ASM Metals HandBook Vol. 17 - Nondestructive Evaluation and Quality Control

Подождите немного. Документ загружается.

These correlations between the signal amplitude and the interfacial area of the crack and between the peak-to-peak

separation and the surface length demonstrate that crack parameters can be obtained from signal features, thus

establishing the feasibility of crack characterization. The results were obtained with half-penny-shaped defects. However,

flaw characterization has also been demonstrated with other aspect ratios (Ref 3).

References cited in this section

R.E. Beissner and M.J. Sablik, Theory of Electric Current Perturbation Probe Optimization,

Rev.Prog.

Quant. NDE, Vol 3A, 1984, p 633-641

R.E. Beissner, G.L. Burkhardt, and F.N. Kusenberger, "Exploratory Development of Advanced Surface Flaw

Detection Methods," Final Report, Contract No. F33615-82-6-5020, Report No. AFWAL-TR-84-

4121,

Wright-Patterson Air Force Base, Sept 1984

C.M. Teller and G.L. Burkhardt, Small Defect Characterization by the Electric Current Perturbation Method,

in Proc. of the Thirteenth Symposium on Nondestructive Evaluation,

Nondestructive Testing Information

Analysis Center, April 1981, p 443-453

Electric Current Perturbation NDE

Gary L. Burkhardt and R.E. Beissner, Southwest Research Institute

References

1. R.E. Beissner and M.J. Sablik, Theory of Electric Current Perturbation Probe Optimization,

Rev.Prog.

Quant. NDE, Vol 3A, 1984, p 633-641

2. F. Forster and H.L. Libby, Probe Coil-Detection of Cracks in Nonmagnetic Materials, in

Nondestructive

Testing Handbook, Vol 4, Electromagnetic Testing,

R.C. McMaster, P. McIntire, and M.L. Mester, Ed., 2nd

ed., American Society for Nondestructive Testing, 1986, p 185-187

R.E. Beissner, G.L. Burkhardt, and F.N. Kusenberger, "Exploratory Development of Advanced Surface

Flaw Detection Methods," Final Report, Contract No. F33615-82-6-5020, Report No. AFWAL-TR-84-

4121, Wright-Patterson Air Force Base, Sept 1984

C.M. Teller and G.L. Burkhardt, Application of the Electric Current Perturbation Method to the Detection

of Fatigue Cracks in a Complex Geometry Titanium Part, Rev. Prog. Quant. NDE, Vol 2B, 1983, p 1203-

1217

C.M. Teller and G.L. Burkhardt, NDE of Fastener Hole Cracks by the Electric Current Perturbation

Method, Rev. Prog. Quant. NDE, Vol 1, 1982, p 399-403

6. C.M. Teller and G.L. Burkha

rdt, Small Defect Characterization by the Electric Current Perturbation

Method, in Proc. of the Thirteenth Symposium on Nondestructive Evaluation,

Nondestructive Testing

Information Analysis Center, April 1981, p 443-453

Electric Current Perturbation NDE

Gary L. Burkhardt and R.E. Beissner, Southwest Research Institute

Selected References

•

R.E. Beissner and G.L. Burkhardt, Application of a Computer Model to Electric Current Perturbation Probe

Design, Rev.Prog. Quant. NDE, Vol 4, 1985, p 371-378

• R.E. Be

issner, C.M. Teller, G.L. Burkhardt, R.T. Smith, and J.R. Barton, Detection and Analysis of

Electric-Current Perturbation Caused by Defects, in Eddy-

Current Characterization of Materials and

Structures, STP 722, F. Birnbaum and G. Free, Ed., American Socie

ty for Testing and Materials, 1981, p

428-446

•

G.L. Burkhardt and R.E. Beissner, Probability of Detection of Flaws in a Gas Turbine Engine Component

Using Electric Current Perturbation, Rev. Prog. Quant. NDE, Vol 4, 1985, p 333-341

• G.L. Burkhardt, F.N

. Kusenberger, and R.E. Beissner, Electric Current Perturbation Inspection of Selected

Retirement-For-Cause Turbine Engine Components, Rev.Prog. Quant. NDE, Vol 3, 1984, p 1377-1387

