Development of a novel mr clutch featuring tooth-shaped disc

Nội dung text: Development of a novel mr clutch featuring tooth-shaped disc

1. Vietnam Journal of Mechanics, VAST, Vol.43, No. 3 (2021), pp. 265 – 276 DOI: DEVELOPMENT OF A NOVEL MR CLUTCH FEATURING TOOTH-SHAPED DISC Quoc Hung Nguyen1,∗, Bao Tri Diep2,3, Duy Hung Nguyen1, Van Bien Nguyen3, Van Bo Vu3, Qui Duyen Do1 1Faculty of Engineering, Vietnamese-German University, Binh Duong, Vietnam 2Faculty of Civil Engineering, HCMC University of Technology and Education, Vietnam 3Faculty of Mechanical Engineering, Industrial University of Ho Chi Minh City, Vietnam ∗E-mail: hung.nq@vgu.edu.vn Received: 14 February 2021 / Published online: 30 September 2021 Abstract. In this research, we focus on development of a new configuration of magneto- rheological fluid (MRF) based clutch (MRC) featuring a tooth-shaped disc with multiple teeth acting as multiple magnetic poles of the clutch. The tooth-shaped disc is placed in a clutch housing composed of the left housing and the right housing. The inner face the housing also has tooth shaped features mating with the teeth of the disc through the working MRF. Excitation coils are placed directly on stationary winding cores placed on both side of the clutch housing. An air gap of 0.3 mm is left between the housing and the winding cores to ensure the housing can freely rotate against the winding cores. After the introductory part, configuration of the MRC is introduced and the transmitted torque of the MRC is derived. An optimization process to minimize the overall volume of the pro- posed clutch, which can generate a required maximum braking torque, is then conducted. The optimal results show that the overall volume of the proposed MRC is significantly re- duced compared to a referenced conventional MRC (0.159 m3 vs. 0.295 m3). A prototype of the proposed MRC is fabricated for experimental works and good agreement between the experimental results and simulated ones is archived. Keywords: magnetorheological fluid (MRF), MR clutch, tooth-shaped rotor, optimal de- sign. 1. INTRODUCTION Magnetorheological fluid, called MRF for short, is a type of smart fluids, composed of tiny magnetic particles (micro to nanometer size) dispersed in a base (carrier) fluid. In case of nonexistence of magnetic field, the particles are dispersed randomly and MRF rheologically behaves as Newtonian fluid like carrier fluid. In the existence of a magnetic field, the particles are magnetized, which consequently experienced attractive forces, forming chain-like structure. At high field intensity, the MRF is almost solidified. This change is very fast (few milliseconds) and almost reversible. With this especial property, © 2021 Vietnam Academy of Science and Technology
2. 266 Quoc Hung Nguyen, Bao Tri Diep, Duy Hung Nguyen, Van Bien Nguyen, Van Bo Vu, Qui Duyen Do the MRF is very potential for various applications such as valves, dampers, clutches, brakes and engine mount [1,2]. There have been a number of researches on MRF based clutch (called MR clutch or MRC for short in this paper). In the early stage, most of the researchers proposed the MRC configurations with rotating coils [4–6]. Obvi- ously, this configuration possesses several dis- advantages such as difficulties in manufactur- ing, unsteady and high friction due to brushes, “bottle-neck” problems of magnetic circuits. To overcome the above disadvantages, new configurations with stationary winding hous- ing have been recently implemented. In this new configuration, the coils of MRCs are wounded on a fixed housing, on which the in- put shaft (usually connected to the disc of the MRC) and the output shaft (usually connected to the housing of the MRC) are assembled as shown in Fig.1[3,7,8]. Thanks to this new con- figuration, the coil is stationary and brushes can be eliminated. Hence, some disadvan- tages of conventional MRC such as magnetic flux “bottle-neck” problems, unsteady contact of the brushes, manufacturing difficulties, etc. Fig. 1. Configuration of the previous can be handled. In previous researches, the MRC [3] clutch discs have cylindrical shape and the torque is transmitted from the input shaft to the output shaft via the MRF in the ducts at the end-faces and outer cylindrical face of the disc. Therefore, in order to archive large transmitted torque, the disc should have large size, which results in large size of the MRC and high inertia effect. The main purpose of this study is to develop a compact size MRC featuring a tooth- shaped disc. With the toothed shaped disc, the working area between the disc and the MRF is larger and a high transmitted torque can be archived while size of the MRC is still kept compact. 