2. Technical Discussion

2.1 The Materials Selection and Design of the Test Cylinders

Several different phases of tests were conducted to demonstrate the effectiveness of the robust lightweight technology. First, an orthogonal fractional factorial test matrix was organized and populated with composite over wrapped cylinders. Variables in this test matrix included varying lay-ups, different percentages of Zylon, varying ways of intermixing the Zylon with the carbon, and varying overall composite thickness. Also fabricated in this phase, were all-carbon and all-Zylon cylinders for comparison purposes. Finally, verification cylinders were fabricated as well in order to assure that the correct principles were being applied in an "optimized" cylinder. In this case, an "optimized" cylinder means a cylinder that applies the right combination of principles that shows the most promise of increased performance. This first set of cylinders shall subsequently be referred to as Test Matrix Phase I. In the next phase, several cylinders of the "optimum" design were built and tested. These cylinders incorporated the "optimum" design, but used different types of carbon combined with Zylon. This set of cylinders shall be referred to as Test Matrix Phase II. Finally by applying all of the principles learned in the previous phases, an SCBA cylinder was designed and fabricated. These SCBA cylinders shall be referred to as Test Matrix Phase III.

2.2 Test Matrix Phase

This first test matrix was designed to provide enough data to optimize a hybrid design.

2.2.1 Material and Liner Selection

Toray's T-1000 was initially chosen to be hybridized with Toyobo's Zylon, because of the similarities of the delivered fiber strengths of the two fibers as observed in previous applications by HEI and because of similar moduli of elasticity of the two fibers. A 9.0-liter water volume SCBA-size aluminum liner manufactured by Hydrospin, Inc. of Huntington Beach California was used in all of the cylinders from this Test Matrix Phase I.

2.2.2 Design of Cylinders

Following the 1/2 orthogonal fractional factorial experiment layout, as developed by Genichi Taguchi (3), several cylinders were designed. The parameters to be varied were selected as being lay-up sequence, percentage of Zylon, manner of intermixing Zylon, and composite thickness. As is typical with the orthogonal array, variations of each parameter were given a "high" and a "low" and three cylinders of each configuration were filament wound. The following table shows the layout of the orthogonal test matrix.

At least three cylinders of all carbon and three cylinders of all Zylon were also designed to be approximately the same as those in the orthogonal matrix. These cylinders were to be used for comparison purposes.

After evaluating the different variables in the orthogonal test matrix an "optimum" design showing the most promise was discovered and three sets of three cylinders each were wound to verify that the design performed well. Each set used a different percentage of Zylon.

All Test Matrix Phase I cylinders were designed to fail in the sidewall region of the cylinder in a hoop failure mode.

2.2.3 Fabrication

All of the cylinders in this test matrix and subsequent test sets were wound at HEI's composite development lab in Brigham City, Utah using Entec Composite filament winding machines. The cylinders were filament wound with four tows with 8-10 lbs of tension each. A description of the cylinders fabricated in this Phase I can be seen in the following chart.

2.2.4 Testing

Low-velocity impact testing was conducted on these Phase I cylinders. One cylinder from each test set received no impact and as such was a "virgin" test. One cylinder from each set was impacted at an energy level of 6.8 J, and the remaining cylinder from each set was impacted at an energy level of 13.6 J. Since six cylinders of the T-1000 set and six cylinders of the Zylon set were fabricated, one more cylinder of each set was tested at 6.8 J and 13.6 J, while the third cylinder was tested at 27.1 J. The impacted cylinders were impacted by filling them ? full of water, laying them on their sides onto concrete, assembling a drop test apparatus around them, and dropping a 4.5 kg steel cylindrical test "impact torpedo" from an appropriate height directly onto the sidewall from above. The following schematic demonstrates the test setup.

Schematic of Drop Test Setup

All of the cylinders were subsequently hydrostatically burst tested at HEI's test facility and results were obtained.

2.3 Test Matrix Phase II

The main purpose of this second test matrix was to discover if certain types of carbon fiber were more compatible with the hybridizing technique than were others.

2.3.1 Material and Liner Selection

Several different carbon fibers ranging from aerospace to commercial grades were chosen and hybridized with Zylon. The same 9.0-liter water volume Hydrospin liners used in Test Matrix Phase I were used in this test matrix.

2.3.2 Design of Cylinders

The design of the composite pressure vessels used in this second test matrix incorporated the principles that were learned from the first test matrix and as such was an "optimized" design. The cylinders incorporated the optimized lay-up sequence, the optimized percentage of Zylon, the optimized manner of intermixing Zylon, and the optimized composite thickness that performed the best. All the cylinders were designed to fail in a hoop failure mode in the sidewall region of the cylinders.

2.3.3 Fabrication

Four cylinders from each test configuration were wound. The following chart describes the different test sets that were fabricated in this test matrix.

2.3.4 Testing

The cylinders in this test matrix were tested similar to those in the Test Matrix Phase I. One cylinder from each set was set aside to be "virgin" burst tested, while the remaining cylinders from each set were individually impact tested at 6.8, 13.6, and then 27.1 J. Following impact testing the cylinders were hydrostatically burst tested and results were recorded.

2.4 Test Matrix Phase III

The purpose of the third set of tests was to demonstrate that the principles learned from previous testing also applied well to an actual SCBA design.

2.4.1 Material and Liner Selection

The liner that was selected for this set of tests was a 6.8-liter water volume SCBA liner manufactured by Samtech International Inc. of Carson, California. It was desired to have a "state of the art" composite pressure vessel and as such Toray's aerospace T-1000 was chosen and hybridized with Zylon.

2.4.2 Design of Cylinders

The design of this SCBA cylinder was based upon HEI's years of experience in designing composite pressure vessels. It was designed to have a service pressure of 30 MPa (300 bar) and to meet the European EN 12245 specification. The cylinders were designed to fail in a hoop failure mode in the sidewall region. Significant effort was given to create no-fragment hydro-burst.

2.4.3 Fabrication

Two basic sets of designs were used in this Phase III test matrix. Both sets incorporated optimum design characteristics learned from the previous tests discussed earlier. The first set also included an E-glass over-wrap as is typical in the composite SCBA industry. The second set was not over-wrapped with any E-glass. The following chart describes the different tests to be evaluated in this third phase of tests.

2.4.4 Testing

The goal in this phase was to get an SCBA cylinder that could be qualified under the EN12245 specification. The first test conducted on the SCBA cylinders was that of a hydrostatic burst. Following a successful burst test, a .30 Caliber gunfire test was performed. The cylinder with the E-Glass over-wrap fragmented into two pieces. Extreme effort was put into causing the burst without E-Glass to be fragment free. The successful hydrostatic bursts can be seen in the following photos.

Hybrid With Glass Hydro-burst	Hybrid without Glass Hydro-burst

As required by the European Norm 12245, in order for a cylinder to pass the .30 caliber gunfire test, it must first be pneumatically pressurized to service pressure (300 bar in this case) and a .30 caliber armor piercing round meeting certain speed and distance requirements must penetrate the cylinder at a 45Á angle to the centerline entering the dome and proceeding towards the sidewall portion of the cylinder. In order to pass, the cylinder must show no evidence of fragmentation. Typically thin walled very high performance cylinders have an extremely difficult time passing this test, because of their sensitivity to impact damage. A photo demonstrating the test setup can be seen below.

30 caliber gunfire test setup

Copyright 2004 by HyPerComp Engineering, Inc. Published by Society for the Advancement of Material and Process Engineering with permission.

<--Previous Abstract Introduction Technical Discussion Test Results Conclusion and References Next-->