DOT/FAA/CT-94/79 FAA Technical Center Atlantic City International Airport, N.J. 08405

Evaluation of Scanners for C-Scan Imaging for Nondestructive Inspection of Aircraft Author: John H. Gieske Sandia National Laboratories Airworthiness Assurance NDI Validation Center

October 1994 Final Report U.S. Departrnent of Transportation Federal Aviation Administration. NOTICE

Three companies submitted an update related to the Sandia Scanner Survey.
1. ULTRA IMAGE INTERNATIONAL
2. ABB AMDATA with its new Catamaran Scanner
3. Panametrics

ABSTRACT

The goal of this project was to produce a document that contains information on the usability and performance of commercially available, fieldable, and portable scanner systems as they apply to aircraft nondestructive inspections. In particular, the scanners are used to generate images of eddy current, ultrasonic, or bond tester inspection data. The scanner designs include manual scanners, semiautomated scanners, and fully automated scanners. A brief description of the functionality of each scanner type, a sketch, and a list of the companies that support the particular design are provided. Vendors of each scanner type provided hands-on demonstrations of their equipment on aircraft samples in the Federal Aviation Administration (FAA) Aging Aircraft Nondestructive Inspection Validation Center in Albuquerque, New Mexico. From evaluations recorded during the demonstrations, a matrix of scanner features and factors and ranking of the capabilities and limitations of the design, portability, articulation, performance, usability, and computer hardware/software was constructed to provide a quick reference to compare the different scanner types. Illustrations of Cscan images obtained during the demonstration are shown.

PREFACE

This work was performed at the Federal Aviation Administration (FAA) Aging Aircraft Nondestructive Inspection Validation Center (AANC) in Albuquerque, NM, and sponsored by the FAA Technical Center in Atlantic City, NJ. The author was able to conduct the scanner evaluations through the cooperation of vendors who participated in the demonstrations. The author expresses his appreciation for the time, effort, and expense the vendor participants and sales representatives incurred while performing the hands-on demonstrations. The author also thanks other members of the AANC staff for their support.

TABLE OF CONTENTS EXECUTIVE SUMMARY INTRODUCTION Background, Goal, Objectives DEMONSTRATION AND EVALUATION PROCESS Scanner System Design Identification Choosing Vendors for Hands-on Demonstrations Aircraft Samples Steps of the Performance Demonstration SCANNER EVALUATION MATRIX Evaluation Matrix Contents Table 1 (sep. page): Scanner Evaluation Matrix for Eddy Current and Ultrasonic C-Scan Imaging of Inspection Results EXAMPLE C-SCAN IMAGES DISCUSSION CONCLUSIONS APPENDICES .... A - SCANNER DESCRIPTIONS (sep. page) B - LIST OF VENDORS WHO PARTICIPATED IN THE DEMONSTRATIONS (sep. page) C - AANC AIRCRAFT SAMPLE DESCRIPTIONS (sep. page) D - EVALUATION MATRIX FEATURES AND FACTORS RANKING CRITERIA (sep. page) E - EXAMPLES OF C-SCAN IMAGES (sep. page) F - GENERAL COMMENTS ON THE SCANNER TYPES (sep. page) NOMENCLATURE

APPENDICE E - EXAMPLES OF C-SCAN IMAGES Krautkramer, Branson, Hocking with Paul Martin articulating the manual scanner

EXECUTIVE SUMMARY

Fieldable nondestructive inspection (NDI) systems based on eddy current and ultrasonic inspection methods, and utilizing scanners to produce images, have been used in the nuclear and petrochemical industry for years to detect cracks, corrosion, and disbands. Similar systems have the same potential in the airline industry for early detection of hidden damage in aircraft structures.

The images produced by the scanning systems mentioned above are called C-scans. C-scans are 2-D images produced by digitizing the point-by-point signal variations of an interrogating sensor while it is scanned over a surface.

To provide the encoded sensor position for the computer during C-scan imaging, a number of portable scanner designs and scanner methodologies have been developed in recent years. Both manual and automated portable scanners have been developed that may be useful for aircraft NDI.

The goal of this project was to produce a document that contains information on the evaluation of scanner systems as they apply to aircraft inspections.

