The two-poster rig plays an important role in accelerated durability evaluation in the motorcycle industry, similar to what a four-poster rig does in a car industry. The rig simulates the exact road conditions in the vertical direction through tire coupling by applying feedback control on displacement. On account of its ability to simulate to the exact customer usage conditions, it reproduces the failures realistically as they happen in the field. However, as a complete vehicle is required for testing on the rig, the testing happens mostly in the advanced stages of product development. Any failures beyond the concept stage have a huge impact on the development time and cost.
Many vehicle manufacturers and component suppliers are moving toward a laboratory test as the signoff durability test. Some are also thinking that to replicate this final laboratory test in the analytical world would be a valuable tool. Before this can be possible, a sufficiently accurate model of the specimen and test rig must be developed. This requires conducting both physical and virtual tests. Where results are not sufficiently matched, an explanation for the difference must be determined and either the physical or virtual system adjusted.
Virtual testing provides an opportunity for the development of such a characterization. Due to the repeatability of the virtual testing process, analysis will be able to discern small changes to the vehicle/motorcycle model.
Correct modeling of front and rear motorcycle shocks is critical to obtain good analysis results. Each shock provides a spring force, friction force, and damping force.
One of the most difficult tasks in virtual testing is to model elastomer isolators such as bushings and rubber mounts accurately. Elastomer isolators usually have nonlinear load deflection relationship. The relationship is also frequency dependent. With small deflect, however, the isolators can be simplified as linear and frequency independent.
For the motorcycle studied, most of the isolators were made of hard rubber and the deflection of these isolators was relatively small and quite linear. Therefore, these isolators were modeled by linear bushing elements.
In tire-coupled and hub-coupled lab tests, usually, a dummy rider is used to simulate the real rider and some static weights are placed on the passenger seat to simulate the passenger. An MSC.Software Corp. Adams model was created according to information provided by the dummy rider manufacturer. The dummy rider includes head, lower arm, upper arm, lower leg, thigh, head, and torso as separate parts. Between the rider and seat, there is a spring damper simulating the seat isolation. On the passenger seat of the motorcycle model, parts with proper weight were added to simulate the static weight in the lab.
Rider and passenger simulation in lab testing can have a major impact on lab testing results. Virtual testing could be an effective tool to evaluate different rider and passenger simulation methods and help to select active rider and passenger simulation designs.
Road load data acquisition was done on a rough road, and the same was simulated on a two-poster rig. To validate the model, the motorcycle was tested in the lab using the simulated road load on an MTS Systems Corp. two-poster rig. Satisfactory virtual to physical validation results were achieved. After model validation, the motorcycle model was integrated with an MTS tire-coupled virtual test rig to form a virtual testing system.
Using MTS’ RPC Pro simulation software, different iterations were conducted using the virtual testing system. Front and rear wheel hub acceleration and shock displacement signals were used as control signals for tire-coupled virtual testing. A comparison was done on physical and virtual acceleration, potentiometer, and strain signals. The results indicate that there is a fairly good match for acceleration signals between the tire-coupled virtual test system and the two-poster rig.
To better understand the limitations of a two-poster system, the same motorcycle model was integrated with a hub-coupled system to form a hub-coupled virtual test system. Front and rear wheel hub vertical accelerations, fork bending moment, swing arm pivot joint longitudinal force, and front and rear shock displacements were used for hub-coupled virtual testing. On comparison with the road, a definite improvement was seen in the results. In general, it was observed that virtual testing with a hub-coupled system has better acceleration and strain correlation results than a tire-coupled system.
Also, a comparison study was conducted to find out the difference between virtual testing using tire-coupled and hub-coupled motorcycle virtual testing systems. The results showed that the tire-coupled system could only reproduce vertical signals accurately, which is what a two-poster rig does. The hub-coupled system can reproduce vertical and longitudinal signals accurately, a combination of which is what is seen on the road. As a result, damage associated with longitudinal loads for components such as front fork can be reproduced more accurately by using a hub-coupled testing system.
This article was based on SAE International technical paper 2010-01-0925 by Ravi Kharul, Sivakumar Balakrishnan, and Dora Karedla of TVS Motor Co. and Dr. Shawn You of MTS Systems Corp. The paper is being presented at the SAE 2010 World Congress technical session “Automotive Engineering Testing and Test Methods” on April 15.