Withstanding The Elements With FEA

By on October 7, 2009

Anchor Industries, Inc. of Evansville, Indiana has produced tents and other lightweight, portable buildings and structures for over a century.   Anchor first established its reputation with riverboat, carnival and circus tents and is now a leading supplier of tents and fabric products to the outdoor amusement and entertainment industries.

Their products can be found at water parks, municipalities, theme parks, circuses and amusement parks around the world.  Given their expertise in producing high-quality, easy-to-assemble, portable structures for numerous industries, Anchor decided to re-enter the burgeoning military market of vehicle maintenance shelters (VMS).  The first Anchor VMS, designed several years ago to vie against a competitor’s lightweight maintenance enclosure (LME), was overpriced compared to similar products in the U.S. and Europe.  They turned to ALGOR FEA to create a more competitive, efficient and optimized shelter design.

The task of designing a new VMS was given to Richard D. Cook, the company’s Senior Structural Engineer.  Cook had used ALGOR FEA software on numerous occasions.  Most recently, he designed Anchor’s Funbrella Palm—a new shade device that benefited from design enhancements driven by FEA results.  From experience, Cook knew that he could quickly and inexpensively model and test various VMS concepts using FEA software.

Design Goals

The VMS is used by most branches of the military to house and service vehicles.  It may also be used for portable military command posts and for short-term housing of troops.  The VMS must withstand the elements wherever troops are deployed.  Specifically, it must withstand wind, snow and the weight of suspended accessories.  The design constraints include an 8’ maximum length for all parts, since the structure must be packed within a container with inside dimensions of 99 ½” L x 34 ¼” W x 38” H.

As the above analysis results show, the frame included separate beam elements defined for the splices.The competitor’s LME, weighing over 1600 pounds, required a minimum of sixteen people to carry and assemble and included a 276-page instruction manual.  Design goals involved a drastic reduction in the weight of the tent’s frame and covering, while generally keeping stresses below 60 to 70% of yield.  Design goals also included easy assembly of the building by eliminating the need for external parts such as pins, guy ropes and tie-downs, all of which were part of the previous structure.  Also, to make production cost-effective, the design had to use extrusions from the company’s inventory, rather than new parts.  The portable building also had to be competitively priced, while meeting the standards of the supply center for the Defense Logistics Agency (DLA).  Getting the VMS to meet these specifications would ensure that the company’s product would be considered for purchase by the U.S. Military.

FEA Models and Analyses

Using ALGOR, Cook created an FEA model based on the use of existing extrusions.  The frame was composed of several parts, including arches and purlins.  The purlins, or horizontal beams, were based on proprietary square aluminum tubes.  The vertical beams forming the arches were modeled using proprietary aluminum extrusions designed with channels into which fabric panels slide.  The channels eliminate the need for securing the vinyl panels to each other and to the frame, allowing for faster, simplified installation as well as a smooth, weather-tight fit and maximum stability.  Given the length of the tent and the 8’ maximum length of each part, the beams and purlins were necessarily spliced for disassembly.

As this analysis result shows, the frame included separate beam elements defined for the splices.  As the above analysis results show, the frame included separate beam elements defined for the splices.

Dead loads were applied to the center of each ceiling beam of the new VMS model, with boundary conditions defined for the ground.Cook defined the legs, arches and purlins as beam elements, using elements from 1’ to 2’ in length.  Given that the splices were critical stress points, he used different elements for both the leg and purlin splices.  Other parts of the frame included 8 X-cables made of ¼” steel tension cables to provide stability to the walls.  They were modeled using truss elements.  The walls and roof of the VMS were modeled using membrane elements copied from one arch and then assembled to represent the .02”-thick, vinyl-coated fabric.  The complete model incorporated approximately 1200 membrane elements for the covering and 600 beam elements for the frame.  He used a rectangular mesh on the sides, roof and lower ends of the cover and a triangular mesh for the apexes of the ends.  He then meshed the structural components uniformly, with refinements at the splices of the vertical legs.

With the FEA model initially defined and meshed, Dr. Cook began to define loads.  The dead loads, representing the hanging accessories inside the structure, were constant forces of 100 pounds applied at the center of each rafter and at the peak, for a total of 900 pounds.  He applied wind loads as high as 7 psf and a snow load of 10 psf loaded vertically on the roof.  There were five load configurations applied to each model, including three separate wind loadings, one snow load and one combination of wind and dead loads.  The ground was defined using fixed boundary conditions.  He used ALGOR’s Mechanical Event Simulation to consider the nonlinear effects of the X-cables included in the concepts.  Although the forces were steady, each loading comprised a timed event lasting twenty seconds, consisting of one second of rest to permit initial tension in the X-cables to distribute itself, seventeen seconds of increasingly applied load, followed by two seconds of rest.  The rest period allowed transients in the model to settle.  The model was tested several times.  Cook looked at the effects of the loads on the beam and membrane elements and examined the nonlinear effects on the truss elements (X-cables).