From a literature survey and discussions with vendors, a variety of different portable scanner designs were identified. The designs include manual scanners, semiautomated scanners, and fully automated scanners. Scanners included both mechanized and nonmechanized designs.

The basic scanner designs were divided for the purposes of this report into eight different types. These are dual axis, tilting arm and bridge, manual (mechanized); dual axis, tilting arm and bridge, automated (mechanized); radial axis, tilting arm with rotation axis bridge, manual (mechanized); dual axis, cantilever arm bridge, manual and automated (mechanized); mobile, automated, ultrasonic scanner (mechanized semiautomated); dual axis, rectangular bridge, automated (mechanized); hands free x-y digitizer (nonmechanized acoustic or video tracking); and square transducer array (nonmechanized electronic switching).

Appendix A includes a brief description of the functionality of each scanner type, a sketch, and a list of the companies that support the particular design.

Vendors provided hands-on demonstrations of their equipment on aircraft samples in the Federal Aviation Administration (FAA) Aging Aircraft Nondestructive Inspection Validation Center (AANC) in Albuquerque, NM. The aircraft samples and the Boeing 737 (B737) airplane used in the demonstrations contained known areas of corrosion damage and disbands from in-service conditions. Capabilities and limitations of the design, portability, articulation, performance, usability, and computer hardware/software were recorded during the demonstrations.

From observations and information recorded during the demonstrations, a matrix of features, factors, and their respective evaluations for each scanner tested was constructed to provide a quick reference for comparing the different scanner systems. A table containing the evaluations and ranking of each feature or factor for the scanners demonstrated is provided.

No attempt is made to rank the scanner systems overall with comparative scores. This is left to potential users. The users should consider features and factors that are most important for their respective applications.

Excellent C-scan images of eddy current and ultrasonic inspection data were obtained during the performance demonstrations. Illustrations of the C-scan images obtained from examinations of the five AANC aircraft samples used in the evaluation are shown in Appendix E. Pictures of the attachment of a number of the scanners on the B737 airplane are also included.

A discussion of the strong points and weak points of the eight scanner types is given in Appendix F. Suggestions for improvements are also provided there.

INTRODUCTION

BACKGROUND
Fieldable nondestructive inspection (NDI) systems based on eddy current and ultrasonic inspection methods, and utilizing scanners to produce images, have been used in the nuclear and petrochemical industry for years to detect cracks, corrosion, and disbands. Similar systems have the same potential in the airline industry for early detection of hidden damage in aircraft structures. Corrective repairs initiated by early detection of damage can be cost effective by reducing the need for subsequent major repairs that impact the availability of the aircraft for revenue.

Another possible application area aside from aluminum structures is composites. New aircraft rely increasingly on composite technology. Periodic inspections of a composite structure for delaminations and impact damage during the service life of the aircraft are essential for safety. Ultrasonic imaging of composites has the potential to provide the inspection data needed to detect these defects and assess the structural integrity of the composite during the life of the aircraft. Thus, imaging technology is applicable to both new and aging aircraft.

The images produced by the scanning systems mentioned above are called C-scans. C-scans are 2-D images produced by digitizing the point-by-point signal variations of an interrogating sensor while it is scanned over a surface. The X-Y position of the sensor is recorded simultaneously with the signal variations. A computer converts the point-by-point data into a color representation and displays it at the appropriate point in an image. This image usually makes it much easier to interpret defects than the individual measurements.

To provide the encoded sensor position for the computer, a number of portable scanner designs and scanner methodologies have been developed in recent years. Both manual and automated portable scanners have been developed that may be useful for aircraft NDI.

GOAL
The goal of this project was to produce a document that contains information on the evaluation of scanner systems as they apply to aircraft inspections. The document is based on demonstrations of commercially available, portable inspection systems that were observed while scanning representative aircraft structures.

OBJECTIVES The objectives of this project were to: Demonstrate and evaluate the capability of commercially available portable scanner systems to generate C-scan images on representative aircraft structures. Evaluate the usability and performance of the different scanner types to help inspection personnel choose an appropriate scanner and to help scanner vendors to improve the usability and performance of the scanners for aircraft inspection requirements.

