Compilation and verification of dynamic torque load spectrum of tractor power take-off

The power take-off (PTO) of the tractor is the key power transmission component for the tractor unit when used in the rotary tillage and driven harrow operations. It is subject to the random loads of different characteristics from the farm implements, and is the main loading part of the bench test for the tractor drive train. Torque load plays a decisive role in the service life of the PTO. The torque load spectrum is an important basis for the strength design and durability test verification of the PTO. In the indoor durability bench test of the tractor PTO, the torque loading is mainly based on static loading or static step-by-step loading. The dynamic characteristics of PTO load output by this mode is different from that of PTO load in actual field operation, which leads to the difference both in the failure mode and the test life of the PTO between the bench test and the actual field operation. Aiming at this problem, in this paper, the compiling method of the PTO dynamic torque load spectrum under multiple working conditions was studied. Firstly, the PTO torque in the rotary tillage and driven harrow operations under two soil types of sand and clay was measured. Secondly, the test data were preprocessed by the Hilbert Huang algorithm, and the torque frequency was distributed and statistically analyzed by the rain flow counting method. The variance matrix statistics were used to test the torque mean value and amplitude value correlation of four working conditions while the max correction decision coefficient was 0.009 99. The torque mean value followed a normal distribution and the amplitude value obeyed the Weibull distribution. Based on Miner's law and 20 CrMnTi modified S-N curve, the sample capacity of the torque data was determined under statistical conditions of 50% damage probability, 5% relative error and 95% confidence level. The PTO torque data was extrapolated using the time domain extrapolation method. The variance of the mean value of the peak-to-valley value in the threshold interval was taken as an index to check whether the threshold value selected in the extrapolation process caused the mean value of the peak-to-valley value to fluctuate drastically or not. Then, the trend of load time history and time history of original load after extrapolation was compared. The PTO integrated dynamic torque load spectrum was obtained when the time domain frequency extrapolation factor was 240, the torque cumulative frequency reached 106, and the average weighted four types of torque. The mean value and amplitude value two-parameter rain flow counts were performed on the PTO torque. The counting results showed that the PTO torque mean value, amplitude value and cycle number were synchronously extrapolated. Compared with the original data rain flow counting result, the mean value and amplitude value distribution law had similarity. Finally, the dynamic load spectrum compiled was used to drive the dynamic loading test bench to test the durability of the PTO. The engine torque was simulated by the motor in the driving unit while the PTO dynamic torque load spectrum was simulated by the motor in the PTO loading unit. During the test, a sudden drop of the torque measurement value in the PTO loading unit occurred at the moment of 642 h as the PTO shaft spline was worn and could not effectively transmit power. The failure time of PTO shaft spline wear in tractor actual field operation was 644.5 h. The test results showed that the failure time of the PTO was 642 h, and the failure mode was PTO shaft spline wear. Both the failure time and failure mode were consistent with the actual field operation, which verified the effectiveness of the integrated dynamic torque load spectrum. © 2019, Editorial Department of the Transactions of the Chinese Society of Agricultural Engineering. All right reserved.