Nondestructive Evaluation in the Aircraft Industry
Wednesday 4 June, 10:20--12:10
- Organizer: Bill TUCKER
- General Electric/CRD
- Moderator: Bill TUCKER
- General Electric/CRD
- James P. FULTON, Tim K. KEYES
- General Electric/CRD, Schenectady NY, USA
NDE methods are very important to the manufacture and maintenance of
aircraft engines. In this regard, it is critical to find various
surface and subsurface defects. This talk will discuss the development
of improved ultrasonic (UT) methods used for the detection of subsurface
defects and analytical methods to derive improved estimation procedures
for determining probability of detection (POD) and probability of false
alarm (PFA) curves associated with the inspection methods. It is
important to find and remove defects as early as possible in the
production process. To this end, the development of a multizone UT
system for the inspection of titanium billets will be covered. A
multizone system has multiple transducers focused at increasing depths,
which gives a much improved detection capability. The quantification of
use of such a system to mitigate risk is determined by the resulting POD
and PFA curves. The development of improved estimation procedures for
POD and PFA based on signal-to-noise ratio concepts also will be
discussed. This work is sponsored by the Engine Titanium Consortium
under the auspices of the Federal Aviation Administration's program on
Aging Aircraft.
[James P. Fulton,
General Electric/CRD,
KW-D244, PO Box 8,
Schenectady NY,
USA
12301]
- Floyd W. SPENCER
- Sandia National Laboratories, Albuquerque, NM, USA
In the maintenance of aging aircraft it is important to be able to find
fatigue cracks in aircraft skins. Eddy current nondestructive
inspection techniques are the method of choice for directed inspections
in areas of an aircraft likely to develop fatigue cracking. Cracks that
develop in skins beneath countersunk rivet heads are of specific
concern, as these cracks will be "hidden" in the early stages of their
growth. Results of validation efforts to quantify the capabilities of
various existing and newly developed eddy current techniques will be
reported. The validations consisted of inspections of test samples with
cracks known only to the experimenter and not to the inspector.
Regression techniques were used to fit signal data obtained against the
known crack lengths. The results were used to derive probability of
detection (PoD) curves as functions of crack length. Inspection factors
influencing the capability estimates will be discussed. This work is
sponsored through the Aviation R & D Program of the Federal Aviation
Administration's William J. Hughes Technical Center.
[Floyd W. Spencer, MS 0829, Sandia National Laboratories, PO Box 5800,
Albuquerque, NM 87185-0829, fwspenc@sandia.gov]
Back to JRC program
Back to JRC Home Page