DEMONSTRATION AND EVALUATION PROCESS

SCANNER SYSTEM DESIGN IDENTIFICATION
From a literature survey and discussions with vendors, a variety of different portable scanner designs was identified. The designs include manual scanners, semiautomated scanners, and fully automated scanners. Scanners included both mechanized and nonmechanized designs.

All mechanized scanners employ optical encoders on one or more of the moving parts of the scanner to indicate the sensor position. Nonmechanized scanners employ diverse techniques to encode the sensor positions. An example of a nonmechanized scanner involved transmitting a high frequency acoustic pulse at the sensor from a distance and detecting the propagating pulse through the air with a pair of microphones. The position of the sensor is calculated by triangulation techniques from arrival time data. Another example employed a light-emitting diode (LED) at the sensor with a video camera encoding system positioned above the sensor for tracking and coding the position of the sensor. A third example employed a 2-D array of small transducer elements embedded in a flexible vacuum blanket that is applied in contact with the surface; the C-scan image is formed by electronic switching through the transducer elements of the array.

The basic scanner designs were divided for the purposes of this report into eight different types. These are dual axis, tilting arm and bridge, manual (mechanized); dual axis, tilting arm and bridge, automated (mechanized); radial axis, tilting arm with rotation axis bridge, manual (mechanized); dual axis, cantilever arm bridge, manual and automated (mechanized); . mobile automated ultrasonic scanner (mechanized semiautomated); dual axis, rectangular bridge, automated (mechanized); hands free x-y digitizer (nonmechanized acoustic or video tracking); and square transducer array (nonmechanized electronic switching).

These scanner types are described in Appendix A. Each entry in Appendix Aincludes a brief description of the functionality of each scanner type, a sketch, and a list of the companies that support the particular design.

CHOOSING VENDORS FOR HANDS-ON DEMONSTRATIONS
Vendors provided hands-on demonstrations of their equipment on real aircraft samples in the Federal Aviation Administration (FAA) Aging Aircraft Nondestructive Inspection Validation Center (AANC) in Albuquerque, NM. The aircraft samples and the Boeing 737 (B737) airplane used in the demonstrations contained known areas of corrosion damage and disbands.

Capabilities and limitations of the design, portability, articulation, performance, usability, and computer hardware/software were recorded during the demonstrations for later evaluation.

For each scanner type, vendors were contacted and performance demonstrations of their equipment were discussed. If the vendor was receptive and volunteered to conduct the hands-on demonstrations, arrangements were made to perform the demonstrations in the FAA/AANC hangar. Each vendor was asked to bring its own eddy current, ultrasonic, and bond tester equipment to be used with the scanners. Multimode scans using the different NDI techniques could then be evaluated at the same time. In some cases, two demonstrations were scheduled at different times for a given NDI technique when two different vendor representatives of the respective techniques were involved.

Priority was given to vendors who supported both eddy current and ultrasonic testing equipment with their scanner systems. Their integrated system would have the best chance of performance and largest potential payback for providing significant information on the capabilities and limitations of the scanner design type for the different NDI modes.

The scope of this project was to evaluate all the basic scanner types that are appropriate for aircraft NDI examinations. A number of vendors sell very similar scanners of the same basic design. They have integrated the scanner with their data acquisition and software system but they do not sell NDI equipment. In this case, only one or two scanners of the same basic design were evaluated with vendors who offered the most integrated NDI capability. It is expected that similar results would be obtained with other scanners of the same basic design.

A list of the vendors and participants who took part in the performance demonstrations is provided in Appendix B.

AIRCRAFT SAMPLES
The demonstrations were performed on a group of samples that represented defects from in-service conditions. Samples with lap splice joint corrosion, various surface conditions and thickness, and disband conditions were chosen. Also, various geometric configurations on the B737 aircraft where the scanner must be mounted in a vertical or upside down (overhead) position were chosen for the evaluation.

Five samples used in the evaluation were:

1. A..D. Little aluminum lap splice joint intergranular corrosion attack specimens of 0.04-inch thickness. (AANC Test Specimen Library Numbers 115 through 122)
2. Large 0.07-inch-thick aluminum panel with visible pitting and intergranular/exfoliation corrosion and pillowing of the surface. (AANC Test Specimen Library Number 111)
3. Calibration standards used for setting up tests for circumferential tear strap disband. (AANC Test Specimen Library Numbers 183 through 185)
4. Textron Specialty Materials boron/epoxy composite repair sample with implanted disbands and delaminations on an aluminum skin. (AANC Test Specimen Library Number 152)
5. Various locations on the B737 AANC aircraft with disbands and corrosion. (AANC Test Specimen Library Number 100)
Detailed descriptions of these samples are given in Appendix C.

