damper force. This can be explained as follows. Figure 17 (left) shows a general sketch of the closed circuit. Taking into account the sine displacement excitation, the current i in the circuit is given by An experimental electromagnetic induction device... 945 i= Em √ (Re+Rc) 2+ω2(Le+Lc) 2 sin2πft (5.1) where Em is the maximum value of the induced electromotive force, f is the frequency of displacement excitation of the piston, Re and Le is the resi- stance and the inductance of the device, Rc and Lc is the resistance and the inductance of the damper coil. Fig. 17. Sketch of the closed circuit of the electromotive induction device: general (left), simplified (right) Due to a low operational frequency in the experiments (1Hz), the induc- tive reactance of the device coil and damper coil is reasonably small (when compared to the resistances) 2πfLe+2πfLc ≪Re+Rc (5.2) and therefore can be ignored. Hence the closed circuit can be treated as a circuit with only ohmic resistance, as depicted in Fig.17 (right). Accordingly, we write i= Em Re+Rc sin2πft (5.3) and this yields Umc =RcIm (5.4) where Umc is the maximum voltage at the damper coil terminals and Im is the maximum current in the closed circuit. A similar statement is applicable to the triangular displacement excitation. The conducted tests revealed that introducing ferromagnetic materials in between themagnets sets led to an increase in the damper force when compa- red with the gap filled by air (compare force plots depicted by grey and back 946 B. Sapiński lines inFigs. 15 and16).This contribution, however not significant,was caused by the increase of the magnetic flux density in the gap. This is in agreement with the results of calculations of magnetic fields (Section 3). Also we did not observe almost any difference in the device performance whenusing coil 1 and coil 2.Though the induced electromotive forcewasmuch greater for coil 1 (than that for coil 2) we got a relatively small and almost the same current in the damper. The reason is that the resistance of coil 1 is much bigger than that of coil 2, and this makes that the induced currents of about 0.01A are too low to affect the damper force significantly (compare plots in Figs. 15 and 16). 6. Summary The work reports the initial stage of the author’s research devoted to self- powered vibration reduction systems with MR dampers using motion of the structure for generation of a magnetic field. The study presents design con- siderations, calculations of magnetic fields and testing of the experimental electromagnetic induction device in a dynamic testing machine. The investigation revealed that the engineered device was able to produce induced electromotive forces.Unfortunately, theywere too low to influence the damper force significantly. The main reason was that, due to a low efficiency of the machine, we could not achieve a higher piston velocity and, thus, a higher induced current. Therefore, further experiments will be conducted on a machine whose parameters ensure fast relative motion between permanent magnets and coil. The investigation also highlighted directions for the improvement of con- struction of the engineered device. This will be achieved by certain modifica- tions in the magnetic system of the device and also in its electric circuit. Acknowledgements The research work has been supported by the State Committee for Scientific Research as a part of the scientific research project no. N501366934. References 1. Cedrat, 2000,User’s guide FLUX 2D version 7.50–CAD Package for Electro- magnetic and Thermal Analysis Using Finite Element Method An experimental electromagnetic induction device... 947 2. Cho S.W., Jung H.J., Lee I.W., 2005, Smart passive system based on a magnetorheological dampers, Smart Materials and Structures, 1, 707-714 3. Sapiński B., 2006, Magnetorheological Dampers in Vibration Control, AGH University of Science and Technology Press, Cracow, Poland 4. Sapiński B., 2008, An investigation of an electromotive induction device for a small-scale MR damper, The 4th International Conference on Mechatronic Systems and Materials, Białystok, Poland 5. http://www.lord.com 6. http://www.magnesy.pl 7. http://ni.com Doświadczalny generator elektromagnetyczny dla tłumika magnetoreologicznego Streszczenie Praca dotyczy doświadczalnego generator elektromagnetycznego składającego się z magnesów stałych i cewki, który wytwarza energię elektryczną dla dołączonego do niego tłumika MR. W pracy przedstawiono projekt generatora, obliczenia pola magnetycznego, opisano wykonany generator oraz wyniki jego badań na maszynie wytrzymałościowej. Manuscript received March 20, 2008; accepted for print May 21, 2008