Penn State / Visteon Design Project

Low Cycle Bending Fatigue Analysis of SAE 8620 Heat Treated Gear Steel

Team Members: Christopher Ault, Dave Cox, Nathan English

Project Sponsor: Visteon Corporation, Axle and Driveline Systems Division, Sterling Heights, MI

Sponsor Contact: Ernie Besler

Faculty Coach: Dr. John Lamancusa

Course: Mechanical Engineering 415W, Fall 2001

 

  Executive Summary
Visteon is a system supplier of advanced automotive driveline systems and is continually looking for new and better ways to optimize these systems for cost, weight, and performance. The focus of this project is to develop helical gear S/N life curves to be used in combination with vehicle road load data acquisition and cumulative damage analysis for component sizing. The main purpose of the project is to develop a better understanding of the fatigue behavior of SAE 8620 gear steel in the low cycle high load fatigue range (1K – 10K cycles). The analysis would include the fatigue evaluation of different gear tooth modules and heat treat parameters.

  Importance of Project
The motivation to have a PSU senior design team perform this project is the following:
  • Need to develop driveline sizing strategies and tools that provide driveline systems that meet the customer’s needs and provide the optimum balance for cost, weight, and reliability
  • Need to develop analytical and validation tools that provide quick and reliable validation of the base design and allow evaluation of design alternatives
  • Need to develop and link vehicle driveline simulation software tools with customer representative load histories to establish life prediction of driveline components and avoid the need for expensive long duration vehicle and laboratory testing

   

The strain gaged samples pictured (above left), were supposed to be tested until failure like the sample above right.


  Recommendations
Upper fixture and 20kip load cellTo ensure future success of this research, it is the opinion of this semester’s senior design team that the following should be taken under careful consideration. The largest hurdle we had to overcome was the testing category. The MTS load cell was too large (20 kip) for four-point bend testing of this material, therefore we were not using its full range. We would recommend acquiring a smaller load cell (5 kip) to resolve this problem and ensure much better resolution. Another problem arose when we sampled and broke a bar. In this instance the top apparatus of the bending fixture had slop in the pin joint, which caused the sample to vibrate between the fixtures when the peak of the sinusoidal loading function came close to zero. The pin vibrated because there was not enough load being applied to the bar to keep the joint tight. This problem seems to be virtually unsolvable because the slop in the pin joint connecting the top apparatus to the load cell is designed to ensure an even bending moment is applied to both sides of the notched bar. One possible, but extremely unfeasible solution to this problem would be to machine a piece to replace the pin connecting feature, which simply eliminates the pin joint by attaching top fixture to the load cell. However, it is vital to remember that if this is attempted, the piece would have to be precise, to ensure the top fixture sits perfectly level on the sample. Otherwise the downward force (bending load) will not be displaced evenly on the sample and the data will be inaccurate and unusable. Another area that needs further research is whether or not displacement controlled testing is sufficient for the type of data needed. It is our group’s opinion that a controller that provides feedback to the load cell is necessary to constantly monitor and adjust for the desired strain in the notch of the sample. This is necessary because of the high load / low cycle type testing.  The sample will be in the plastic region when it fails and therefore displacing the bar to the same distance every cycle does not yield reliable data. It is hypothesized that a strain (constant rate) control test would be more applicable to this type of testing. A controller that was capable of feedback to the load cell was not available in the facilities at the Engineering Mechanics Department. For the reasons mentioned, Visteon should continue the research of SAE 8620 steel under high load and low cycles. Penn State simply does not have the resources or the budget to complete such research at this time.