STEPS OF THE PERFORMANCE DEMONSTRATION
The performance demonstration of each scanner by the vendor was conducted with the following steps:

1. An overview of the AANC activities was provided to the vendor by a member of the AANC staff.
2. A calibration and initial setup of the equipment were performed on flat horizontal samples on a table top to become familiar with the equipment. C-scan images were generated to demonstrate the general capabilities of the system and scanner operation. At this time, an overview of the equipment hardware and software capabilities was provided by the vendor.
3. For the first eddy current test, the A.D. Little intergranular corrosion attack samples of a lap splice joint of 0.04-inch skin thickness were examined. C-scan images of the hidden corrosion over the 12-inch length joint was recorded and saved in a data file on the computer system. The A.D. Little samples were used to observe the general operation and function of the scanner and observe the effort the examiner needed to obtain meaningful C-scan images representing the areas of corrosion damage.
4. For the second eddy current test, the large panel with skin thickness of 0.07 inch was scanned. This panel contained visible corrosion and pillowing of the surface between the rivet locations. This panel was used to demonstrate how well the scanner functions on wavy and rough surfaces representative of significant pillowing. The time to scan and obtain meaningful C-scan images of the corrosion for an area of 4 and 8 square inches was recorded.
5. The scanner with the eddy current sensor was then attached to the B737 airplane where a demonstration was conducted at the area bounded by body station (BS) 877 and BS 887 and stringers (S) 22R and S 24R. The scanner must be positioned somewhat vertical and upside down to perform this test. This area of the airplane had corrosion and a tear strap disband could be seen by viewing the interior panel surface. This test demonstrated the ability of the scanner and effort required by the examiner to take inspection data on a curved surface and with the scanner in an upside down or overhead position. The scan time to produce a C-scan image of the inspection data was recorded for this area. Notes as to the operation of the equipment under these conditions were recorded and C-scan images of the inspected areas showing the detected damage were saved for comparison with different scanner systems. If time permitted, C-scan images were also obtained at the lap splice joint at S 20R and the butt joint at BS 907. A final test was made on the airplane at the lap splice joint S 10L on the left side of the airplane above the windows between BS 817 and BS 907. The equipment must be carried up a scaffold and attached to the fuselage above the windows for this last eddy current test. This exercise provided information on the portability of the scanner.
6. For the first ultrasonic pulse-echo and resonance evaluations, area scans with C-scan images were made on the tear strap disband calibration standards and the Textron boron/epoxy repair patch calibration standard. These samples were used to observe the general operation of the scanner for ultrasonic inspections. If the vendor also had a bond tester capability, then data were also obtained with the bond tester sensor. If a bond tester was not available, the pulse-echo technique was set up to simulate the resonance bond testing technique to obtain the inspection data.
7. The scanner with the ultrasonic sensor was then attached to the B737 airplane at BS 877 at S 22R. This is the same area on the airplane where eddy current evaluations were accomplished. An ultrasonic scan in this area where corrosion and tear strap disbands have occurred demonstrated the ability of the scanner and effort required by the examiner to take ultrasonic inspection data for the vertical and overhead position of the scanner. The functionality of the scanner under these conditions to maintain ultrasonic couplant and sensor perpendicularity to the surface was observed. The scan time to produce the C-scan image of the inspection data was recorded. C-scan images of the inspected area were saved for comparison of the detected damage with different scanner systems.

Later in the program, a boron/epoxy repair patch was placed on the airplane and on a large lap splice joint fatigue panel. Demonstrations of ultrasonic resonance techniques on these repair patches were made when these became available. This provided additional information on the effort and effectiveness of ultrasonic C-scan imaging for assessing the integrity of the repair patches.

If not made during the demonstration, hard copy images or image files of the C-scans were obtained from the vendor so that copies of the images could be compared at a later date.