2. THE MR CLUTCH FEATURING TOOTH-SHAPED ROTOR The configuration of the proposed MRC with tooth-shaped disc is illustrated in Fig.2. From the figure, it can be seen that a tooth-shaped disc (made of magnetic ma- terial) is fixed to the driving shaft (made of nonmagnetic material). The disc is placed in- side housing of the clutch. The inner face of the clutch housing has teeth as well, mating with the teeth of the disc via MRF medium. One side of the clutch housing is connected to the driven shaft (made of nonmagnetic steel). In order to avoid the MRF leaking, a mechanical lip-seal is used. Two ball bearings are used for relative rotation between the
3. Development of a novel MR clutch featuring tooth-shaped disc 267 housing (output shaft) and the disc input shaft). Two fixed winding cores, on which the coils are wound, are assembled on both side of the clutch housing. The cores are attached to the corresponding fixed supports, on which the clutch shafts are also assembled. By assembly the cores and the clutch shaft on the support, relative position between the clutch housing and the cores can be assured. To ensure a free rotation of the clutch hous- ing relatively to the winding cores, a small air gap (0.3 mm) between the housing and the winding cores is employed. When a current is applied to coils, an induced magnetic field is created causing the MRF becomes solid-like. By control the applied current, a controllable transmitted torque from the input shaft to the output shaft can be archived. Fig. 2. Configuration of the proposed tooth-shaped mRC In order to derive transmitted torque of the MRC, firstly a small ring element in an inclined duct of MRF shown in Fig.3 is analyzed. The friction torque acting to this element can be calculated by 2 2 dT = rτdA = 2πr τdl = 2π (R1 + l sin ϕ) τdl, (1) where r is the radius of the small ring element; R1 and R2 are the radii of the first end and the second end of the inclined duct respectively, and L is the length of the inclined duct. By assuming a linear velocity profile of MRF in the duct, the Bingham model of MRF in the duct can be mathematically expressed by r∆ω δω (R + l sin ϕ) τ = τ + µ = τ + µ 1 , (2) y d y d where τ is the tangent shear stress acting on the MRF element; ϕ is the angle between the inclined duct and the rotational axis; δω is the relative angular velocity between the
4. 268 Quoc Hung Nguyen, Bao Tri Diep, Duy Hung Nguyen, Van Bien Nguyen, Van Bo Vu, Qui Duyen Do two walls of the duct, which is also the relative angular speed between the disc and the housing); d is the duct gap size; τy and µ are in sequence the yield stress and the post- yield viscosity of the MRF. Fig. 3. The ring element of MR fluid in the inclined duct Plug Eq. (2) into Eq. (1), then take integration along the duct, the following equation can be obtained Z L   2 δω (R1 + l sin ϕ) TI = 2π (R1 + l sinϕ) τy + µ dl 0 d  1  = 2πL R2 + R L sinϕ + L2 sin2 ϕ τ (3) 1 1 3 y 1 L∆ω + πµ 4R3 + 6R2L sinϕ + 4R L2 sin2 ϕ + L3 sin3 ϕ
   . 2 d 1 1 1 Eq. (3) is used for calculation induced torque in an inclined duct of MRF. From Eq. (3), the induced torque in an annular and radial duct (end-face duct) can be retrieved respec- tively with ϕ = 0 and ϕ = 90 degrees, which were widely used in many researches [9, 10] as followings  ∆ΩR  T = 2πR2L τ + µ a , (4) a a a y d 4  4! πµeRo Ri 2πτy 3 3
   TE = 1 − ∆Ω + Ro − Ri , (5) 2d Ro 3 where Ta is the friction torque of MRF in the annular duct acting on the cylindrical face of the disc, La and Ra are respectively the length and the radius of the annular duct, Te is the friction torque of MRF in the end-face duct acting on the end-face of the disc, Ri and Ro are respectively the inner and outer radius of the end-face duct. By applying Eqs. (3), (4) and (5) for the MRF ducts of the proposed MRC as shown on Fig.3 and neglecting of friction due to the sealing and bearings, the induced transmitted
5. Development of a novel MR clutch featuring tooth-shaped disc 269 torque can be determined by ! Tb = 2 ∑ TIi + ∑ TEj + Tc, (6) i=1,3,5,7,9 j=0,2,4,6,8,10 where  1  T = 2πl R2 + R lsinφ + l2sin2φ τ Ii i i 3 yIi 1 l∆ω + πµ 4R3 + 6R2lsinφ + 4R l2sin2φ + l3sin3φ