• G.L. Burkhardt, F.N. Kusenberger, R.E. Beissner, and C.M. Teller, Elect

ric Current Perturbation Inspection

of Complex Geometry Features in Gas Turbine Engine Components, in

Proc. of the Fourteenth Symposium

on Nondestructive Evaluation, Nondestructive Testing Information Analysis Center, April 1983, p 468-474

• G.L. Burkhard

t, G.W. Scott, and B.N. Ranganathan, Electric Current Perturbation Method for Inspection of

Aluminum Welds, Rev. Prog. Quant. NDE, Vol 5, 1986, p 1713-1721

• R. Palanisamy, D.O. Thompson, G.L. Burkhardt, and R.E. Beissner, Eddy Current Detection of Subsur

face

Cracks in Engine Disk Bolt Holes, Rev.Prog. Quant. NDE, Vol 3, 1984, p 643-651

Magabsorption NDE

William L. Rollwitz, Southwest Research Institute

Introduction

MAGABSORPTION TECHNIQUES were originated by the author at Southwest Research Institute in June 1957. During

electron spin resonance (ESR) measurements on a rocket propellant, it was noticed that an entirely different ESR signal

was obtained from the propellant with Fe

than from the same propellant without the iron oxide. When the microwave

power for the ESR was turned off but the modulation was left on, the different signal disappeared. When the modulation

power was turned off, the different signal also disappeared. Therefore, the signal appeared to depend not only on the

microwaves but also on the modulation. The frequency of the signal itself was predominantly a second harmonic of the

modulation frequency. When the magnetic field was varied from 0 to 0.4 T (4000 G), the ESR-type signal was obscured

by the different signal. The shape of the different signal did not change as the field varied from 0 to 0.4 T (4000 G), but

the amplitude did. The amplitude decreased almost linearly from 0.4 T (4000 G). The signal had the same shape but at a

much lower level as the field increased. The signal was obtained from the sample with iron oxide but not from the sample

without the Fe

. The same type of signal was obtained from a nuclear magnetic resonance (NMR) detector at 30 MHz

and from the ESR detector at 9240 MHz. When the modulation was stopped with the NMR and ESR detectors, the signal

disappeared.

It was concluded that the shape of this different signal was independent of the frequency used; it was a function of the

presence of Fe

, and it had a fundamental frequency that was twice the modulation frequency. When the signal was

observed before the amplitude detection in the NMR detector, it was found to be a 30-MHz carrier amplitude and/or

frequency modulated with the different signal whose fundamental frequency was twice the modulation frequency. The

phenomenon was not a resonance function of frequency. It was a function of the modulation field and was an amplitude

modulation of the radio-frequency (RF) signal used.

A schematic diagram of a system specifically made to detect the different signal is shown in Fig. 1. The sample of Fe

in Fig. 1 is in two alternating magnetic fields. The first is an RF field at 30 or 9240 MHz, and the second is a lower-

frequency field at 60 Hz. The peak value of the RF field, H

is much lower than 8 A/m (0.1 Oe). The peak value of the

lower-frequency field is much higher at around 8000 A/m (100 Oe). The RF detection coil is resonated at the radio

frequency by the tuning capacitor in the NMR detector. The output of the NMR or ESR amplifier is the special signal

amplitude modulated on the RF voltage. The different signal is separated from the RF by an amplitude detector. Because

the shape of the different signal was found to be independent of frequency, a much lower frequency was used. For the

first detection system specifically made to detect the different signal, a frequency of 0.5 MHz was chosen as shown in

Fig. 1. An NMR-type detector was used at 0.5 MHz to detect the different signal.

Fig. 1 Schematic of the special test system for the investig

ation of the detection of the different signal with an

NMR-type detector

The resonant circuit of the NMR-type detector can be represented by the diagram in Fig. 2. It is possible to use the

techniques used in nuclear magnetic resonance, which relate the very small changes in the energy absorbed by the nuclei

from the RF magnetic field as an amplitude modulation of the RF voltage V

across the RF coil in Fig. 2, where L is the

inductance of the RF coil, R

is the resistance representing the losses in the coil, and C is the capacitance required to tune

the circuit to

= 2 · 0.5 · 10

radians/s. The impedance, Z

, with no sample in the two fields, is:

(Eq 1)

