Model validation of the output reciprocating dynamic responses of a twin electrorheological (ER) clutch mechanism

1、model validation of the output reciprocating dynamic responses of a twin electro-rheological (er) clutch mechanismk.p. tan *, a.r. johnson, r. stanway, w.a. bulloughdepartment of mechanical engineering, university of sheffield, mappin building, mappin street, ukabstractelectro-rheological (er) fluid

2、 devices are becoming more popular in the industrial applications. this is due to the fast speed of response and large output dynamics of the er actuators. the usefulness of this er dynamic response is considered in the material winding processes where fast output bi-directional responses are essent

3、ial. therefore in the present paper, an er twin clutch mechanism is proposed. this clutch mechanism consists of two identical clutches that rotate in opposite directions. but the bi-directional output dynamics of the clutch mechanism is not well understood due to its non-validation in the past. the

4、main aim of this paper is to model the reciprocating responses of the clutch mechanism and then perform model validation with the measured test results. the close agreements between the modeled and experimental data indicate that the er output angular velocity and displacement responses models of th

5、e clutch mechanism are validated. these validated models can then be used to predict accurately the reciprocating output responses of the twin er clutch mechanism for future research studies. 2007 elsevier ltd. all rights reserved.keywords: electro-rheological; twin clutch mechanism; bi-directional;

6、 model validation; er angular velocity; displacement responses1. introductionindustrial applications such as the yarn production in the textile industry and the winding of filaments onto bobbins normally require actuators that can provide fast reciprocating speeds of responses and large output torqu

7、es. to resolve this issue, various devices are considered. the present servomotors are unsuitable in this application due to their cogging torque responses and large output inertia 16. next, the smart fluid actuators are considered. by using the electro-rheological (er) fluid as the working medium,

8、an er clutch can be a potential rotary device. this analysis is obtained by considering the following research investigations. in the past, carlson et al. 7, firoozian et al. 8 and tan et al. 9 tested the er clutch and they reported the er fast speed of response and huge dynamic responses. next, joh

9、nson et al. 10 and bullough et al. 11 employed inertia loadings at the output shaft of the er clutch. the pick-up and drop-load velocity responses delivered. by the er clutch were significantly fast in terms of tens of milliseconds. these fascinating er dynamics motivated brookfield and dlodlo 12,13

10、. they employed a proportional-integral-derivative (pid) controller to activate the er clutch for driving a robot arm to reach a robotic positioning error of 0.75. a limitation of their er application is that the inertia arm was driven at fast speed of response in one direction. therefore to achieve

11、 fast bi-directional responses, the following er research works are considered.in order to actuate the inertia loads in both directions, choi et al. 14 used an er clutch-brake mechanism to achieve start-stop motion responses. although this clutch-brake mechanism could perform bi-directional response

12、s, but the turn-round dynamics or reciprocating dynamics were slow. therefore for the purpose of actuating the output inertia at fast speed of response in the forward and reversal motions, the authors studied a twin er clutch mechanism 15-17. this clutch mechanism consists of two identical clutches

13、that are driven by their respective electric motors to rotate in different directions. in the working process, one of the energized rotary er clutches actuates the output inertias to travel in one direction when the other clutch is non-energized. to generate reciprocating responses on the inertias,

14、both the clutches are activated alternately by the electric fields. but, there is no detailed understanding of the bi-directional response of this clutch mechanism particularly on the er speeds of responses or turn-round dynamics of the inertias. therefore in this present paper, two mathematical mod

15、els are derived to understand the output angular velocity and displacement responses of the reciprocating clutch mechanism. to prove its validity, these kinematic models of the clutch mechanism undergo validations with the experimental er output angular velocity and displacement responses. this vali

16、dation work is necessary due to it can provide key knowledge on the reciprocating dynamics of the smart actuators. next, the authors describe the organization of this paper as follows.section 2 describes the test facility of the clutch mechanism. section 3 illustrates the derivations of the mathemat

17、ical models for the er output angular velocity and displacement behaviors. section 4 validates these models by using the measured kinematic responses of the er clutch mechanism. section 5 is the respective conclusion. appendix 1 tabulates the measured results of the er torque and passive fluid visco

18、sities.2. the reciprocating twin er clutch mechanismfig. 1 is the overall mechanical assembly of the er clutch mechanism. in this clutch mechanism, two identical co-axial er clutches are arranged vertically. the detailed assembly of each er clutch is presented in fig. 2. section 2.1 describes the me

19、chanical assembly of the twin clutch mechanism. section 2.2 presents the electrical facility that is used to energize the er fluid clutches alternately. section 2.3 is a detailed description of the er reciprocating process in the clutch mechanism.2.1. description of the mechanical arrangement of the

20、 twin er clutch mechanismin fig. 2, the two fundamental components of the co-axial er clutch are its input rotor (1) and output rotor (2). the former rotor is the ground potential electrode and the latter rotor is the high voltage electrode. both these rotors are made of stainless steel. the input r