   , (7) 2 Ii d i i i 4  !4 πµEjR R 2πτ   = j+1 − j + yEj 3 − 3 TEj 1 4  ∆Ω Rj+1 Rj , (8) 2d Rj+1 3 ∆ΩR T = 2πR2 (b + 2h)τ + µ 11 , (9) c 11 yc c d In the above, TEj is the friction torque induced by MRF in the duct Ej, TIi is the friction torque induced by MRF in the duct Ii, Tc is the friction torque due to MRF in the cylindri- cal duct C, Ri is the radius of point i shown in Fig.4, l and φ are respectively the length and angle of the inclined duct, h is the tooth height, τEi, τIi and τc respectively are the yield stress of the MRF in the Ei, Ii and C ducts, µEi, µIi and µc are the post-yield viscosity of the MRF in the Ei, Ii and C ducts respectively. Fig. 4. Illustration of MRF ducts for evaluating transmitted torque
6. 270 Quoc Hung Nguyen, Bao Tri Diep, Duy Hung Nguyen, Van Bien Nguyen, Van Bo Vu, Qui Duyen Do It is noted that the transmitted torque of the conventional MRC [3] is determined by 4  4!   πµRdo Rdi 4πτye 3 3
   2 ∆ΩRdo Tbc = 1 − ∆Ω + Rdo − Rdi + 2πRdotd τy0 + µ , d Rdo 3 d (10) where Rdi and Rdo are the inner and outer radius of the disc, d is the MRF gap size, td is the thickness of the disc, τye is the average yield stress of MRF in the in the end-face ducts of the MRC, τy0 is the zero-field yield stress and µ is post-yield viscosity of the MRF 3. OPTIMAL DESIGN OF THE MR CLUTCH In this study, the optimization of the MRC is stated as following: Find optimized value of geometric dimensions of the clutch structure in order to archive a required trans- mitted torque considering the objective to minimize overall volume of the MRC. The overall volume of the MRC is determined by 2 VMRC = πR L, (11) where R is the overall radius of the MRC and L is the overall length of the MRC. The de- sign variables are: the disc thickness td, the tooth-height h, the tooth angle φ, tooth-peak length ltp, tooth bottom length ltb, the inner disc radius Rdi (= R0), radius of first tooth R1, the outer disc radius Rdo(= R11), outer housing thickness of the clutch to, the winding core thickness th, the coil width tc, the coil height hc. The MRF gap size is empirically chosen by 0.8 mm. The thickness of the housing wall is set by 1.25 mm in consideration of manufacturing issues. In order to solve the optimization problem of the MRC, the ANSYS commercial software is employed, in which the axisymmetric element (PLANE 13) and the steepest descent algorithm (first order method) are used. In this study, the silicon steel is used for magnetic components of the MRC such as the side housing, the envelope and the disc. As abovementioned, in the optimization, the maximum achiev- able transmitted torque of the MRC is constrained to be greater than a required torque. Obviously, the higher applied current is, the stronger magnetic field is generated which in turn results in a higher transmitted torque. Therefore, in order to archive the maxi- mum transmitted torque, the maximum working current (depend on the coil gauge size) is applied to the coils. In this research, the 24-gauge copper wire (the coil wires diame- ter is 0.511 mm, maximum working current is 2.5A) is used. Hence, the applied current used in the optimization process is 2.5A. The commercial MR fluid, MRF132-DG, is used. For comparison purpose, optimization of the conventional MRC with plain cylindrical disc [3] is also conducted. Fig.5 shows the optimal solution of the plain disc MRC while Fig.6 shows that of the proposed tooth-shaped disc MRC. In the optimization, the required transmitted torque is 10 Nm with an accuracy of 2%, the convergence condition is set by 0.1%. In addition, radius of the clutch shaft is chosen by Rs = 6 mm accounting for the strength of the shaft. From the results, it is found that, at the optimum, transmitted torque of the MRCs is almost 10 Nm as constrained and the overall volume of the conventional MRC is 0.295E-3 m3 while that of the proposed tooth-shaped disc MRC is 0.159E-3 m3. Thus, the overall volume of the proposed MRC is much improved (almost half). The reasons
7. power consumption of the MRCs. Therefore, in the application where high power is available ant not significant, the proposed MRC can be employed. In order to compromise between the size and the power consumption of the proposed MRC, the power consumption should be taken into account as a constraint function in the optimization. Figure 7 show the optimal solution of the proposed MRC with the required transmitted torque is 10Nm and the power consumption is constrained to be smaller than that of the conventional one (44W). It is noted that in this case the applied current is also considered as a design variable. From the figure it is observed that, with the above transmitted torque and power consumption, the overall volume of the proposed MRC is 0.166E-3m3, which is still significantly smaller than that of the conventional one (0.295E-3m3). It is noteworthy that at the optimum the applied current is power consumption of the MRCs. Therefore, in1 .the77A application. From the whereabove, high it is power remarked is available that in order to compromise between the size and ant not significant, the proposed MRC can be employed.power consumption In order to of compromise the MRC, the bet powerween theconsumption should be conserved as a constraint size and the power consumption of the proposedfunction MRC, and the the powerapplied consumption current should should be treated be as a design variable. taken into account as a constraint function in the optimization. Figure 720 show the optimal Rdi 0.1xRdo td th solution of the proposed MRC with the required transmitted torque is 10Nm and the powerh w t t 16 c c es ec consumption is constrained to be smaller than that of the conventional one (44W). It is noted that in this case the applied current is also considered as a design variable.12 From the figure it is observed that, with the above transmitted torque and power consumption,8 the overall 3 volume of the proposed MRC is 0.166E-3m , which is still significantly Design Variables [mm] smaller4 than that of the conventional one (0.295E-3m3). It is noteworthy that at the optimum the applied current is 0 1.77A. From the above, it is remarked that in order to compromise between the5 size10 and15 20 25 30 Iteration power consumption of the MRC, the power consumptionDevelopment should of a novel be MR conserved clutch featuring as tooth-shaped a constraint disc 271 function and the applied current should be treated as a design variable. (a) design variables ) 3 15 20 m R 0.1xR t t Overall Volume di do d h -4 3.6 Transmitting Torque ) hc wc tes tec 16 x10 12 Nm 3.0 12 2.4 9 8 Design Variables [mm] 1.8 4 6 Transmitted torque ( Overall volume of MRC ( 1.2 0 3 5 10 15 20 25 30 5 10 15 20 25 30 Iteration Iteration (a) Design variables (b) Overall volume and transmitted torque (a) design variables (b) overall volume and transmitted torque ) 3 15 m Overall Volume -4 3.6 Transmitting Torque ) x10 12 Nm 3.0 2.4 9 1.8 6 Transmitted torque ( 8 Overall volume of MRC ( 1.2 3 5 10 15 20 25 30 Iteration (c) Magnetic density at the optimum (b) overall volume and transmitted torque Fig. 5. Optimization solution of the conventional MRC for this significant volume reduction come from the larger contact area between the rotor and the working MRF and the higher magnetic density across the MRF ducts of the pro- posed MRC compared to those of the conventional one as shown are shown in Figs. 5(c) and 6(c). The optimal solution is summarized in Table1. From the table it is also noted that the mass of the proposed MRC is also much smaller than that of the conventional MRC, however, the power consumption of the proposed MRC is almost twice as high as that of the conventional8 one. This is obvious because in the proposed MRC, two magnetic coils are employed (one on each side) and the power consumption is not considered in the optimization of the MRC. Thus, there is a trade-off between the size and the power con- sumption of the MRCs. Therefore, in the application where high power is available ant not significant, the proposed MRC can be employed. In order to compromise between the size and the power consumption of the proposed MRC, the power consumption should be taken into account as a constraint function in the optimization. Fig.7 show the opti- mal solution of the proposed MRC with the required transmitted torque is 10 Nm and the power consumption is constrained to be smaller than that of the conventional one
8. (c) magnetic density at the optimum (c) magnetic density at the optimum Figure272 5. Optimization Quoc Hung Nguyen, solution Bao Tri of Diep, the Duy conventional Hung Nguyen, VanMRC Bien Nguyen, Van Bo Vu, Qui Duyen Do Figure 5. Optimization solution of the conventional MRC 55 8 hc 55 tc Ri th 8 h ltp hc tc Ri th h ltp Ro R L ltb lto td 44 R R L l l t 44 o 6 tb to d 6 33 33 4 4 22 22 Design Variables [mm] Design Variables [mm] 2 11 Design Variables [mm] Design Variables [mm] 2 11 0 0 0 0 5 10 15 20 5 25 10 30 15 20 525 1030 15 20 5 2510 3015 20 25 30 Iteration Iteration Iteration Iteration (a) design variables (a)(a) design Design variables ) ) 3 3.5 3 14 m 3.5 14 Overallm Volume Overall Volume -4 -4 ) Transmitting Torque Transmitting Torque ) x10 3.0 x10 3.0 12 12 Nm Nm 2.5 2.5 10 10 2.0 2.0 8 8 1.5 1.5 6 6 Transmitted torque ( Transmitted torque ( Overall volume of MRC ( Overall volume of MRC ( 1.0 1.0 4 4 5 10 15 20 525 1030 15 20 25 30 Iteration Iteration (b)(b) overall Overall volume volume and and(b) transmitted transmittedoverall volume torque torque and transmitted(c) Magnetic torque density at the optimum Fig. 6. Optimization solution of the proposed MRC ) 60 3 Overall Volume m 5 -4 Transmitted torque (Nm) 50 x10 Power consumption (W) 4 40 30 3 20 2 10 9 Overall volume of MRC ( 1 0 9 5 10 15 20 25 30 35 40 Iteration Transmitted torque & Power consumption Figure 7. OptimizationFig. 