Fig. 2 Resonant circuit showing the sample in the RF magnetic field

When there is a sample in the two magnetic fields, the impedance is changed to Z

, which is:

(Eq 2)

where F is the filling factor of the sample in the RF magnetic field, P is the power absorbed by the resonated nuclei

sample, Q is the figure of merit of the resonant circuit where Q =

L/R, and H

is the peak amplitude of the sinusoidal

RF magnetic field. The change of impedance caused by the power absorption by the sample is:

(Eq 3)

where

and

Equation 3 shows that the change in the resonant impedance of the detection circuit is directly proportional to the power

absorption (the rate of energy absorption) by the sample from the RF magnetic field. The nuclear magnetic resonance

signal is also directly proportional to the power absorbed by the chosen nuclei from an RF magnetic field. The RF power

absorbed by the nuclei is proportional to the complex susceptibility of the sample, which is:

= ' - ''

(Eq 4)

The power absorbed from the RF coil by the sample will therefore have two components--one in phase with the RF

magnetic field proportional to '' and a second one 90° out of phase with the RF magnetic field proportional to '. The

first is called the in-phase or absorption component of the NMR signal, and the second is called the quadrature-phase or

dispersion component of the signal. The different signal also has in-phase and quadrature-phase components. Therefore,

the different signal is directly proportional to the power absorbed by the susceptibility of the sample of iron oxide. The

susceptibility has both an in-phase component and a quadrature-phase component.

In October 1957, the author received funding for a program to study the phenomenon in more detail. Just before the

project was undertaken, it was found that W.E. Bell of Varian Associates had discovered a similar different signal. He had

termed the phenomenon magnetoabsorption and had written a paper with that title in 1956 (Ref 1). Bell attributed the

phenomenon to magnetoresistance and called it magnetoabsorption because the apparatus measures the RF absorption in

the test specimen as a function of the applied magnetic field.

The magnetoresistance phenomenon is described by Bozorth as the change in the electrical resistivity of iron when it is

magnetized (Ref 2). In most magnetic materials, this magnetoresistance is an increase of resistivity with magnetization

when the current and magnetization are parallel and a decrease when they are at right angles to each other; the change in

resistivity is independent of the sense of field or induction and, in this respect, is like magnetostriction (Ref 2). Reference

2 also gives the curve of Δρ/ρ as a function of the applied field strength; this magnetoresistance curve for nickel is shown

in Fig. 3. The magnetoresistance curve has a butterfly shape and hysteresis. The magnetoresistance change is 10 to 15%

of the resistivity at saturation, Δρ

. Reference 2 also states that the change in resistance (increase or decrease) is beyond

the knee of the magnetization curve.

Fig. 3 Hysteresis of the magnetoresistance of nickel as plotted against applied field strength. Source: Ref 2

The effects of both tension and magnetic field on the resistivity of ferromagnetic materials are discussed in Ref 2 with the

aid of the domain illustrations in Fig. 4. According to Ref 2, domains having an initially haphazard distribution of

orientation (Fig. 4a) are aligned parallel by a strong (saturating) magnetic field (Fig. 4b). Tension applied to an

unmagnetized material, with positive magnetostriction, orients the domains along the axis of tension with one-half of the

domains antiparallel to the others (Fig. 4c). Because the resistivity depends on magnetization and increases with

increasing magnetization, each domain will contribute to the resistivity an amount depending on its orientation. This

contribution will be the same for parallel and antiparallel orientations; for the specimen as a whole, Δρ will be the same

for Fig. 4(b) and 4(c) and will be Δρ

. When the magnetostriction is negative, tension orients the domains at right angles

to the direction of tension, and the resistivity decreases (Fig. 4d). A magnetic field applied subsequently causes the

domains to rotate so as to increase the resistivity (Fig. 4e). When a material having positive magnetostriction is subjected

to strong tension, subsequent application of a magnetic field will cause 180° rotations (of the domains) almost

exclusively, and these will have no effect on resistivity (going from Fig. 4c to e).

Fig. 4

Domain interpretation of the effects of field and tension on resistance. At right, domain vectors have

been moved to a common origin. (a) Unm

agnetized. (b) Saturated. (c) Tension applied (positive

magnetostriction). (d) Tension applied (negative magnetostriction).