The results of snow loading on the complete structure showed that the model is well within yield standards.Cook ran models with 8’ and 16’ bays.  He refined the 16’ model using lighter aluminum extrusions for the frame and purlins and then repeated the analysis, checking each model to see if it withstood the loads and remained within 60-70% of yield.  By the fifth concept, he arrived at a model that optimized materials and remained within yield standards.  He also benchmarked his models against a model of a competitive product.  The final concept included two side purlins to withstand the loads from side winds.  He detailed this concept using Autodesk Inventor and then performed final analysis of that design in ALGOR.

The new VMS design is easy to assemble and requires far fewer than the 16-person minimum required by a competitor’s design.  The assembly process is shown here.  The new VMS design is easy to assemble and requires far fewer than the 16-person minimum required by a competitor’s design.  The assembly process is shown above.

The final purlin was a 2” proprietary square 6061-T6 aluminum tube with .125” wall thickness.  The frames were a proprietary 6061-T6 aluminum extrusion with channels.  The leg splices were designed as .625” x 2.625” A36 steel.  The eave and ridge weldments were primarily cut-outs of A36 steel.  The purlin splices were 1.68” square aluminum tubes with a proprietary shape, with ridges about 1.72” square for a slip fit when in the field.

Producing and Testing the Prototype

Next, Cook ran field tests of portions of the frame to verify the ALGOR results.  He applied a 240-pound load at the splices of the purlin, the most vulnerable stress point in the design.  This load was previously hand-calculated and run in ALGOR to ascertain that it produced stresses exceeding those revealed in any of the five load combinations tested in ALGOR.  “In the worst case, one purlin splice was very near the yield strength under the snow load.  If this configuration was not adequate, there was a heavier alternative,” said Cook.  “I was anxious to see if the splices were as strong as the FEA software predicted that they were and found through direct application of weight that the software was correct.  The beams that the ALGOR analysis allowed us to use were much lighter than expected, representing a 40% overall weight reduction.” The company then built a prototype of the VMS based on the FEA analyses.  The 40% reduction in aluminum was the greatest of the cost savings over Anchor’s first VMS and the competitor’s LME structure.  This gave Anchor an extremely competitive product.  Anchor Industries expects to deliver the first ten units for military use by August 1st.

The new VMS design is easy to assemble and requires far fewer than the 16-person minimum required by a competitor’s design.  The assembly process is shown above.

The fully-assembled VMS stores and shelters military vehicles, protecting personnel working on the vehicles from extreme weather conditions.  It may also be used for a portable military command post and for short-term housing of troops.  The fully-assembled VMS stores and shelters military vehicles, protecting personnel working on the vehicles from extreme weather conditions.  It may also be used for a portable military command post and for short-term housing of troops.

Future Plans

Given the design success of the VMS, Anchor intends to use ALGOR for any new designs within this family of products.  “I will use ALGOR for future projects that require material optimization—to save time, reduce cost and build competitive products.  ALGOR software is user-friendly and represents an excellent value,” said Cook.

Senior Structural Engineer Dr. Richard D. Cook used ALGOR finite element analysis software to design a new vehicle maintenance shelter for the U.S. Military.Senior Structural Engineer Dr. Richard D. Cook used ALGOR finite element analysis software to design a new vehicle maintenance shelter for the U.S. Military.  Richard Cook earned a B.S. in Aeronautical Engineering, an M.S. in Engineering and a Ph.D. in Engineering Science from Purdue University.  He also holds an M.B.A. from the University of Evansville.  Before becoming a Senior Structural Engineer for Anchor Industries, he held numerous positions in academia and industry.  Most recently, he was an Assistant Professor at the University of Southern Indiana, where he taught computer science.  His most recent industry positions before Anchor Industries include Manager of Mechanical Engineering and Projects Coordinator at Integrated Systems Manufacturing, Inc. and Manager of Engineering for Faultless Caster Division.  His use of FEA software dates to the late 1970s.  He first used ALGOR software in 1991 at Faultless.



 

About CADinfo

Online CAD information and resources since 1990

Leave a Reply

Your email address will not be published. Required fields are marked *