Experimental Study on Truck Related Power Losses: The Churning Losses in a Transmission Model and Active Flow Control at an A-pillar of Generic Truck Cabin Model

Doktorsavhandling, 2019

The fight with global warming constantly forces vehicle manufacturers to innovate, in order to be able to reduce the CO2 emissions of their product. This means that marginal efficiency gains in every component are considered beneficial for total reduction of CO2 emissions. For long-haul heavy-duty trucks, the transmission losses and the aerodynamic drag are two of the important losses.

If we look at the truck transmission, the losses in form of churning losses stems from the interaction between oil and partially immersed rotating gear wheel. It is obvious that reducing the amount of oil inside the transmission lower the churning losses. However, the presence of oil inside a transmission is not solely for lubrication purposes but also for heat dissipation. Excessive heat on the components impacts greatly on their durability / lifetime. Finding a balance between low loss and high heat dissipation requires further studies in the area of fluid dynamics.

In this study, oil flow inside a transmission was studied by means of flow visual- ization, velocity measurement, and torque measurement. A simplified version of a transmission was specially built to be able to study in more detail the oil flow inside the gearbox. The test section, test object, and test oil have clear appearance for ensuring good optical access. Flash photography and high speed recording were used to visualize the oil flow. Particle image velocimetry was used to measure the oil flow velocity. The torque loss was measured for different oil level and gear geometry. The torque loss estimation formula was fitted with the help of machine learning algorithm in search of general equation that described the churning losses.

The results shows that rotational speed, immersion depth and geometry have dominant effect to splash pattern and oil distribution inside the transmission. The 2C2D-PIV measured the oil flow velocity in the mid-plane and revealed recirculation regions below the gear and the pinion. The 3C2D-PIV was done in several different planes and reconstruction of the data shows the three dimensional flow below the gear wheel. Cross plane measurement revealed vortices below the gear wheel and oil flow velocity in the gear meshing region. Air entrainment into the oil sump (aeration) was observed during the measurement. Aeration was caused by solid and liquid impingement to the free surface. The aeration level was estimated up to 20%. Torque measurement shows up to 9% increase in torque was found when comparing torque data between aerated and non-aerated oil. Curve fitting to the torque measurement data was done with help of machine learning algorithm.

If we look at the truck aerodynamics drag, it stems from flow separation. Trucks are considered as a bluff body from fluid dynamics perspective. The aerodynamic drag of a bluff body is dominated by the pressure drag, which is the pressure difference between the stagnation point and the wake. The earlier the flow separate from the body, the larger the wake size. There are several locations in the truck where the flow separates. The front of the truck, the A-pillar, the under-body and the wheel, the gap between tractor and trailer, and the wake behind the trailer.

In this study, flow separation at the A-pillar was investigated. An active flow control in term of zero net flux synthetic jets was applied. The synthetic jets role were to energize the boundary layer which in return suppress the flow separation. The study was performed in a generic truck cabin model specially build for this study. The measurements were done at Chalmers wind tunnel with inlet velocity of 20 m/s. Combining the inlet speed with the 0.4 m characteristic length of the model, the resulting Reynolds number was 5 × 105. Pressure measurement and velocity measurement using time-resolved PIV were done to quantify the flow velocity field with and without the active flow control at the side and at the wake region of the test object. Four different actuation cases (F + = 1, 2.1, 3.1, and 6.2) of the synthetic jet are studied and measured. Hot wire anemometry was used to characterize the actuators that produce the synthetic jet.

The results suggest that the receptive band of the shear layer and wake in this study was in the range between 0.7 < F+ < 3.1. The chosen F+ value of 1, 2.1, 3.1, and 6.2 successfully showed suppression of the flow separation at the A-pillar. The F+ = 1 shows the lowest absolute streamwise velocity and the shortest wake. The F+ = 2.1 and 3.1 show a lower base pressure region and a very similar wake configuration. The strongest level of flow separation suppression happened at F+ =3.1 . The F+ = 6.2 was chosen to confirm previous study that the velocity fields in the wake becomes independent from the actuation frequency.

The experimental data gathered in this study were proven to be useful and were used to validate a numerical simulation. The experimental data from the churning losses study were used to validate a numerical method, smoothed particle hydrodynamics. The data from active flow control study were used to validate CFD simulation using PANS turbulence model. Additional knowledge to the fields of fluid dynamics especially in oil flow around a spur gear inside a transmission and active flow control using synthetic jets for A-pillar were added. The work has laid a good foundation for future study in these topics.