SCANNER EVALUATION MATRIX

EVALUATION MATRIX CONTENTS
As a result of the observations and information recorded during the demonstrations, a matrix of features, factors, and their respective evaluations for each scanner tested was constructed to provide a quick reference for comparing the different scanner systems. Table 1 contains the evaluations and ranking of each feature or factor for the scanners demonstrated.

The evaluations concentrated on the mechanics and efficiency of the scanner to provide XY position data while maintaining proper sensor orientation and articulation so that meaningful C-scan images were obtained. The matrix contains observations made by the author while witnessing the demonstrations for the different NDI methods of eddy current scans, ultrasonic pulse-echo scans, or ultrasonic bond testing scans.

Each feature or factor in Table 1 is ranked from 1 (not applicable for aircraft applications) to 5 (ideal for aircraft applications). The ranking criteria for each feature or factor is given in Appendix D. The purpose of ranking the features is meant as an aid to document observations made during the hands-on demonstrations and to differentiate them from the characteristics of the author's idea of an ideal scanner system, which is given in Appendix F. The ranking is meant to point out differences observed by the author during the hands-on demonstrations and is not meant to be a recommendation of one system over another. Each system has certain merits that may make it useful in one application but undesirable in another application. Every feature of the ideal scanner system is not attainable in any one scanner design. The characteristics of an ideal scanner are discussed in Appendix F.

All systems evaluated contained software that generated basic C-scan images. The basic C-scan images were quite adequate for aircraft applications. Some systems contained software tools for advanced image processing that could be used to enhance interpretation of a particular inspection data set. These tools are valuable, but the evaluation of the imaging tools available in the various systems was not attempted.

No attempt is made to rank the scanner systems overall with comparative scores. This is left to potential users. The users should consider features and factors that are most important for their respective applications. The evaluation of the features and factors for all of the scanners demonstrated are given in Table 1.

EXAMPLE C-SCAN IMAGES

Excellent C-scan images of eddy current and ultrasonic inspection data were obtained during the performance demonstrations. Illustrations of the C-scan images obtained from examinations of the five AANC library samples used in the evaluation are shown in Appendix E. Pictures of the attachment of a number of the scanners on the B737 airplane are also included. In some cases, the color palette of the original C-scan images was changed so that black and white reproductions of the illustrations would show the inspection results clearly. The C-scan images are provided to show the potential benefits of C-scan imaging in inspection of aircraft structures.

DISCUSSION

Commercially available portable scanners can provide excellent C-scan imaging of NDI data. The images shown in Appendix E illustrate the potential of C-scan imaging for providing quantitative measurements of hidden corrosion and disbands for aircraft applications.

Setup of the eddy current and ultrasonic equipment was done by using the experience gained from testing similar structures by the vendor representatives and the author. The parameters used may not have been optimal for quantitative NDI results especially since only limited time was available to demonstrate the equipment. Quantification of corrosion damage can be done through proper calibration procedures. The purpose of this study was to evaluate the usability and performance of scanner systems to acquire and display meaningful inspection data of corrosion damage and disbands. No attempt was made to calibrate and optimize equipment parameters or quantify the corrosion damage detected.

When applications are identified and the use of C-scan imaging is concurred by industry to be valuable for future NDI aircraft applications, then the test parameters, calibration, and test procedures must be developed and established for these defined applications. The development of these optimum test parameters, procedures, and reliability of inspection results on the variability of surface conditions, paint thickness, etc., would be the subject of possible future work for knowledgeable researchers in the field.

