This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.

Creating accurate wear-test rigs using simulation

Author : Aliihsan Karamavruc is Senior Computer Aided Engineering Analyst at BorgWarner Turbo Systems, USA.

07 April 2017

BorgWarner is a global product leader in powertrain solutions, with a focus on developing powertrain technologies to improve fuel economy, emissions and performance.

Typical BorgWarner turbocharger

When designing wastegate (WG) actuators, BorgWarner must ensure the life of the actuators over millions of cycles of operation despite non-uniform loading that contributes to the difficulty of predicting wear patterns. A new ANSYS Mechanical feature that analyses wear between sliding parts provides accurate predictions in one day - a big time saving compared to the several days that were required in the past to perform physical testing. This capability will expedite design iterations between the designer and analysts and help to design more accurate wear test rigs.  

In a very popular style of BorgWarner turbocharger, the WG actuator shaft is mounted vertically inside the turbocharger. One end of the shaft is connected to a flapper in the exhaust gas flow stream that seals the WG. The exhaust gas exerts a continual force on the flapper, while a spring at the opposite end of the shaft resists the force to keep the WG closed. The internal spring of an actuator is calibrated to a predetermined boost level. When this boost level is reached, the flapper opens and allows exhaust gas to bypass the turbine. From a wear standpoint, the primary concern is wear on the shaft and bushing, which is difficult to predict because of the non-uniform contact due to force exerted by the flow stream and rotational motion of the shaft.

In the test rig, a motor drives an eccentric crank shaft connected to a crank arm that moves the shaft back and forth over the range that would be driven by the flapper. A 12.8kg mass hangs on the end of the shaft to represent the force exerted by the flow stream on the flapper. Each revolution of the motor represents one open-and-close cycle of the device. The shaft is also maintained at a temperature of around 450°C to replicate the real-life operating temperature. The test rig accurately predicts the wear experienced by the WG actuator during turbocharger operation, but it requires building an expensive prototype. In addition, running the test rig through enough cycles to predict the wear on the actuator takes about a week.

Without analytically predicting the wear, BorgWarner engineers often found that their first design did not meet wear-life specifications, so the entire design, build and test process needed to be repeated, often several times. The ability to determine wear on the shaft and bushing prior to building a prototype would save time and avoid multiple prototypes. Until recently, the only method available to analytically determine wear had been to use the Archard wear equation, which describes sliding wear based on the load, sliding distance, hardness of the contacting surfaces and a dimensionless constant K. While this equation is useful in predicting wear on evenly loaded surfaces, it does not take non-uniform loading into account so it cannot be used in this case.

ANSYS Mechanical now calculates wear

ANSYS Mechanical mesh of key components of the test rig

Recent releases of ANSYS Mechanical have given engineers the ability, for the first time, to calculate wear based on non-uniform loading. In this case, BorgWarner engineers began with a computer-aided design (CAD) model of the test rig, including the crankshaft, crank arm, bushing, shaft and pendulum assembly (which holds the weight representing the flow stream pressure). The boundary conditions for the model included a fixed support holding the bushing in place, a mass connected to the end of the shaft, and rotational joints in the bushing and crank arm. The material properties were defined as a function of temperature. Material hardness was defined as a function of the yield stress of the underlying elements, but temperature was not included in this simulation. The generalised Archard wear model was used to predict wear based on the loads calculated at each point in the contact zone by the ANSYS Mechanical simulation. The value of K was determined based on an engineering handbook. A frictional contact was used between the shaft and the bushing.

Engineers ran the simulation over 720° of motor rotation, which amounts to two open and close cycles of the actuator. The contact nodes were moved as per the wear increment at each time step. Additional equilibrium iterations for the corrected deformation were then performed. The software performed rezoning whenever the mesh became distorted due to wear.

Accurate prediction of wear

The simulation results included deflection of the components as a function of crank shaft rotation angle, which plays an import role in the resulting wear pattern. The simulation also calculated contact pressure as a function of crank rotation angle, an important predictor of wear. The contact pressure over the surface of the shaft as predicted by ANSYS software matched the wear patterns on a shaft that had undergone physical testing. The software also predicted the wear generated on each node of the shaft during the two cycles of rotation. Engineers plotted the accumulated wear over the two cycles and extrapolated this information for the full one-week test period. 

The new simulation capability in ANSYS Mechanical can predict wear with a high level of accuracy. This process reduces the time needed to investigate a design from several days to just one day. This capability will save BorgWarner time and money by making it possible to evaluate different design alternatives based on their wear performance prior to the prototype stage so that just one prototype can be built with a high degree of confidence.


Contact Details and Archive...

Print this page | E-mail this page