Macroslip detection and clamping force control for a metal V-belt continuously variable transmission

Abstract

In this paper, a macroslip detection method is proposed for a metal V-belt continuously variable transmission, and a clamping force control strategy is suggested on the basis of a macroslip detection method. Using the rotational accelerations of the primary and secondary pulleys, the velocity of the secondary pulley and the speed ratio of the continuously variable transmission variator, observation signals are defined. The characteristics of the observation signals are investigated by simulations in a vehicle-driving environment. It is found that two observation signals became unsynchronized when macroslip occurs. Considering the oscillations of the acceleration signal and the noise from the sensors or the differentiating process, a signal-processing method for the observation signals is suggested. Based on the signal-processing results, a variable called the ‘amplitude difference rate’ is introduced for slip evaluation. A macroslip detection method that uses the amplitude difference rate is proposed. The effectiveness of the macroslip detection method is validated by experiments. It is found from the experiments that the proposed method can effectively detect macroslip with an acceptable time delay. In addition, a clamping force control strategy based on the macroslip detection method is developed. In this strategy, the clamping force is maintained at a minimum value with a safety factor of 1 in normal driving conditions but an additional clamping force is applied when macroslip is detected. It is found from the simulation results that macroslip is eliminated by clamping force control. It is expected that the efficiency of the continuously variable transmission system can be improved by reducing the marginal clamping force using the proposed macroslip detection method and the clamping force control strategy.

Macroslip detection for a metal V-belt continuously variable transmission using the period difference

“…The parameter U and the sampling time are selected as U = 3 and Dt = 0.01 s respectively. It is seen from Figure 9 that the PD stays below the threshold value b until t = 0.87 s and then becomes larger than b, which means that macroslip occurs according to the macroslip detection criterion in equation (12). When the occurrence time for macroslip is compared with that in Figure 6, it is noted that a time delay of 0.03 s exists for the detection of macroslip due to the signal processing.…”

“…It is noted that the primary pulley's accelerations and the secondary pulley's accelerations show similar oscillation periods in the microslip region and become out of synchronization in the macroslip region (Figure 11(c)). The IP (Figure 11(d)) of the primary pulley's acceleration and the IP of the secondary pulley's acceleration have almost the same values until t = 0.87 s for Dt = 0.01 s. Using the PD, macroslip is detected at t = 0.91 s from the macroslip detection criterion in equation (12) for U = 3 (Figure 11(e)). A time delay of 0.06 s exists for macroslip detection because of the signal processing.…”

“…3,7 To overcome these drawbacks, several methods without using additional sensors have been introduced. [8][9][10][11][12] In these methods, fluctuations were applied to the torques or the clamping forces, which generally originate from the road load, the engine and the hydraulic system. One method 8,9 identified the slip state of the CVT variator by the normalized slip state G n , which is defined on the basis of the power spectra of the fluctuations in the rotational speeds of the primary pulley and the secondary pulley.…”

“…A similar method to determine the efficiency-optimized region of the CVT using the speed differential was also presented. 11 In a previous study, 12 an observation signal was introduced, which can be obtained from the rotational accelerations of the primary pulley and the secondary pulley, the velocity of the secondary pulley and the speed ratio of the CVT. It was found that the primary and secondary observation signals are out of synchronization when macroslip occurs.…”

Force-tracking control of a novel electric parking brake actuator based on a load-sensing, continuously variable transmission

“…The load-sensing CVT works on the same type of actuating principle as the push-belt CVT, in which the reduction ratio changes with the load torque. 25,26 The structure is shown in Figure 3. The torsional spring has a very small deformation due to the pre-load torque in the original state of the load-sensing CVT, the metallic ball is in the shallow chute, and the rotating radius of the belt is large on the drive shaft in comparison with the rotating radius of the belt on the driven shaft.…”