WALKING A FINE LINE

Reliability engineers at Intermec Technologies Corporation, Cedar Rapids, Iowa, used Mechanical Event Simulation software from ALGOR, Inc., to evaluate the suitability of an integrated scanner module design (see inset) for this mobile hand-held computer. The module is designed as a pod that attaches to the back of the computer and consists of a laser scanner for bar code recognition and a micro-switch for activation. Product testing revealed that the new activation micro-switch would fail to meet an operational life requirement for field use.

By: Bill Woodburn

Reliability Engineer, Intermec Technologies Corporation

Each day engineers and business managers across all industries are faced with a common dilemma - weighing design optimization vs. manufacturing costs while meeting time-to-market demands. As a reliability engineer at Intermec Technologies Corporation (Intermec) in Cedar Rapids, Iowa, I have dealt with this dilemma first-hand. Fortunately, designers and engineers can use CAD and FEA software to shorten design times by reducing physical prototyping. In addition, the animation of FEA results is a powerful tool for presenting engineering data to both technical and non-technical personnel about the outcome of design changes.

A large part of my job is to ensure that Intermec抯 supply chain system solutions (mobile hand-held computers and automated data collection devices) can withstand the wear-and-tear of use in distribution, field service, warehousing, utilities and transportation industries. Recently, I used Mechanical Event Simulation software from Pittsburgh-based ALGOR, Inc. (ALGOR), to evaluate the suitability of an integrated scanner module design for an Intermec mobile hand-held computer. The analyses helped my company make prudent "design vs. cost" decisions that balanced gains in product life with production costs.

DEFINING DESIGN ISSUES
Intermec抯 reliability team performs a full suite of environmental tests, such as operating and storage temperatures, temperature cycling, humidity and water resistance, on new product designs. We also perform durability and life tests such as packaged and unprotected drop, vehicle vibration, key/button life tests and touch panel durability.

For this particular integrated scanner module, extensive durability and life tests were performed. The module was designed as a pod that attaches to the back of the hand-held computer. The pod consists of a laser scanner for bar code recognition and a micro-switch housed within a durable polycarbonate/ABS plastic blend. The operator activates the scanner by pressing a large curved button attached to the housing. The scanner then reads the bar code into the hand-held device.

During preliminary testing of this module, it was discovered that the scanner activation button failed to meet the three million depression requirement needed to achieve an operational life of three to five years. In some instances, the buttons experienced cracking at the connection points within the housing and failed at around one million depressions (estimated to be one to one and a half years of operational use).

The next logical step was to perform a Mechanical Event Simulation on the original CAD solid assembly to simulate the mechanical failure and then develop a proposal for design modification based on the findings.

ALGOR抯 Mechanical Event Simulation software was chosen as our best analysis option because of the physics involved in pressing the scanner button. A typical linear static stress analysis determines stresses based on a calculated force applied to an area of the model. For the scanner button, the force applied could not be accurately determined due to the fact that different operators will press the button with varying force each time. However, we could easily measure the maximum displacement of the button when it is fully depressed. Using this known value, we could specify a time-varying prescribed displacement in a Mechanical Event Simulation. The software determined the bending stress results based on the motion and displacement that occurred over time. We could be confident that the stress results were accurate because we did not need to make assumptions about the forces at work in this design scenario.

DEVELOPING DESIGN OPTIONS
To begin the Mechanical Event Simulation, the reliability team acquired the CAD assembly, which was created in CoCreate Solid Designer, and imported the model into ALGOR using an IGES universal file. We used a combination of automatic and hand-meshing techniques to create a consistent solid brick FEA mesh throughout the entire model.

The assembly model consisted of two parts: the large curved button and the connecting piece of housing. The button model used six heat stakes that protrude through holes in the housing. During the manufacturing process, the heat stakes are melted down onto the housing to bind the two components. For the FEA model, truss elements were used to effectively bind the heat stakes to the housing. Fully constrained boundary conditions were added to the back wall of the housing component to represent the portion of the housing that was not modeled.