7. Optimization solution solution of the of proposed the proposed MRC MRC with with two two constraints: constraints: transmitted transmitted torque ≥ 10 Nm, power consumption ≤ 44 W torque 10Nm, power consumption 44W In order to investigate performance of the proposed MRC at different maximum transmitted torques, the optimization of the MRCs at different values of the required transmitted torque are conducted and the results are shown in Figure 8. As shown in Figure 8a, the overall volume of the MRCs increases with the required transmitted torque. The overall volume of the proposed MRC is significantly smaller than that of the conventional one with a volume ratio ranging from 60% at the required torque of 5Nm to 45% at the required torque of 100Nm. In contrary, the power consumption of the proposed MRC is almost as twice as that of the proposed one at different value of the required torque as shown in Figure 8b. ) 3 m -4 20 70 300 18 Tooth shaped Conventional [9] Tooth-shaped MRC Conventional MRC [9] 16 Volume ratio 240 60 14 12 180 10 50 8 120 6 40 4 Volume ratio (%) 60 2 Power Consuming (W) 0 30 0 0 20 40 60 80 100 0 20 40 60 80 100 Overall volume of the MRCs (x10 Required transmitted torque (Nm) Required transmitted torque (Nm) (a) overall volume b) power consumption Figure 8. Optimal results of the MRCs as different values of the required transmitted torque 4. Experimental Validation In this part, the above optimized MRC is fabricated and its performance characteristics are tested by experimental works. Figure 9 shows experiment apparatus to evaluate transmitted torque of the MRC. The 500W-DC servo motor with gearbox is controlled by a computer to constantly rotate at 300 rpm, the output shaft of the motor is connected to the input shaft of 11
9. Development of a novel MR clutch featuring tooth-shaped disc 273 (44W). It is noted that in this case the applied current is also considered as a design vari- able. From the figure it is observed that, with the above transmitted torque and power consumption, the overall volume of the proposed MRC is 0.166E-3 m3, which is still sig- nificantly smaller than that of the conventional one (0.295E-3 m3). It is noteworthy that at the optimum the applied current is 1.77A. From the above, it is remarked that in order to compromise between the size and power consumption of the MRC, the power con- sumption should be conserved as a constraint function and the applied current should be treated as a design variable. Table 1. Significant parameters of the optimized MRCs MRC types Design parameter (mm) Characteristics (at I = 2.5A) Conventional Coil: width tc = 10.6; height hc = 5.75; Max. Torque: 10 Nm No. turns: 238 Overall Volume (m3): 0.295E-3 Housing thickness: side th = 3, cylin- Off-state Torque: 0.3 Nm drical to = 3 Mass (kg): 2.3 Fixed Housing: side t f h = 5.9, cylin- Power Cons.: 44 W drical t f o = 5.3, overall size R = Coil Resistance(Ω): Rc = 7.0 63.8, L = 23.0 Disc: inner radius Rdi = 16.0, outer ra- dius Rdo = 48.9; thickness td = 3 MRF duct gap: 0.8 Proposed Coil: width tc = 7.7; height hc = 21.6; Max. Torque: 10 Nm No. of turns: 2*500 Overall Volume (m3): 0.159E-3 Fixed Housing: thickness th = 4.5, Mass: 1.24 kg overall size R = 38.4, L = 34.5 Off-state Torque: 0.293 Nm Disc: Ri = 8.0, Rd = 36.4, td = Power Cons.: 85 W 2.0, ht = 2.0, ltp = 3.9, ltb = 4.5, lto = Coil Resistance(Ω): Rc = 6.8 2.5, φ = 1.07 rad MRF duct gap: 0.8 In order to investigate performance of the proposed MRC at different maximum transmitted torques, the optimization of the MRCs at different values of the required transmitted torque are conducted and the results are shown in Fig.8. As shown in Fig. 8(a), the overall volume of the MRCs increases with the required transmitted torque. The overall volume of the proposed MRC is significantly smaller than that of the conven- tional one with a volume ratio ranging from 60% at the required torque of 5 Nm to 45% at the required torque of 100 Nm. In contrary, the power consumption of the proposed MRC is almost as twice as that of the proposed one at different value of the required torque as shown in Fig. 8(b).
