<div class="section abstract"><div class="htmlview paragraph">Electric vehicles (EV’s) present new challenges to achieving the required noise, vibration &amp; harshness performance (NVH) compared with conventional vehicles. Specifically, high-frequency noise and unexpected noise phenomenon, previously masked by the internal combustion engine can cause annoyance in an EV. Electric motor (E-motor) whine noise caused by electromagnetic excitation during E-motor operation is caused by torque ripple and radial excitation. Under high speed and high load operating conditions, the overall sound level may be low, however high frequency whine noise can impair the vehicle level NVH performance. An example of a previously masked unexpected noise phenomenon is a droning noise that can be caused by manufacturing quality variation of the spline coupling between the rotor shaft of the E-motor and the input shaft of the reducer. It is dominated by multiple higher orders of the E-motor rotation frequency.</div><div class="htmlview paragraph">In this study, the high speed and high load condition whine noise problem was reproduced through electromagnetic and structural analysis. The countermeasures (E-motor geometry refinements to reduce the excitations and mechanical system transfer path modifications to reduce the vibration response) were defined and the effects investigated. Mechanical system modification to improve NVH performance without increasing the mass is challenging. However, E-motor air-gap geometry optimization, such as slot opening modulation and rotor notch modifications achieved significant noise reduction without critical trade-off of other performance.</div><div class="htmlview paragraph">This paper also describes the basic improvement plan proposed through the simulation of the droning noise behavior. The main two manufacturing errors (pitch error in the spline coupling and the alignment error between the rotor and the gearbox shafts) that are responsible for the unexpected droning noise were identified. Using simulation it was shown that the droning noise can be reduced through improved quality control and tolerance design optimization of factors such as axis self-alignment and spline gear tooth quality.</div></div>

<div class="section abstract"><div class="htmlview paragraph">In electric-powertrains, noise and vibration can be generated by components such as gears and motors. Often a noise phenomenon known as rumble or droning noise can occur due to low shaft order excitation at the spline. In this study, we identified the excitation source for spline induced rumble noise and developed a novel analysis method. First, a detailed spline model, believed to be the key factor for rumble noise, has been developed and verified by comparison with Finite Element Method(FEM) analysis. In order to identify an excitation source, a typical electric-powertrain assembly model including the developed spline model was constructed and simulated. Results according to changes of key factors including spline pitch errors and shaft alignment errors were analyzed. Spline radial force has been identified as an excitation source of spline induced rumble noise. This was verified through comparison with the forced vibration analysis result and time domain analysis result. This paper presents two methods for simulating spline rumble, including the key contributing factors of spline pitch errors, shaft alignment errors and unbalanced mass. Time domain method is accurate but slow. So, a faster, lower fidelity quasi-static approach has been developed and verified using time domain analysis results. Finally, an analysis process for each development stage has been established utilizing the two simulation methods. In concept design stage, quasi-static analysis is appropriate to identify robust structure for rumble noise. In detailed design stage, manufacturing tolerances could be adjusted by reviewing various conditions for rumble noise through quasi-static analysis, and a direct response review through time domain analysis is also useful to check for problems.</div></div>