(e) Tension and strong field. (f) Residual

induction

Because of the above discussion from Ref 2, the term magnetoabsorption was adopted by the author for the phenomenon,

and the internally sponsored work at Southwest Research Institute was undertaken using the title "Magnetoabsorption"

(Ref 3). However, the term magabsorption is now being used because the techniques have developed differently from the

magnetoabsorption described by Bell (Ref 1).

One of the magabsorption measurements described in Ref 3 involved measurements on iron carbonyl powders with a

bridge-type NMR detector. The in-phase or the loss component and the quadrature or the reactive component of the

magabsorption signal were separately detected and recorded. They were displayed as Lissajous figures where the

magabsorption signal (in-phase or quadrature-phase) was on the vertical axis of the oscilloscope and a voltage

proportional to the 60-Hz modulation was fed to the horizontal axis. The in-phase (resistive) and the quadrature-phase

(reactive) components of the magabsorption signals from a 3 m (120 in.) sample of carbonyl iron are shown in Fig. 5.

The shapes of the quadrature-phase signals are quite different from the in-phase signals in Fig. 5. The resistivity or the in-

phase signals have higher harmonic content than the quadrature-phase or reactive signals. Other results obtained during

the research summarized in Ref 3 are discussed in the section "Measurements and Applications" in this article.

Fig. 5 Comparison of resistivity and permeability components for a 3 m (120 in.) sample of iron carbonyl

The above discussion of the effect on the magnetoresistance caused by an applied magnetic field and an applied stress on

a piece of ferromagnetic material, coupled with the differences between the magabsorption signals from nickel powder

and nickel plating on a brass screw, led to the belief that magabsorption could be used to measure the stress and material

properties of ferromagnetic materials and nickel-plated nonferromagnetic materials. Research in this area was conducted,

and it resulted in four final reports (Ref 4, 5, 6, 7). This work is briefly described in this article, along with information

from Ref 8. Two patents have also been issued for this technology (Ref 9, 10).

References

1. W.E. Bell, Magnetoabsorption, Vol 2, Proceedings of

the Conference on Magnetism and Magnetic

Materials, American Institute of Physics, 1956, p 305

2. R.M. Bozorth, Magnetism and Electrical Properties, in Ferromagnetism, D. Van Nostrand, 1951, p 745-768

3. W.L. Rollwitz, "Magnetoabsorption," Final Report, Research Project No. 712-

4, Southwest Research

Institute, 1958

W.L. Rollwitz and A.W. Whitney, "Special Techniques for Measuring Material Properties," Technical

Report ASD-TDR-64-123, USAF Contract No. AF-33(657)-10326, Air Force Materials Laboratory, 1964

W.L. Rollwitz and J.P. Classen, "Magnetoabsorption Techniques for Measuring Material Properties,"

Technical Report AFML-TR-65-17, USAF Contract No. AF-33(657)-

10326, Air Force Materials

Laboratory, 1965

6. W.L. Rollwitz and J.P. Classen, "Magneto

absorption Techniques for Measuring Material Properties,"

Technical Report AFML-TR-66-76 (Part I), USAF Contract No. AF-33(657)-

10326, Air Force Materials

Laboratory, 1966

7. W.L. Rollwitz, "Magnetoabsorption Techniques for Measuring Material Properties.

Part II. Measurements

of Residual and Applied Stress." Technical Report AFML-TR-66-76 (Part II), USAF Contract No. AF-

33(615)-5068, Air Force Materials Laboratory, 1968

8. W.L. Rollwitz, Magnetoabsorption, Progress in Applied Materials Research, Vol 6, E.

G. Stanford, J.H.

Fearon, and W.J. McGonnagle, Ed., Heywood, 1964

W.L. Rollwitz, Sensing Apparatus for Use With Magnetoabsorption Apparatus, U.S. Patent 3,612,968,

1971

10.