10. ) ) 60 60 3 3 OverallOverall Volume Volume m m 5 5 -4 -4 TransmittedTransmitted torque torque (Nm) (Nm) 50 50 x10 x10 PowerPower consumption consumption (W) (W) 4 4 40 40 30 30 3 3 20 20 2 2 10 10 Overall volume of MRC ( 1 Overall volume of MRC ( 1 0 0 5 510 1015 1520 2025 2530 3035 3540 40 Transmitted torque & Power consumption IterationIteration Transmitted torque & Power consumption FigureFigure 7. Optimization7. Optimization solution solution of theof the proposed proposed MRC MRC with with two two constraints: constraints: transmitted transmitted torquetorque 10 Nm10Nm, power, power consumption consumption 44 W44 W In In order order to to investigate investigate performance performance of of the the proposed proposed MRC MRC at at different different maximum maximum transmittedtransmitted torques, torques, the the optimization optimization of of the the MRCs MRCs at at different different values values of of the the required required transmittedtransmitted torque torque are are conducted conducted and and the the results results are are shown shown in Figurein Figure 8. As8. Asshown shown in Figurein Figure 8a, 8a, thethe overall overall volume volume of ofthe the MRCs MRCs increases increases with with the the required required transmitted transmitted torque. torque. The The overall overall volumevolume of theof the proposed proposed MRC MRC is significantlyis significantly smal smaller lerthan than that that of theof the conventional conventional one one with with a a volumevolume ratio ratio ranging ranging from from 60% 60% at theat the required required torque torque of 5ofNm 5Nm to 45%to 45% at theat the required required torque torque of of 100100Nm274Nm. In. contrary,In Quoccontrary, Hung the Nguyen, the power power Bao Triconsumption Diep, consumption Duy Hung Nguyen,of theof Vanthe proposed Bien proposed Nguyen, MRC Van MRC Bo Vu,is Quialmostis Duyenalmost as Do twiceas twice as thatas that of theof the proposed proposed one one at differentat different value value of theof the required required torque torque as shownas shown in Figurein Figure 8b .8 b. ) ) 3 3 m m -4 20-4 20 70 70 300 300 18 18 Tooth Tooth shaped shaped Conventional Conventional [9] [9] Tooth-shaped Tooth-shaped MRC MRC Conventional Conventional MRC MRC [9] [9] 16 16 Volume Volume ratio ratio 240 240 60 60 14 14 12 12 180 180 10 10 50 50 8 8 120 120 6 6 40 40 Volume ratio (%) 4 4 Volume ratio (%) 60 60 Power Consuming (W) 2 2 Power Consuming (W) 0 0 30 30 0 0 0 0 20 20 40 40 60 60 80 80 100 100 0 0 20 20 40 40 60 60 80 80 100 100 Overall volume of the MRCs (x10 Overall volume of the MRCs (x10 RequiredRequired transmitted transmitted torque torque (Nm) (Nm) RequiredRequired transmitted transmitted torque torque (Nm) (Nm) (a) (a)(a)overall Overalloverall volume volumevolume b)(b) powerb) Power power consumption consumption consumption Fig. 8. Optimal results of the MRCs as different values of the required transmitted torque Figure Figure 8. Optimal 8. Optimal results results of theof theMR MRCs Cass differentas different values values of theof therequired required transmitted transmitted torque torque 4. EXPERIMENTAL VALIDATION In this part, the above optimized MRC is fabricated and its performance character- 4. 4Experimental.isticsExperimental are tested byValidation Validation experimental works. Fig.9 shows experiment apparatus to evaluate Intransmitted thisIn this part, part, the torque the above above of theoptimized optimized MRC. The MRC MRC 500W-DC is fabricatedis fabricated servo motorand and its with itsperformance performance gearbox is characteristics controlled characteristics by are are thea computer MRC while to constantly the output rotate shaft ofat 300 the MRCrpm, the is connected output shaft to an of thestationary motor torque is connected sensor. testedtestedto by the byexperimental input experimental shaft of works the works MRC. Figure. whileFigure 9 the shows9 outputshows experiment shaftexperiment of the apparatus MRCapparatus is connectedto evaluateto evaluate to transmitted an transmitted sta- When the experimental process is commenced, a step current with different magnitude (0.5A, torquetorquetionary of theof torquethe MRC. MRC. sensor. The The 500W When 500W-DC the-DC experimentalservo servo motor motor processwith with gearbox isgearbox commenced, is controlledis controlled a step by current bya computer a computer with to to 0.75A,different 1.0A, magnitude 1.25A, 1.5A,1.75A, (0.5A, 0.75A, 2.0A, 1.0A, 2.5A) 1.25A, is applied 1.5A, 1.75A, to the 2.0A,two coil 2.5A)s of isthe applied MRC and to the the constantly rotate at 300 rpm, the output shaft of the motor is connected to the input shaft of constantlytransmittedtwo coils rotate of torque the MRCat is 300 measured and rpm the, bythe transmitted the output torque shaft torquesensor. of isItthe measuredis motornoted that is by connectedthe the torque to sensor. the input shaft of 11 11 FigFig.ure 9 .9. Experiment Experiment