Once the modeling was complete, we identified the material properties for polycarbonate/ABS blend polymers, which were provided by Intermec抯 plastic molding supplier. We decided to use ALGOR抯 von Mises nonlinear material model with kinematic hardening to account for possible yielding in small areas of the part that exhibited cracking during laboratory testing. Finally, we added a prescribed displacement of .105 inches, the maximum button travel measured from test units, and defined an event duration of .01 seconds, the approximate time it takes to press the button, with a capture rate of 10,000 per second. This rate was chosen to ensure that the analysis would adequately capture the assembly抯 behavior throughout the entire event.

A Mechanical Event Simulation of the original geometry, shown above, was performed with ALGOR software to determine the maximum von Mises stresses that result when the button is activated. Based on these ALGOR results, the reliability team proposed four geometry modifications to reduce the bending stresses at the back wall of the housing. The inset shows the final housing geometry, which features an increase in material at the intersection with the back wall. The final design balances design optimization and production costs by satisfying life actuation requirements with the least amount of cost.

The Mechanical Event Simulation in ALGOR yielded stress results for each of the 100 timesteps. The maximum von Mises stresses and displacements occurred near the end of the event (timestep 80) on three narrow connection points at the back wall of the housing. We used the von Mises stress results as input for fatigue calculations to determine the Modified Goodman Safety Factor and the estimated fatigue life. The existing geometry had a safety factor of 0.69. A Modified Goodman Safety Factor less than 1.0 gives an indication of finite life while a value greater than 1.0 indicates near-infinite life. The calculated assembly life was found to be below one million depressions. I was confident in these results as the calculated value of depressions agreed well with empirical life test results.

Based on the location of the maximum stresses, the reliability team made a series of four proposed geometry modifications to reduce the bending stresses. These included removing one row of heat stakes, centering the material that intersects with the back wall of the housing and flattening the area of intersection, thinning and lengthening the intersecting material region and integrating the button modification with all of the housing modifications.

Safety factors for the four new designs ranged from 0.72 to 1.26. Estimated assembly life calculations ranged from just under one million to 5.8 million depressions. ALGOR抯 precision contouring was used to ensure that it was appropriate to compare results between the original model and the new designs. This feature gauges the quality of the finite element mesh, revealing any discontinuities that may affect the accuracy of results. The precision contours for each analysis revealed similar values. The final proposed design, which integrated the button and housing modifications, exhibited the greatest reduction in bending stresses.

The reliability team animated the Mechanical Event Simulation results using ALGOR抯 built-in animation tools. Previous experience has shown that our audience, both research and development engineers and business managers, can more readily absorb the quantitative and qualitative aspects of FEA results when they can view an animation of the simulated mechanical events. In addition, we could easily illustrate how the stress distribution changed as the modifications to the geometry were implemented. By showing the research and development engineers the animated results, we were able to present them with different options to consider - what to try and what not to try. This was the intent of performing the Mechanical Event Simulations with ALGOR.

DEFINING THE FINAL DESIGN
In determining the final design, Intermec抯 research and development engineers had to take into account the manufacturing process already in place. Plastic molding tools had already been created for the manufacturer of the scanner housing. The reliability team抯 best-engineered design would have required the manufacturing equipment to be retooled at a cost of tens of thousands of dollars. Whereas, a lesser modification of the placement of material at the connection points of the button with the back wall of the housing could achieve a satisfactory safety factor and estimated assembly life performance, while only costing a few thousand dollars in modifications to the production tools.

Decisions for changes in the final design were based on the stress contours from the original analysis. More material was added to the intersection of the housing back wall and housing and generous fillets were added to reduce stress concentrations. The cost of adding material to the part (or removing material from the tool) was minimal in comparison to the cost and time needed for a more substantial retooling. In the end, Intermec struck the optimal balance between design optimization and production costs by choosing a design that satisfied life actuation requirements with the least amount of cost.

By using ALGOR抯 Mechanical Event Simulation, the reliability team was able to circulate design optimization ideas quickly across several departments within the company. Presenting animated results also enabled individuals with all levels of engineering experience to understand of the outcome and implications of our stress analysis. In the future, we will continue to use ALGOR抯 Mechanical Event Simulation to troubleshoot product designs as well as to design adequate vibration test fixtures for Intermec抯 reliability testing laboratory.