21、otor has outer and inner diameters of 52 mm and46 mm, respectively. the er active external and internal diameters of the output rotor are 51mm and47 mm, respectively. when coupled together, both these rotors have er active lengths of 30 mm across two inter-electrode gap sizes of 0.5 mm each. due to

22、the embedded er fluid between both the electrode gaps, the viscous active inner and outer lengths of the output rotors are 30 and 40 mm, respectively. after the descriptions of both the rotors, the mechanical assembly of the twin er clutch mechanism is stated below.the output rotor of the er clutch

23、is assembled to a mild steel shaft (3) by means of two ball bearings. this shaft is then engaged to the input rotor co-axially using a key and keyway. this forms a sub-assembly of the er clutch. next, each of the two sub-assemblies is coupled tightly to a mild steel column by employing two ball bear

24、ings and a locknut. each er clutch-column assembly is fitted rigidly to a high voltage isolator (4). the latter is placed between an ac motor and the clutch-column assembly. this isolator serves to protect the electrical circuitry of the ac motor from the effects of the high voltages that are applie

25、d to the clutch. two isolators are placed below the twin clutch-column assemblies. the ac motor is described as follows.the employed motor is a 3 phases, ac induction type (electrodrives limited, serial number: bh114203). for this motor, it can deliver a maximum mechanical power and an angular speed

26、 of 0.25 kw and 2800 rpm, respectively. since each of the er clutches is driven by the output shaft of the motor (5), two identical ac induction motors are employed. they are arranged in vertical positions with a horizontal distance of 0.175 m apart as shown in fig. 1. by using a manual speed regula

27、tor, the motors are controlled to spin at similar output rotational speeds but in the opposite angular directions. this implies that the left and right motors undergo clockwise and anti-clockwise rotations respectively. in order to integrate these motor dynamics into the overall performances of the

28、clutches, the following tasks are performed.by using its output shaft (5) and an oldham coupling (6), each of these motors transmits its output dynamics to the respective coupled clutch. however prior to the clutch rotation, a nylon cover and an o-ring rubber seal (7) are used to seal the er fluid (

29、8) securely within each clutch compartment. then, a nylon, spur type, gear wheel (9) of 80 mm pitch circle diameter and 125 external gear teeth is tightened rigidly to the output rotor of each er clutch. for the sub-assembly of each clutch, the total output moment of inertia is 3.83 x 10 kgm inclusi

30、ve of its attached gear wheel. next, the twin clutch-column assemblies are coupled with the two motors. finally, a toothed timing belt (13) of 16 g mass and 630 mm in length is engaged with the external gear teeth of the two gear wheels. the gear wheels and timing belt are used to integrate the rota

31、tional dynamics of the output rotors of both the er clutches. next in order to achieve reciprocating output responses of this clutch mechanism, the following facilities are described.2.2. description of the electrical facilities for the er reciprocating responsesfor the purpose of activating the er

32、clutches electrically, three sets of connections are required. initially, two carbon brushes are used to contact the output rotors of the er clutches. then, two output probes from the eht switch boxes are fitted firmly to the carbon brushes. this is to link the high voltage supply to the clutches. t

33、he source of these probes originates from the electrical circuitry as shown in fig. 3. next, two carbon probes contact the external surfaces of the two input rotors to form the earth connection. thus, the electrical path of the twin clutch mechanism is completed. in fig. 3, the detailed descriptions

34、 of the ksm power supply, farnell function generator, voltage and current measurement units, lecroy oscilloscope, farnell power supply and eht switch boxes 18,19 had been stated previously. similar descriptions had been given for the tested er fluids of lipol l30155, l40155 and bayer ai3565 50% volu

35、me fraction fluids. their volume fractions are 30%, 40% and 50% respectively. the temperature ranges of these er fluids are between 0 c and 80 c. the largest electric field strength inputted to all these er fluids is 5 kv/mm. these er fluids (8) fill the er clutch compartment as shown in fig. 2. in

36、order to activate the fluid clutches with the electrical facilities of fig. 3, the output high voltage waveform of fig. 4 is employed.the usage of the above bi-polar voltages is to fulfill two main aims. the first aim is to energize one er clutch and de-activate the other clutch in order to achieve

37、belt reciprocating response. however, the main objective of switching the polarity differences is to prevent the occurrence of electrophoresis of the tested er fluid. this electrophoresis can breakdown the er effect and it produces inaccurate output dynamic results 19. by applying the above bi-polar

38、 voltages to the twin er clutch mechanism in fig. 1, the belt traverses with linear reciprocating motion. this er reciprocating output dynamics are next described as follows.2.3. description of the er reciprocating process in the twin clutch mechanismin this section, the er reciprocating dynamics ar

39、e described by means of the key positions of the reciprocating belt as shown in fig. 5. the mid and two extreme positions on the belt are points 0/5, 3 and 8, respectively. these two end points are produced due to the fast speeds of response of the er clutches during their individual bi-polar, on-of