apparatus apparatus to to test test the the proposed proposed MRC MRC Fig. 10 show the results obtained from experiments. From Fig. 10(a), it is observed Figurethat the 10 measured show the transmittedresults obtained torque from at experiments. the applied currentFrom Figure of 2.5A 10 isa, aroundit is observed 10.4 Nm, that which is a bit greater than that of the simulation (9.9 Nm). The main reason may come the measured transmitted torque at the applied current of 2.5 A is around 10.4Nm, which is a bit greater than that of the simulation (9.9 Nm). The main reason may come from the neglecting of the friction torque of the sealing and bearings in simulated results. It is also observed from Figure 10b that at different applied currents, different values of output torque can be archived and very closed to the simulated ones; the error is within 5%. It is also found the torque response time is around 0.25s, which is very good time response for application in industry. . 12 12 2.5A 2.0A 1.75A 1.5A Measured Simulated 1.25A 1.0A 0.75A 0.5A 10 10 8 8 6 6 4 4 Output Torque (Nm) Output Torque (Nm) Output 2 2 0 0 0 1 2 3 4 0.0 0.5 1.0 1.5 2.0 2.5 Time (s) Applied Current (A) (a) measured torque (b) simulated vs. measured torque Figure 10. Experiment results of step response of the MRC 5. Conclusions In this study, a novel configuration of magneto-rheological clutch (MRC) featuring a tooth shaped disc was proposed. The inner face of the clutch housing also has tooth shaped features mating with the teeth of the disc via MRF layer. Excitation magnetic coils are assembled on stationary winding cores placed on both side of the clutch housing. An air gap of 0.3mm is left between the housing and the winding cores to ensure the housing can freely rotate against 12
11. thethe MRC MRC while while the the output output shaft shaft of of the the MRC MRC is is connected connected to to an an stationary stationary torque torque sensor. sensor. WhenWhen the the experimental experimental process process is is commenced commenced, a, astep step current current with with different different magnitude magnitude (0.5A, (0.5A, 0.75A,0.75A, 1.0A, 1.0A, 1.25A, 1.25A, 1.5A,1.75A, 1.5A,1.75A, 2.0A, 2.0A, 2.5A) 2.5A) is is applied applied to to the the two two coil coils sof of the the MRC MRC and and the the transmittedtransmitted torque torque is is measured measured by by the the torque torque sensor. sensor. It Itis is noted noted that that the the FigFigureure 9 .9 Experiment. Experiment apparatus apparatus to to test test the the proposed proposed MRC MRC FigureFigure 10 10 show show the the results results obtained obtained from from experiments. experiments. From From Figure Figure 10 10a,a it, itis is observed observed that that thethe measured measured transmitted transmitted torque torque at at the the applied applied current current of of 2.5 2.5 A A is is around around 10.4 10.4NmNm, which, which is is a a bitbit greater greater tha than n that that of of the the simulation simulation (9.9 (9.9 Nm Nm). ). The The main main reason reason may may come come from from the the neglectingneglecting of of the the friction friction torque torqueDevelopment of of the of the a novel sealing sealing MR clutch and featuring and bearings tooth-shaped bearings disc in in simulated simulated results results 275 . It. It is is also also observedobserved from fromfrom Figure theFigure neglecting 10 10b bthat ofthat the at frictionat different different torque applied ofapplied the sealing currents, currents, and bearings different different in simulated values values results. of of output output torque torque It is also observed from Fig. 10(b) that at different applied currents, different values of cancan be be archived archivedoutput and and torque very very can closed beclosed archived to to the andthe simulated verysimulated closed ones; to ones; the simulatedthe the error error ones; is is within thewithin error 5%. is5%. within It Iist is also also found found the torque response5%. It is alsotime found is around the torque 0.25s, response which time is is aroundvery good 0.25 s, time which response is very good for time application in the torque response for time application is around in industry. 0.25s, which is very good time response for application in industry.industry. . . 