The towers that carry transmission signals are getting taller, bigger and heavier to bear HDTV抯 dramatically improved, wide-screen digital audio/video information. This tower for WOAC Channel 67 in Kent, Ohio transmits the Shop At Home Network. It is supported by guy wires and the ROCKET-SOCKET Dead-end designed by PLP of Cleveland, Ohio.
	Guyed towers consist of a narrow structure supported by guy wires. Large guyed towers may be up to 2,000 feet tall and often have numerous guy wires, which connect to the tower at different heights.
	A dead-end is a component at the end of the guy wire that helps to anchor the guyed tower. It is almost 2 feet long and about 60 pounds.
	Phil Pisczak of Preformed Line Products used MES software on a PC running Windows NT. Here he views the results of the simulation.

  
Gearing up for High-Definition Television

National HDTV Conversion Effort Requires Re-Engineered Transmission Towers

The national conversion from analog TV to high-definition television (HDTV) will require just about every piece of hardware in the television industry to be replaced or upgraded, from television sets to the international broadcast infrastructure, including cameras, transmitters and telecommunication lines. The towers that carry transmission signals are getting taller, bigger and heavier to bear HDTV抯 dramatically improved, wide-screen digital audio/video information. Preformed Line Products (PLP? of Cleveland, Ohio is contributing to this technology with its ROCKET-SOCKET Dead-end for guy wires, which supports these larger communications/broadcast transmission towers. PLP is a $200 million global leader in the manufacture of cable anchoring systems. Its customer base includes most, if not all, of the nation抯 power utility and communication providers such as Verizon, Bell South and Adelphia in addition to a variety of resellers. They provide high-quality electrical conductor and optical fiber cable anchoring and control hardware and systems, overhead and underground splice cases and related products and high-speed cross-connect devices.

揥e offer highly-engineered, quality-tested solutions,?said Project Engineer Phil Pisczak. 揙ur dead-end housings are large, heavy and can be loaded up to 252,000 pounds, so it is critical to ensure that they will hold up well and provide lasting service.?The engineering of the ROCKET-SOCKET design involved ALGOR抯 PC-based Mechanical Event Simulation (MES) and laboratory testing. The result was a product that withstands higher mechanical loadings. By using the software, PLP engineers were able to expedite the testing, reduce the number of iterations in the laboratory and get their product to market more quickly. 

Providing a Firm Foundation for Telecommunication Towers

Guyed towers are constructed as a fairly narrow structure that must be supported by guy wires. These guyed towers must have a minimum of three guy wires to stabilize the structure from wind and weather. Larger guyed towers ?some up to 2,000 feet tall ?often have additional guy strand wires of greater strength with diameters in excess of 1.5 inches, which connect to the tower at different heights. 

As telecommunication towers increase in size to accommodate new technologies, all the components must be stronger, including the dead-end ?a component at the top and bottom of the guy wire that helps to anchor the guyed tower. The guy wire is fitted with a cone-shaped wedge that fits inside the dead-end housing. Two 揺ars?extend beyond the housing to attach to other components. This configuration has been used in a variety of environments for over 20 years in the PLP products. Previous products were made of ductile iron and terminated galvanized-coated, steel-based strand of up to 1-1/4" in diameter. As civil structures and antennas have increased in size and height, so has the need for guy strands of greater strength and diameters, up to 1 5/16" through 1 7/16". The challenge on this project was to design a new dead-end to support these larger communication/broadcast transmission towers that would withstand typical loads including 252,000 pounds of structural weight and wind loading as well as dynamic loads that might result from accidental impact. 

PLP engineers decided to investigate the possibility of manufacturing the dead-ends from austempered ductile iron (ADI) in order to make them stronger, without increasing their size. ADI is a heat-treatment process applied to ductile iron material (cast iron with nickel) for increased strength and toughness. The heat treatment controls the formation of the material, which contributes to the improved material properties. Depending on the exact heat-treat specification, different elongation and surface hardness characteristics can be accomplished. ADI is comparable in strength to cast steel but is not as heavy. Bearing surfaces, worked by relative movement, will develop an increase in surface durability. For these reasons, ADI is the material of choice for use in automotive brake calipers and other 搈ission critical?components in a variety of industries. Although more time-consuming and expensive to produce, ADI was a good candidate for bearing surfaces exhibiting high stress loadings, such as pin connections in the housing as well as the surface where the wedges are seated. Because ADI is a new material application, the performance of the Dead-end geometry needed to be tested, especially at the cold temperatures sometimes found at installation sites.

Pisczak considered both static and dynamic forces in analyzing and testing the design. 揥e first did a linear static stress analysis and it looked good, especially considering that we have a high safety factor built into the product,?said Pisczak. 揃ut in addition, we were concerned about dynamic stresses. One possible source of dynamic stresses is normally wind loading that causes wire to vibrate. However, based on years of field experience with this type of design, we knew we had a strong product capable of withstanding this loading. The source of dynamic stresses we were most concerned about was the possibility of abuse in the field. The housing is shaped like a tuning fork and could produce high resonances if struck. Pure resonances produce high mechanical loading, creating stress risers of typically 2 to 10 times the resonance stress that the part would otherwise experience. We wanted to ensure that if a ROCKET-SOCKET housing did resonate as the result of some type of impact, it would not be damaged.?h4> Solving the Problem of Looking at Dynamic Events