W.L. Rollwitz, J. Arambula, and J. Classen, Method of Determining Stress in Fe

rromagnetic Members

Using Magnetoabsorption, 1974

Magabsorption NDE

William L. Rollwitz, Southwest Research Institute

Magabsorption Theory

Magabsorption involves the measurement of the change in the energy absorbed from the field of an RF coil, by a

ferromagnetic material placed in the field of the coil, as a function of the strength of a magnetic bias field also applied to

the materials. The general concept of magabsorption is shown in Fig. 6. The ferromagnetic material is in both the RF

magnetic field of strength H

and the bias magnetic field of strength H

. The value of H

is much smaller than H

Fig. 6 General concept of magabsorption showing the RF field and the bias field in the material

A thorough understanding of the basic electrical and magnetic properties of ferromagnetic materials is essential to the

theoretical treatment of magabsorption. With these properties, a solution can be obtained that has correlated very well

with experimental measurements. However, because some of these properties are difficult to describe mathematically, an

exact solution is difficult if not impossible. Therefore, simplifying approximations are required to obtain a useful solution.

B-H Characteristics. When a magnetic field is applied to any material, a resulting magnetic flux density, B, is induced.

The intensity of the applied magnetic field is called the magnetic field intensity, H. The resulting flux density, B, is

directly proportional to the magnetic field intensity, H.

This relationship is normally written in the following form:

B = μH

(Eq 5)

The constant of proportionality, μ, is called the permeability, which may be nonlinear. When a ferromagnetic material is

subjected to a magnetic field, the flux density, B, is not a linear function of the magnetic field intensity, H. In other words,

the permeability, μ, is not a constant. In addition to this nonlinearity, the permeability depends on the past magnetic and

stress history of the ferromagnetic material. This dependence on past magnetic and stress history is known as hysteresis.

The hysteresis loop is explained by the concept of magnetic domains. Each domain consists of a group of atoms acting as

a small magnet. The group of atoms in a small region forms a small magnet because molecular interactions cause the

molecular magnetic moments to be aligned parallel to one another. In other words, each domain is spontaneously

magnetized to saturation even in the absence of an external magnetic field. Because the directions of magnetization in

different domains are not necessarily parallel to one another, the resultant magnetic flux density may be zero. When the

specimen is placed in a magnetic field, the resultant magnetic flux density can be changed by two different mechanisms.

The first is movement of a domain wall, and the second is directional rotation of the magnetic field of the domain. These

two domain processes must stop at a high magnetic field intensity because all the domains become aligned with the

applied magnetic field. This limiting process causes the nonlinearity of the B-H characteristics. Also, the stability of

different domain orientations causes the hysteresis of the B-H characteristics. Therefore, the shape of the hysteresis loop

and the domain orientation are dependent on the type of material, its stress history, and its magnetic history.

Reversible Permeability. If at some point on the hysteresis loop the magnetic field intensity is reversed from its

normal direction, the resulting magnetic flux density will not follow the original hysteresis loop. This deviation from the

original hysteresis is explained by the concept of the domain. If the magnetic field intensity is decreased before it reaches

the previous limit, the orientation stability of the domain will cause hysteresis.

If the deviation of the magnetic field intensity is a small cyclic field, ΔH, super-imposed on a large bias field, H

then a

smaller inner hysteresis loop will be formed, as shown in Fig. 7. Note that the magnetic biasing field, H

, is the magnetic

field that determines the larger external hysteresis loop. As the size of the small cyclic deviation in the magnetic field

intensity, ΔH, is decreased, the resulting change in the magnetic flux density, ΔB, is decreased. As ΔH approaches zero,

the ratio ΔB/ΔH approaches a limiting value μ

(Eq 6)

where

is the permeability of free space (in henries/meter),

, is the relative reversible permeability (unitless), and is

the reversible permeability (in henries/meter).

Fig. 7 Superimposed hysteresis loops

The reversible, or alternating current (ac), permeability is the property of great interest in the study of magabsorption.

Reversible permeability is dependent on the reversible boundary displacement and reversible rotation of the domains in

the ferromagnetic material. Therefore, the reversible permeability is dependent on the type of material and its magnetic

and stress history. The relative reversible permeabilities for unannealed nickel, annealed nickel, and iron are shown in

Fig. 8. For all the relative reversible permeability curves in Fig. 8, the upper portion of the curve is observed for an

increasing bias field, and the lower portion of the curve is observed for a decreasing bias field. Assuming that the

ferromagnetic material was originally demagnetized, the plot of

versus H

is symmetrical about H

= 0.

Fig. 8

Relative reversible permeability curves for (a) unannealed nickel wire, (b) annealed nickel wire, and (c)

iron thermocouple wire