|
Other SAE Magazines Sites
|
Optimizing composite inspections
-
Composite reference standards will facilitate nondestructive test equipment calibration.
Information for this article was provided by Dennis Roach and
Larry Dorrell, FAA Airworthiness Assurance NDI Validation Center, Sandia National Laboratories; Jeff Kollgaard, Boeing Commercial Aircraft Co.; and Tom Dreher, United Airlines.
As the demand for composite materials
technology increases, inspection of
composite structures remains critical to ensure their continued airworthiness. The FAA's Airworthiness Assurance Nondestructive Inspection Validation Center and the Commercial Aircraft Composite Repair Committee (CACRC) are developing a set of composite reference standards to be used in calibration of nondestructive test (NDT) equipment for performing damage assessment and post-repair inspection of commercial aircraft composites. For global acceptance, the standards must incorporate the necessary structural configurations of many aircraft. They must also be representative of damage found in the field and include typical flaw scenarios such as disbonds and delaminations. Also under evaluation is the workable number of reference specimens for optimum evaluation. Standards will encompass both composite-honeycomb and -laminate configurations.
The composite-honeycomb task began by engineers fabricating a set of 64 specimens to isolate the effects of the following variables and boundary conditions on NDT:
- laminate material (carbon; fiberglass)
- honeycomb core material
(Nomex; fiberglass)
- laminate thickness (3 plies; 12 plies)
- honeycomb core thickness (0.25 in.; 2 in.)
- honeycomb cell size (0.125 in.; 0.25 in.),
- honeycomb core density (2-8 lb/ft3)
- disbond and delamination flaws
Figure 1. Composite honeycomb panel containing four different construction types and engineered flaws.
|
Sixteen panels used in the study contained four different construction types and isolated the effects of each variable (Figure 1). Nondestructive inspection (NDI) was applied to the specimens to assess the difficulties presented by the engineered flaws. The inspection results were used to identify the key variables that should be included in composite honeycomb reference standards.
The NDI methods and equipment used were low-/high-frequency bond testers (S-9 Sondicator, Bondmaster, and MAUS in resonance mode), through-transmission and pulse-echo ultrasonics (Staveley 136 and MAUS in PE mode), tap test (Mitsui Woodpecker), thermography (Thermal Wave Imaging), and mechanical impedance analysis (MIA-3000, V-95 Bondcheck). To allow comparison of results from different NDI methods, which use varying indicators to infer the presence of defects, each inspection was measured using the signal-to-noise ratio of each defect versus the surrounding structure. An unchanged signal-to-noise ratio indicates that increasing cell size has no effect on defect detectability. Reference standards must have skins that closely represent the structure to be inspected.
Evaluation of NDT data led to the production of a prototype minimum-reference standard set that includes the important variables for inspecting composite honeycomb structures—laminate thickness, laminate type, and honeycomb type. Table 1 summarizes the construction characteristics of the prototype honeycomb set. Eight standards were manufactured—a 3-,
6-, 9-, and 12-ply laminate with carbon or fiberglass skins, with both Nomex and fiberglass cores. Figure 2 illustrates the basic honeycomb design approach.
Figure 2. Basic honeycomb standard design sample.
|
Signal-to-noise results from the panels indicated acceptable flaw detection over the entire range of honeycomb types. Researchers also verified that an exact laminate thickness match was necessary for accurate test results. Structures containing 12 or more plies did not provide consistent inspection results using bond testers, pulse-echo ultrasonics, and mechanical impedance analysis.
Several of the NDI tests revealed inconsistencies in the flaw manufacturing methods. Pillow-insert flaws were used because they were thought to provide realistic responses. However, engineers determined that the response from the disbonds and delaminations engineered with pillow inserts sometimes did not provide sufficient deviation from the noise floor to allow for clear flaw detection. Inspection results indicated that machining the honeycomb core (recessing) away from the laminate provides the best method for producing reliable skin-to-core disbond flaws. It also produces flaw sites that can support tap testing.
The remaining question was how best to produce interply delamination flaws. To answer this question, two trial standards were made that included three candidate methods for engineering delamination flaws. Figure 3 shows the engineering drawing for these honeycomb specimens; one carbon and one fiberglass skin specimen was produced with this flaw layout. The three methods used to make the delamination flaws were: a pillow insert consisting of Kapton tape around four layers of tissue paper, brass shims coated with a silicon mold release to prevent bonding to the plies, and Teflon inserts. Each method was used to generate three similar delamination flaws to test for repeatability and determine the amount of NDI signal disruption generated. Each specimen includes potted core and core splice areas to aid in interpretation of NDI signals.
Figure 3. Flaw layout for honeycomb reference standard evaluation.
|
Future tasks are planned to complete validation of the minimum honeycomb NDI reference standard set. One task includes finalizing the standard fabrication process by determining the optimum way to engineer flaws. The S-9 Sondicator, Bondmaster, and through-transmission ultrasonics in an immersion tank will be used for this assessment. At least 18 dB will be required at the flaw sites. Following this task, the 64 aircraft panels will be inspected for the honeycomb reference standards. Then the prototype honeycomb reference standard set will undergo field testing with United and Northwest Airlines. Lastly, the design will be optimized to minimize the overall size of each standard and achieve the fewest number of separate honeycomb standards. The final specimen size must accommodate probe deployment on both good and flawed structure and eliminate any edge effects or effects from adjacent flaws.
The goal of the composite-laminate task was to establish a single, generic composite laminate reference standard that will accommodate inspections on the full array of fiberglass and carbon laminates found on aircraft. Researchers would prefer to substitute a single material for both carbon and fiberglass solid laminate inspections; however, the material would need to provide the same response for both. In addition, to improve on existing solid laminate standards, the material should be inexpensive, reliably manufactured, and easy to machine into a solid laminate standard.
The first step in the task was to apply through-transmission ultrasonics to the series of existing laminate specimens (step wedges of various materials at different thicknesses) to measure the key velocity, acoustic impedance, and attenuation characteristics in the laminates. The laminate standard design includes thickness ranges from one ply (0.010 in.) to 1.0 in. A material search identified what appears to be a good candidate as a generic solid laminate reference standard material. Testing thus far has determined matches in key velocity and acoustic impedance properties, as well as low attenuation relative to carbon laminates. Additionally, comparisons of resonance-testing response curves from the G11 phenolic prototype standard was similar to the resonance response curves measured on the existing carbon and fiberglass laminates. Resonance tests on three carbon composite standards revealed that variability across similar standards was similar to the variability observed between G11 and carbon or fiberglass. Table 2 presents velocity, acoustic impedance, and attenuation characteristics for the candidate laminate standard materials.
Future activities are planned for a final assessment of the G11 material's suitability as a generic solid laminate standard. Personnel from Boeing's NDT Engineering Department will use the generic standard to support inspections on various in-house solid laminate aircraft structures with flaws. Additional insights will be obtained from industry and aircraft inspectors. Solid laminate inspection procedures will be examined to determine if any modifications or additions are necessary to accommodate the use of a generic laminate standard. Issues to be addressed regarding optimizing the generic laminate design include minimization of size and weight, ease of machining and handling, and ease of use in the field.
Following final validation, field testing, and design optimization on both solid laminate and honeycomb reference standards, formal modifications to appropriate OEM manuals will be addressed. Active participation of the OEMs provides a harmonized approach worldwide. The end result will be more streamlined inspection setups for aircraft maintenance depots and improved inspections using optimized NDI reference standards.
|
|
|
