T.O. 33B-1-1
4-52
increases. This decreases the signal-to-noise ratio, making it more difficult to observe the small flaw indication. The
decrease in signal-to-noise ratio lowers the reliability of the inspection. Therefore, an increase in gain will increase the
amplitude of the flaw signal as well as increase the level of noise. Thus, useful sensitivity must be measured in relation
to the noise of the test system.
4.5.2.10
Influence Of Frequency On Noise.
Increasing the operating frequency for eddy current inspection improves the sensitivity to near-surface defects but also
tends to increase noise from surface related factors such as lift-off scratches, rough surface, and probe wobble.
4.5.3
Lift Off Effects.
4.5.3.1
Sources Of Lift Off Variations.
During eddy current inspection, changes in spacing between the probe coil and the inspection surface cause variations
in test coil impedance. These changes in lift-off result from surface roughness, slight contour changes, probe wobble,
probe bounce, and inconsistent thickness of nonmetallic coatings, such a paint, primer, and anodic coatings. The
magnitude of impedance changes resulting from small amounts of lift-off variations can exceed the response from a
relatively large crack. Consequently, some means of eliminating or separating this effect must be provided.
4.5.3.2
Lift Off Suppression.
One option for eliminating lift-off effects from the variable to be measured is the use of impedance plane analysis,
where the phase direction of the response from the desired variable is separated from the phase direction of signals
caused by lift-off variations. This type of analysis can be performed using any of the waveform display instruments that
provide amplitude and phase of the signal. The small, meter readout type battery-powered instruments provide only a
total amplitude measurement and require some means of lift-off suppression. For these instruments, lift-off
compensation is obtained by selection of an off null operating point. The off null operating point is selected to provide
equal current flow (meter reading) with the probe on bare metal and at a designated amount of liftoff. Eddy current
inspection using small amounts of lift-off compensation or adjustment is also termed intermediate layer technique. The
amount of lift-off adjustment is selected to minimize any surface roughness or variation in coating thickness on the
part.
4.5.4
Lift Off Compensation Methods.
4.5.4.1
Impedance Plane Analysis Instruments.
Instruments that present the phase and amplitude of the signal on a CRT have phase rotation controls which allow the
eddy current signal to be rotated until the phase is in a particular orientation. For instance, the phase can be rotated
until the lift-off signals move in a horizontal motion, with increasing lift-off represented by movement to the left on the
screen. Flaw signals or loss of conductivity will generally be in a vertical direction. The phase angle and amplitude of
an indication will depend upon the depth of the flaw and the frequency of the test.
4.5.4.2
Meter Type Instruments.
Meter type instruments utilize frequency selection, off-null settings, and other electrical compensation procedures to
minimize lift-off. The frequency selected provides the same meter response when the probe is in contact with bare
metal or separated from the metal by a nonconductive shim (usually paper or plastic) equal in thickness to the expected
maximum variation in gap between the probe and the metal. For a specific amount of lift-off adjustment, more than
one frequency or operating point may be available. For maximum sensitivity to surface or near surface cracks, the lift-
off compensation point occurring at the highest frequency should be used.
4.5.4.3
Lift Off Effects On Sensitivity.
As lift-off increases, sensitivity of the eddy current system decreases. The magnitude of the response from a crack or
other defect decreases continuously as the distance between the cracked metal and the probe increases. The typical
effect of increasing lift-off on crack response is shown in Figure 4-42. The magnitude of the total response obtained
from two cracks is plotted against the controlled thickness of an intermediate layer between the probe and the part.
