INTRODUCTION As emission requirements become more stringent, fuel economy becomes increasingly important. The transmission plays an important part in improving vehicle fuel economy. The transmission efficiency and the spin loss has great impact on the fuel economy prediction. Many people tried different ways to estimate transmission efficiency [1]-[4]. A lot of research has been done for the transmission components [5][6]. However, the studies on the transmission output chain churning have rarely been reported. The preliminary studies show this churning loss may have 10~15% contribution to the total transmission spin loss. In normal transmission design, a chain may be used to transfer the power from one shaft to the output shaft. The driven chain sprocket is partially immerged inside the sump oil. When the chain passes through the oil, it creates the churning losses. It is really difficult to calculate the drag force analytically using fluid dynamics with an accepted accuracy for the real world applications. Part of the reason is that it is hard to consider the flow separations. This causes wakes which build the low pressure area and result in high drag losses. Current technology still relies on the experiments [ 9]. A complex CFD analysis may help to find the solution, but it involves high computational time and efforts [1]. The accuracy of the modeling is still not clear. In this paper, an experiment study was carried out with the testing fixture in Southwest Research Institute (SwRI). Based on the testing data, the relationship of the drag coefficient and the Reynolds number were derived for the chain. A closed form equation of chain spinloss was developed.TESTING SETUP The chain and sprocket assembly was installed inside an aluminum chamber shown in Figure 1. The sprockets were mounted to shafts which extend outside of the box for support and alignment. The back baffle and scoop snub was installed against the chamber back wall. The front side of the baffle was attached using brackets to simulate the real transmission environment. In addition, a cylindrical enclosure was fabricated to represent the final drive housing and prevent oil from spilling onto the chain. A ruler was mounted to the chamber to determine the fill level relative to the lower driven shaft centerline. The dynamic oil levels during the tests were measured through an ACT fluid level sensor. Fluid density changes due to oil aeration were measured through use of a Micro Motion density meter installed in line with an external oil circulation pump. Careful attention was paid to maintain clearances that represent the actual transmission environment. The chain assembly was driven by a 15KW, 6000rpm motor mounted outside of the chamber. A high accuracy 20Nm Futek torque meter was used to measure the torque required to drive the chain assembly. A magnetic pickup was used for speed measurement. Four oil lubricated bearings were used to provide alignment and assembly load support. Each bearing has a K type thermocouple tapped for temperature monitoring and recording. A picture of the test stand is shown in Figure 1.Transmission Output Chain Spin Loss Study Sen Zhou General Motors LLC Bryan Williams Southwest Research Institute ABSTRACT Transmission spin loss has significant influence on the vehicle fuel economy. Transmission output chain may contribute up to 10~15% of the total spin loss. However, the chain spin loss information is not well documented. An experimental study was carried out with several transmission output chains and simulated transmission environment in a testing box. The studies build the bases for the chain spin loss modeling and depicted the influences of the speed, the sprocket sizes, the oil levels, the viscosity , the temperatures and the baffle. The kriging method was employed for the parameter sensitivity study. A closed form of empirical model was developed. Good correlation was achieved. CITATION: Zhou, S. and Williams, B., "Tra