     
|
|
|
Technostar strives to develop an original CAE software that can truly contribute to the strengthening of product development capabilities of our customers in the manufacturing industry.
This article introduces a user example that was provided by Nissan Motor Co., Ltd., in which the total analysis time required for the analysis process was reduced by 80%.
Source : Automotive Information Platform
MarkLines Co., Ltd.( www.marklines.com) 2006.05
Extracted from an automotive solution report
|
The expanse of CAE use in component design
- A system develpment application
to promote CAE for designers -
Nissan Motor Co., Ltd. (Transmission vibration stress prediction system)
Figure 7 Development process
In order to reduce the time and costs for transmission design, Nissan is shifting its development process from a test oriented process in which prototyping and physical tests are repeated in order to decide the specifications, to a design oriented process, which uses computer simulation to identify problems in the early stage of development (Figure 7).
Transmission vibration stress is caused by strain amplification due to the bending resonance of powerplants, and affects the durability of the transmission.
The traditional performance design only focused on the bending resonance frequency, and the vibration stress, which is the final evaluation item, was detected by physical tests. In some cases, the cycle of prototyping and physical testings was repeated several times until the problems were solved.
In order to achieve more efficiency for analysis work, Nissan developed a new analysis system, as well as the transmission vibration stress prediction technology using large scale structural analysis models (Nissan technical paper, 2006).
Figure 8 Shell model and solid model
In traditional performance design, a normal mode analysis of the powerplant was performed using simplified shell models. In this application example, the stress concentration area needed to be identified in detail to obtain the transmission vibration stress. Therefore, the box geometry was modeled using 300,000 solid elements (Figure 8).
There is hardly any difference between the bending normal mode results of the two models, which indicates that shell elements have sufficient accuracy for normal model analysis. The solid model result shows that the crack location from the physical test coincide well with the stress concentration area obtained from the vibration stress analysis. In order to match the absolute stress value and the actual measurement value, the element size must be smaller than 1mm. However this will result in an immensely long calculation time.
Studies confirmed that the realistic method is to perform vibration stress analysis using a normal mesh (8mm), after which the area of stress concentration is remeshed using a fine mesh (0.5~1mm).
Although this method enables the estimation of vibration stress, there still remains a problem for its application to development projects, due to the long analysis time. Therefore, a new system was developed to largely reduce the analysis time.
The analysis flow of this estimation method is: (1) Creating the analysis model -> (2) Assembly-> (3) Normal mode analysis-> (4) Vibration stress analysis-> (5) Stress evaluation-> (6) Remesh-> (7) Assembly-> (8) Vibration stress analysis-> (9) Stress evaluation. Steps (1), (5) and (6) make up over 80% of the total process.
A meshing software with outstanding automatic meshing function was introduced in order to reduce the time required for creating an analysis model based on 3D CAD data.
A stress evaluation software was developed that can automatically detect areas of stress concentration, stress values, and areas above a defined threshold, in order to reduce the time required for evaluating vibration stress analysis results. Also, an integrated function of meshing and stress evaluation applications was developed that can automatically generate a fine mesh for the areas that were detected, in order to reduce the time required for remeshing areas of stress concentration.
By developing this new analysis system, Nissan managed to reduce their total analysis time to 40 hours, which is 20% of the previous number of man hours.
Written by Mizuho Fukuda, CAE consultant
|
|
|
|