40、f energisations and passive slippages. in order to understand the belts dynamics from points 0 to 9, table 1 shows the behavior of the output angular velocity responses of the two er clutches, states of energisations and slippages at different belt positions. the term energisation is defined as the

41、application of high voltages to activate the er fluid in a particular er clutch. the term slippages of the input and output rotors occur when the er clutch is non-er active.the detailed descriptions of table 1 are stated as follows. by influencing the rotational er clutches a and b with consistent z

42、ero and high voltages, the rotors of the former and latter devices are disengaged and engaged, respectively. these occurrences are due to the presences of fluid slippages and bingham-plastic fluid within the respective clutches a and b. using this bingham-plastic fluid, the energized clutch b output

43、s larger torque than clutch a by means of the er torque. using its larger output torque dynamics, the clutch bs output rotor spins in the anti-clockwise direction at a constant angular velocity, q. this velocity is delivered to the output shaft of the de-energized clutch a when the belt traverses li

44、nearly from points 0 to 1.in order to reverse the direction of belt motion, the er clutch a is suddenly energized and the er clutch b is de-activated electrically. however, the output angular velocity of the clutch a remains unchanged due to the presence of an electron-hydraulic time period. this ti

45、me delay causes the absences of the output er torque responses from both the clutches. in previous studies, this electron-hydraulic time delay 8,9 is virtually negligible at the consideration of the output er velocity response 10,11,20. therefore, the fluid slippages occur at the output viscous geom

46、etries of both the disengaged clutches. since the generated viscous dynamic responses are insufficient to alter the output angular momentums of both the er clutches, its output rotors maintain at the constant angular velocity of q. as a result of these occurrences, the belt translates from points 1

47、to 2.after the electron-hydraulic time period is fulfilled, the disengaged clutch b experiences the fluid slippages between its rotors and thus, it outputs viscous torque response. for the energized clutch a, both its rotors are coupled together and this er activity leads to the production of both t

48、he er torque 7,15-17 and viscous torque. due to the net output torque response of both the clutches, the er clutch a delivers larger dynamics than clutch b. using its er torque response to overcome the angular momentum, the clutch as output shaft produces an er kinematics of turn-round motion. in ot

49、her words, the clutch as output rotor decelerates from q to a halt and then accelerates from zero angular velocity to +q. this er kinematics dominates the overall output response of the clutch mechanism when the output angular velocity response of the er clutch a is delivered to the output rotor of

50、the clutch b via the belt. however in the turn-round dynamics, the fluid slippages of both the clutches co-exist with the er torque due to the output velocity response of the clutch a has not attained the value of its input driving speed. this er kinematics is highlighted as the belt reversal from p

51、oints 2 to 3 and then from points 3 to 4, as seen in fig. 5.at the belt position of point 4, the turn-round motion of the energized er clutch a is completed when its output shaft rotates at a constant angular velocity of q in the clockwise direction. this kinematic response is due to the firm engage

52、ment between the rotors of the clutch a and the fluid slippage in the clutch b. therefore, the belt travels linearly from points 4 to 6. to alter this belt kinematics, the applied voltages of the clutches a and b are reversed. however, the de-energized clutch a maintains the output angular velocity

53、of x during the electron-hydraulic time delay. this kinematics behavior is observed by the belt positions from points 6 to 7 and it is due to the viscous slippages outputted by both the er clutches.when the er activity of the er clutch b is completed, the net torque response between the energized cl

54、utch b and the viscous clutch a is the er torque. after overcoming the angular momentum of the er clutches output rotors, the belt and the viscous torque, the er torque of the clutch b causes its output rotor to experience the dynamics of rapid reciprocating responses. the output rotor of the clutch

55、 b decelerates from the angular velocity of x to zero using points 7 to 8 and then, from zero to x as shown by the points 8 to 9. from the belt points 7 to 9, both the clutches undergo slippages. by points 9 to 0, the output velocity of the clutch b is equivalent to its input angular speed and this

56、kinematics is delivered to the output rotor of the clutch a that has viscous slippages. these dynamic responses of the clutch b are transferred to the output rotor of the clutch a via the belt. the repeatability of this reciprocating er kinematics is available provided that both the clutches receive

57、 continual inputs of the bipolar, on-off voltages and rotations of their respective motors. the above descriptions of the er output kinematic responses of the reciprocating clutch mechanism can be formulated into mathematical models as follows.3. mathematical derivations of the er output displacemen

58、t and velocity responses of the er cluchesin this section, the output angular displacement, velocity and acceleration responses of the er clutches a and b 15-17 are analyzed mathematically. the output kinematics of these clutches are equivalent due to their output rotors are linked together by using

59、 two attached gear wheels and a toothed timing belt in the belt reciprocating process. initially by using fig. 6, the modeling of these er output kinematics are conducted according to newtons second law of motion. therefore, a preliminary dynamic equation is written below where t is the output torque of the er clutch, / the total moment of inertia influencing the belt traversing motion and a is the change in output angular momentum of both clutches.the output