CONCLUSIONS

Conclusions derived from this evaluation study can be summarized as follows:
• Eddy current C-scan imaging can be implemented easily with available commercial equipment and the benefits realized immediately. Ultrasonic resonance techniques may also be implemented in the near future after experienced users have correlated results with calibration samples and gained confidence in its use. Ultrasonic pulse-echo measurements may be used only after the more experienced operators have developed a technique and procedure for each specific inspection application.
• Every mechanized scanner tested scratched the surface of the aluminum panels for both the eddy current and ultrasonic techniques. Automated scanners in some cases scratched the surface more severely since larger forces are needed to keep the sensor holder in contact and perpendicular to the surface. New designs of sensor holders are recommended that in effect would result in more nearly frictionless contact deployment. However, scanning over composite surfaces and composite repair patches with the present scanners did not damage the surface of the composite.
• Eddy current data acquisition with the mechanized scanners was more reliable and easier to obtain than ultrasonic data acquisition. This is because small couplant variations and tilt of the ultrasonic transducer influenced the inspection data significantly, whereas small lift-off variations for the eddy current sensor has little effect on the inspection data.
• Both manual and automated scanners performed well over flat rivets and over surfaces with nominal pillowing between rivets.
• Manual scanners are most useful for small area scans where surface obstructions (raised rivets) may be present. Inspection times for 1 square foot coverage ranges from approximately 10 to 20 minutes depending on the spot size resolution desired.
• Automated scanners are recommended for both small area scans and large area scans where obstructions are not present. Inspection times for 1 square foot coverage varied from approximately 5 minutes to 15 minutes. Spot size resolution is not a major concern for automated scanners since fine and large spot sizes result in approximately the same scan time.
• Hands-free digitizer scanners have the potential of being the most useful manual scanner since they are the least expensive and most versatile for areas of complex curvatures and obstructions. They may also be useful in areas on and around the stringers on the interior surface of the fuselage.
• Manual scanners are more labor intensive and tiring to operate than automated scanners especially in overhead applications. In general, manual ultrasonic C-scan imaging is very difficult to implement for overhead applications. Inspection times longer than 1 hour would be taxing on the examiner.
• Passive rubber cup suction feet are unreliable for vertical and overhead deployment. An active vacuum system either by hand pumps or an AC vacuum pump is more reliable.
• The dual axis tilting arm bridge automated scanner provided the best results for portability, performance, and overall usability of all the automated scanners tested.
• The radial axis tilting arm with rotation manual scanner provided the best results for portability, performance, and overall usability of all the manual scanners tested.
• The heads-up display used to acquire data with the dual axis XY manual scanner did not add to the performance or ease of data acquisition for C-scan imaging.
• The 2-D transducer array system performed very well resolving 0.04-inch aluminum skin thickness for possible quantitative and accurate corrosion damage assessment. Excellent resolution of the implanted delaminations in the thin boron/epoxy repair patches was obtained. Portability and usability of the 2-D array system were excellent.
• The 2-D transducer array system has great potential in the initial and periodic assessment of composite repair patches. It may also be useful for the assessment of impact damage of composite structures.
• An experienced ASNT Level 2 NDI inspector would be required to operate every one of the C-scan systems evaluated for general applications. However, an experienced ASNT Level 1 NDI inspector could operate every one of the C-scan systems with proper training and supervision by Level II or Level III inspectors in specific applications.
• Setup time from off the shelf to start of scan was reasonable (10 to 20 minutes) for all systems evaluated.
• The cost of scanner systems for C-scan imaging ranged from approximately $30,000 to $150,000.

The strong points and weak points of the eight scanner types are discussed in Appendix F. Suggestions for improvements are also provided.

NOMENCLATURE

AANC: Aging Aircraft Nondestructive Inspection Validation Center
ASNT: American Society for Nondestructive Testing
B737: Boeing 737
BS: Body station
FAA: Federal Aviation Administration
LED: Light-emittng diode
NDI: Nondestructive Inspection
S: Stringers

NOTICE

This document is disseminated under the sponsorship of the U. S. Department of Transportation in the interest of information exchange. The United States Government assumes no liability for the contents or use thereof.
The United States Government does not endorse products or manufacturers. Trade or manufacturers' names appear herein solely because they are considered essential to the objective of this report.

Author(s): John H. Gieske
Performing Organizaion Name and Addr ess:
Sandia National Laboratories, Albuquerque New Mexico 87185

Sponsoring Agency Name and Address:
U.S. Department of Transportation, Federal Aviation Administration
Technical Center, Atlantic City International Airport, NJ 08405
Report Date: September 1994
Contract or Grant No: DTFA-03-91-A-0018
Type of Report and Period Covered: Final Report
14 Sponsoring Agency Code: ACD-220
The Sponsoring Agency's Technical Officer was Dave Galella
Key Words: Aging Aircraft, Portable System, Ultrasonic, Eddy Current,
Distribution Statement:
This document is available to the public through the National Technical Information Service, Springfield, Virginia 22161

For more information see: Focus on NDT in Aerospace in UTonline 10/97