2.5A 2.0A 1.75A 1.5A 1212 1212 2.5A 2.0A 1.75A 1.5A Measured Measured Simulated Simulated 1.25A1.25A 1.0A 1.0A 0.75A 0.75A 0.5A 0.5A 1010 1010 8 8 8 8 6 6 6 6 4 4 4 4 Output Torque (Nm) Output Torque (Nm) Output Torque (Nm) Output 2Torque (Nm) Output 2 2 2 0 0 0 0 0 0 1 1 2 2 3 3 4 4 0.00.0 0.50.5 1.01.0 1.51.5 2.02.0 2.52.5 TimeTime (s) (s) AppliedApplied Current Current (A) (A) (a)(a) measured measured(a) torque Measured torque torque (b) (b) simulated simulated(b) Simulated vs. vs.vs. measured measured torque torque torque FigFigureure 10 10. Experiment. Experiment results results of of step step response response of of the the MRC MRC Fig. 10. Experiment results of step response of the MRC 55. .ConclusionConclusionss 5. CONCLUSIONS In this study, a novel configuration of magneto-rheological clutch (MRC) featuring a tooth shaped disc was proposed. The inner face of the clutch housing also has tooth InIn this this study, study,shaped a featuresanovel novel configuration mating configuration with the teeth of of magneto of magneto the disc- viarheological-rheological MRF layer. Excitationclutch clutch (MRC) magnetic(MRC) featuring coilsfeaturing a atooth tooth shapedshaped disc disc wasare was assembled proposed proposed on. stationaryThe. The inner inner winding face face coresof of the placedthe clutch clutch on both housing housing side of also the also clutch has has housing.tooth tooth shaped Anshaped features features matingmating with with airthe the gap teeth ofteeth 0.3 of mm of the isthe left disc disc between via via MRF the MRF housing layer. layer. and Excitation theExcitation winding coresmagnetic magnetic to ensure co co theilsils housingare are assembled assembled on on can freely rotate against the winding cores. Optimal design of the proposed MRC consid- stationarystationary winding windingering transmitted cores cores placed torqueplaced andon on overallboth both side volume side of of ofthe thethe clutch MRC clutch was housing. conductedhousing. An andAn air comparedair gap gap of of 0. 0.3mm3mm is is leftleft between between withthe the housing traditional housing and MRC and the featuringthe winding winding plain cores cylindricalcores to to ensure disc.ensure The the the optimal housing housing results can can showed freely freely that rotate rotate against against at the same required transmitted torque, the overall volume and mass of the proposed MRC are significantly smaller than those of the conventional one (almost half). However, the power consumption of the proposed MRC1212 is around double of the conventional one because two coils are implemented. In order to compromise between the size and the power consumption of the proposed MRC, the optimization with two constraints (trans- mitted torque and power consumption) and was conducted. The results showed that with the power consumption is constrained to be smaller than the power consumption of the conventional one, the overall mass of the MRC is still significantly smaller than that of the conventional one. The optimized MRC was then manufactured for testing. Exper- imental works showed a good agreement between the measured transmitted torque and the simulated one.
12. 276 Quoc Hung Nguyen, Bao Tri Diep, Duy Hung Nguyen, Van Bien Nguyen, Van Bo Vu, Qui Duyen Do ACKNOWLEDGEMENT This work was supported by the Vietnam National Foundation for Science and Tech- nology Development (NAFOSTED) under grant number 107.01-2018.335. REFERENCES [1] J. Wang and G. Meng. Magnetorheological fluid devices: Principles, characteristics and applications in mechanical engineering. Proceedings of the Institution of Mechanical En- gineers, Part L: Journal of Materials: Design and Applications, 215, (2001), pp. 165–174. [2] A. Muhammad, X. liang Yao, and Z. chao Deng. Review of magnetorheological (MR) fluids and its applications in vibration control. Journal of Marine Science and Application, 5, (2006), pp. 17–29. [3] T. D. Truong, V. Q. Nguyen, B. T. Diep, D. H. Q. Le, D. T. Le, and Q. H. Nguyen. Speed control of rotary shaft at different loading torque using MR clutch. In A. Er- turk, editor, Active and Passive Smart Structures and Integrated Systems XIII, SPIE, (2019), [4] U. Lee, D. Kim, N. Hur, and D. Jeon. Design Analysis and Experimental Evaluation of an MR Fluid Clutch. Journal of Intelligent Material Systems and Structures, 10, (1999), pp. 701–707. [5] T. Kikuchi, K. Ikeda, K. Otsuki, T. Kakehashi, and J. Furusho. Compact MR fluid clutch device for human-friendly actuator. Journal of Physics: Conference Series, 149, (2009). [6] D. Wang and Y. Hou. Design and experimental evaluation of a multidisk magnetorheological fluid actuator. Journal of Intelligent Material Systems and Structures, 24, (2012), pp. 640–650. [7] K. H. Latha, P. U. Sri, and N. Seetharamaiah. Design and Manufacturing Aspects of Magneto- rheological Fluid (MRF) Clutch. Materials Today: Proceedings, 4, (2), (2017), pp. 1525–1534. [8] Q. H. Nguyen and S.-B. Choi. A new method for speed control of a DC motor using magne- torheological clutch. In Active and Passive Smart Structures and Integrated Systems 2014, Inter- national Society for Optics and Photonics, (2014), [9] Q. H. Nguyen, N. D. Nguyen, and S. B. Choi. Design and evaluation of a novel magnetorheo- logical brake with coils placed on the side housings. Smart Materials and Structures, 24, (2015). [10] N. D. Nguyen, T. Le-Duc, L. D. Hiep, and Q. H. Nguyen. Development of a new magnetorheological fluid–based brake with multiple coils placed on the side hous- ings. Journal of Intelligent Material Systems and Structures, 30, (2018), pp. 734–748.