The detonation of an underwater plastic explosive rocks a barge containing fluid power equipment fabricated by Huber, Inc. ALGOR's DDAM software accurately predicted that Huber's equipment would pass the shock test requirements. Image courtesy of Hi-Test Laboratories, Inc.

Huber, Inc. Simulates U.S. Navy Shock Testing with ALGOR DDAM Software

"Ten seconds, nine, eight, seven, six," counted down a voice projected via loud-speaker across the Hi-Test Laboratories, Inc. lakeside facility for naval shock testing in Arvonia, Virginia. On the lake, a barge floated as the countdown continued, "Five, four, three, two, one, fire!" Twenty-five feet beneath the lake surface, a 60-pound plastic explosive was detonated, sending a shockwave through the water. The impulse violently rocked the barge and erupted in a spray of water high into the air.

For Huber, Inc., a manufacturer of fluid power systems located in Jefferson, Louisiana, over three years of work hinged on this split-second explosion. Inside the barge, equipment designed and fabricated by Huber was mounted to the hull in accordance with U.S. Navy requirements for shock qualification. Could the equipment survive the shock test? If not, redesigning, refabricating and retesting would be very costly to Huber in both time and money – a structural failure would entail a three-month setback and redoing the shock test alone would cost an additional $50,000.

"It was very important for us to verify that the equipment would survive the shock test on the first try," said Dominic Lorino, Huber Mechanical Engineer. Huber needed to comply with a U.S. Navy requirement that all mission-critical equipment be qualified by the Dynamic Design Analysis Method (DDAM) for "Grade A" shock loading such as could be received from a torpedo, missile or depth charge. "The test specifies a very short, powerful shock impulse. All of the equipment was hull-mounted, which meant it could experience up to 190 Gs in several directions. If the equipment shook loose, it could damage the ship or personnel. We had to show that the equipment would remain intact and keep on working."

Lorino had used finite element analysis (FEA) software from Pittsburgh-based ALGOR, Inc. to verify the designs with DDAM computer simulation prior to the expensive physical shock test. "Because ALGOR FEA had predicted the equipment would survive, we felt confident going into the physical test," he explained. "As it turned out, all of our equipment passed with no structural failures."

	This artist rendition of the LPD 17 ship shows the location of the sterngate. Huber's HPU is designed to control the opening and closing of the sterngate. Images courtesy of NAVSEA (PMS 317) DET New Orleans.

Work on this project began in December 1996, when Avondale Industries, a division of Northrop-Grumman, one of the largest shipbuilders in the United States, was awarded a $641 million contract to construct the U.S. Navy's new Landing Platform Dock (LPD) 17 ship, the first of the San Antonio Class amphibious assault ships. Currently, the LPD 17 is being constructed at the Avondale Shipyard in Louisiana with completion scheduled for 2004.

In 1999, Huber was subcontracted by Avondale to build several pieces of equipment for the LPD 17's sterngate system. The sterngate is located at the rear of the ship, where large upper and lower gates open and close to release or admit Amphibious Assault Vehicles (AAVs) or other craft, typically manned by U.S. Marines for missions on sea or land. The equipment that Huber built included hydraulic power units (HPUs), storage tanks, cylinder boxes and the upper and lower sterngates.

For each piece of equipment, Lorino created computer-aided design (CAD) and FEA models and conducted DDAM simulations, which indicated the need to significantly redesign in some cases. For example, Lorino greatly revised the design of the sterngate HPU. "The first design went way past the yield point," he said. "I redesigned the whole unit, which included using a stronger material, changing the location of equipment, changing the orientation of supports and increasing the number of bolts to make the structure stronger."

	This photograph shows the sterngate HPU at the Huber fabrication shop.

The sterngate HPU is a 14,000-pound fluid power system with many components including two hydraulic manifolds, a 300-gallon reservoir, two motor-and-pump assemblies and various other pieces of equipment all mounted on a skid structure. It is designed to supply hydraulic fluid to the cylinders that operate the upper and lower sterngates.

"One of the biggest problems we had with the sterngate HPU concerned its two manifolds," said Lorino. "The manifold for the upper gate is about 900 pounds and the one for the lower gate is about 1,500 pounds. Controlling and supporting these massive weights during shock loading was a challenge."