揑mpact analysis and testing is more challenging than static testing,?Pisczak points out. 揥hen parts are tested in the lab on the tensile bed, the equipment provides a plot of the part抯 behavior during breakage. With an impact analysis, it抯 more difficult to get that kind of information in the laboratory. Accelerometers can be used, but the sensors may be destroyed in the course of an impact experiment. High-speed cameras are not as effective at capturing small vibrations.?nbsp;

揗ES allows us to look at what is happening in an impact event without high-speed cameras, or complicated laboratory tests,?said Pisczak. 揑t is our microscope into the dynamic impact process. We get to see the motion, dynamic loading and stresses. Those results tell us much of what we could learn with the most sophisticated laboratory tests. That抯 why this technology is really powerful.?p> Working on a PC running Windows NT, Pisczak began by modeling half of the housing and an impact head in Pro/ENGINEER. 揝ince the housing is symmetrical, we could safely and reasonably assume symmetry for purposes of modeling,?said Pisczak. He then used ALGOR抯 InCAD technology to capture the Pro/ENGINEER geometry for the simulation and to automatically generate a solid mesh. 揟he software抯 automatic meshers are very effective on the first pass,?commented Pisczak. 揘o surface mesh enhancement was needed to get a usable mesh. I specified a finer mesh on the housing and a coarser mesh on the falling impact head. Because the impact head is moving, it has more equations associated with it. By applying a larger mesh to the moving part, I was able to reduce the run time.?

Pisczak input the material properties for the impact head and ADI dead-end and defined surface-to-surface contact between the two parts. The model was completely constrained on the bottom and a 0.04 second event was specified. Gravity was applied and the impact head was positioned above the housing model at a distance that Pisczak calculated would result in a 12 mile per hour impact.

In addition to this 12 mile per hour simulation, Pisczak prepared a second model with very little modification, that positioned the impact head at a distance that would result in a 48 mile per hour impact. 

Evaluating Mechanical Event Simulation Results

The results of the simulations showed Pisczak some additional performance behavior. 擲ince the housing is shaped similar to a tuning fork, I expected it to vibrate like one,?said Pisczak. 揟he software allowed us to see the wave of resonance travel up and down the part. Without the use of MES software, it would have been more difficult to make effective changes to the geometry.?nbsp;

	The results of the simulation revealed that the stresses were well below the yield stress of the material, but that they occur at the fillet at the base of the ear. Based on these results, a larger fillet radius was applied to that area to increase the part抯 strength.

In addition to displacement results, Pisczak looked at the location and magnitude of the stresses. 揥here the stresses occur geometrically is very important,?said Pisczak. 揑 look at whether the high stresses are occurring on the surface or through the part. Stresses appearing through the thickness of a part can be evaluated with the software quickly. I look at the magnitude of the stresses in comparison to the yield stress of the material. In this case, the highest stresses were well below the yield point of the material, but it occurred at a feature ?the fillet at the base of the ear.?As a result of the simulation, Pisczak changed the geometry of the ROCKET-SOCKET housing to put a larger fillet radius at the base of the ears. A prototype was constructed for laboratory testing.

揑 use simulation to find the 慼ot spots?and then modify the geometry and do the final testing in the laboratory,?said Pisczak. 揑 use this technique to expedite the process. It is mandatory to do proper laboratory testing, but MES helps me to reduce the number of iterations needed in the laboratory.?nbsp;

In the laboratory, the ROCKET-SOCKET housing prototype was dropped 5 feet onto a steel plate to replicate an impact. The experiment was repeated 20 times at room temperature and 5 times at -40 癋. Although the abuse was severe, the housing performed well.

揟he impact testing is far more abusive than anything we expect this product to see in the field,?said Pisczak. 揥e test so extensively because these product are big and heavy and could possibly be inadvertently dropped.?

PLP Finds Extensive Applications for Mechanical Event Simulation 

	The new ROCKET-SOCKET assemblies are rated to bear up to 252,000 lbs. PLP first tested them in the laboratory on a tensile bed (upper left) and then in a drop test (lower right).

揗ES demonstrated to me that a whole new level of information is available to enhance the design process,?said Pisczak. 揑t is as significant to the way I work as when I discovered in the mid-eighties that FEA software could give me stress information before the part was made. The application of MES is universal ?the only limit is in one抯 imagination.?p> 揑 now perform nearly every analysis with the nonlinear MES solver, because there are very few scenarios that I know upfront will be truly static and linear,?continued Pisczak. Pisczak has used the software on some highly diverse cases that involve a wide variety of materials including metal, plastic, neoprene, rubber, urethanes and others. 揟ake, for example, the case of a plastic communications housing subjected to pressure testing. Linear static FEA is often used for that kind of scenario. Performing a MES with nonlinear material input lets me see how and where the stresses originate and how they propagate, which is more useful information than I can get with linear static stress analysis.?nbsp;

揂LGOR抯 interface has made analysis and simulation easier to use while still allowing the analyst to control even advanced analysis parameters,?said Pisczak. 揟he control the interface gives me over analysis parameters and results enables me to extract more information. For example, I can find the stress value on a specific node on the model just by clicking on it. These kinds of capabilities make ALGOR a very powerful, very professional analysis package.?  

Phil Pisczak earned a Bachelor of Science in Mechanical Engineering from Case Western Reserve University and is presently finishing a Masters of Engineering program at Case Western Reserve University with emphasis on structural and material properties. He began using finite element analysis in 1984 using Intergraph's Rand-Micas FEA software and has been using ALGOR software since 1988. He has worked for Reliance Electric, Allen-Bradley and Sealy as well as for Preformed Line Products and holds several patents.