Lorino created a finite element model of the sterngate HPU using Superdraw, ALGOR's precision FEA model-building tool. "The manifolds were modeled as lumped masses," he said, "and connected to the skid with beam elements to simulate the weight of the objects. Another lumped mass was placed inside the tank to simulate the weight of the fluid, which is about 2,800 pounds. The rest of the structure was modeled with plate elements. The base of the model was fully constrained to simulate the bolts that fix the sterngate HPU to the skid."

	With the roof of the barge removed, Huber's fluid power equipment can be seen mounted to the hull. The sterngate HPU is highlighted in red.

To analyze the model, Lorino first performed a natural frequency (modal) analysis to obtain the mode shapes and natural frequencies. Next, he set up and performed a DDAM analysis that used the results from the natural frequency analysis in conjunction with an input spectrum. "As called out in the Navy shock guidance specification, 80 percent of the modal mass must participate in the model," he said. Lorino defined the cut-off percentage by simply clicking on an option within FEMPRO, ALGOR's finite element modeling, results evaluation and presentation interface. ALGOR's DDAM software is specifically designed to simulate the U.S. Navy DDAM shock qualification test, including options that enabled Lorino to conveniently specify the type of ship, mounting location, shock spectra, type of material and user-specified coefficients. In separate analysis runs, he analyzed the model's response to various shock spectra including in the fore and aft (front and back), athwartships (side to side) and vertical (up and down) directions.

Lorino evaluated the analysis results in FEMPRO's Superview IV Results environment. "Mostly, I checked the stresses," he said. "That's what we were worried about, because stresses show when the material starts to yield and when it goes into plastic deformation, which could indicate a structural failure." Evaluating the analysis results involved Lorino's engineering judgment. "In the sterngate HPU model, there were a few areas of localized stress, but everything else was way under yield. Were the stresses around a bolt hole? Were they around a part that could separate from its foundation? A lot of interest went into looking at the calculated stress distribution to determine whether the stress levels were critical."

	Huber Mechanical Engineer Dominic Lorino used ALGOR software to create a finite element model of the sterngate HPU and perform a DDAM analysis to simulate the required U.S. Navy shock qualification test.

Satisfied with the ALGOR analysis results, Lorino approved the sterngate HPU design and it was sent off to Huber's in-house fabrication shop. After the unit was fabricated and assembled, it was shipped by truck to the Hi-Test Laboratories facility for the required U.S. Navy shock qualification test. The test involved five underwater explosions to simulate various combinations of shock in the fore and aft, athwartships and vertical directions. "We were looking for our equipment to pass the test on the first try, and it did. The units didn't have any structural failures," commented Lorino.

He added, "The only problems were with some lock washers being crushed and bolt heads shearing off. But, these items were corrected during the test procedure in between the first three explosions. We didn't want to simulate details as fine as the washers, bolts and so forth with FEA. Rather, we used FEA to simulate the overall structural response. Doing so helped us to save time and money."

Prior to this project, Lorino had not used FEA before. "I took the ALGOR introductory training course," he said, "and I downloaded and used several of the keystroke-specific tutorials on ALGOR's web site. It was a learn-as-you-go process. ALGOR's training materials quickly got me started in the right direction."

For other projects, Lorino has used ALGOR's InCAD Designer/Autodesk Inventor software for direct CAD/CAE data exchange and full associativity with each design change. "It's very convenient to build a model with thin parts in Autodesk Inventor, bring it into ALGOR and automatically generate a plate element model. I can easily vary the mesh in ALGOR until I converge on the optimal mesh," he said. In the future, Lorino plans to use ALGOR's new NASTRAN products. "Other companies use NASTRAN to design parts that interface with our fluid power systems. ALGOR's NASTRAN products make it possible for us to work with their NASTRAN files within the same familiar interface that we use when performing design and analysis with ALGOR."

	ALGOR's DDAM analysis results for the sterngate HPU model verified that the unit would pass the shock test. This image shows von Mises stress contours resulting from shock loading in the vertical direction only.