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

Подождите немного. Документ загружается.

Fig. 14 Three-stage mechanistic model of diffusion welding. (a) Initial asperity contact. (b) First-

stage

deformation and interfacial boundary formation. (c) Second-stage grain-

boundary migration and pore

elimination. (d) Third-stage volume diffusion and pore elimination. Source: Ref 5

Successful nondestructive evaluation of diffusion-bonded joints requires that the maximum size and distribution of

discontinuities be determined. However, many conventional NDE methods and equipment are not adequate for

discontinuity determination.

Fluorescent penetrants provide excellent detection capability for lack of bonding as long as the interfacial crack breaks the

surface. They are completely ineffective for defects that have no path to the surface.

Conventional film radiography is not suitable for detecting the extremely small defects involved in diffusion bonding, but

the use of x-ray microfocus techniques coupled with digital image enhancement offers an improvement in resolution.

Discontinuities as small as 50 m (0.002 in.) have been detected.

Conventional ultrasonic testing has some applications, although only significant lack of bonding or clusters of smaller

defects can be reliably detected. High-resolution flaw detection involving frequencies approaching 100 MHz appears to

have distinct advantages over conventional testing. Scanning acoustic microscopy also appears to offer excellent

possibilities in diffusion bond inspection. Eddy current and thermal methods are relatively unsatisfactory for most

applications.

Methods of Nondestructive Inspection

The nondestructive inspection of weldments has two functions:

•

Quality control, which is the monitoring of welder and equipment performance and of the quality of the

consumables and the base materials used

• Acceptance or rejection of a weld on the basis of its fitness for p

urpose under the service conditions

imposed on the structure

The appropriate method of inspection is different for each function. If evaluation is a viable option, discontinuities must

be detected, identified, located exactly, sized, and their orientation established, which limits inspection to a volumetric

technique.

Weld discontinuities constitute the center of activity with the inspection of welded constructions. The most widely used

inspection techniques used in the welding industry are visual, liquid penetrant, magnetic particle, radiographic, ultrasonic,

acoustic emission, eddy current, and electric current perturbation methods. Each of these techniques has specific

advantages and limitations. Existing codes and standards that provide guidelines for these various techniques are based on

the capabilities and/or limitations of these nondestructive methods.

Selection of Technique. A number of factors influence selection of the appropriate nondestructive test technique for

inspecting a welded structure, including discontinuity characteristics, fracture mechanics requirements, part size,

portability of equipment, and other application constraints. These categories, although perhaps unique to a specific

inspection problem, may not clearly point the way to the most appropriate technique. It is generally necessary to exercise

engineering judgment in ranking the importance of these criteria and thus determining the optimum inspection technique.

Characteristics of the Discontinuity. Because nondestructive techniques are based on physical phenomena, it is

useful to describe the properties of the discontinuity of interest, such as composition and electrical, magnetic, mechanical,

and thermal properties. Most significant are those properties that are most different from those of the weld or base metal.

It is also necessary to identify a means of discriminating between discontinuities with similar properties.

Fracture mechanics requirements, solely from a discontinuity viewpoint, typically include detection,

identification, location, sizing, and orientation. In addition, complicated configurations may necessitate a non-destructive

assessment of the state of the stress of the region containing the discontinuity. In the selection process, it is important to

establish these requirements correctly. This may involve consultation with stress analysts, materials engineers, and

statisticians.

Often, the criteria may strongly suggest a particular technique. Under ideal conditions, such as in a laboratory, the

application of such a technique might be routine. In the field, however, other factors may force a different choice of

technique.

Constraints tend to be unique to a given application and may be completely different even when the welding process

and metals are the same. Some of these constraints include:

• Access to the region under inspection

• Geometry of the structure (flat, curved, thick, thin)

• Condition of the surface (smooth, irregular)

• Mode of inspection (preservice, in-service, continuous, periodic, spot)

• Environment (hostile, underwater, and so on)

• Time available for inspection (high speed, time intensive)

• Reliability

• Application of multiple techniques

• Cost

Failure to consider adequately the constraints imposed by a specific application can render the most sophisticated

equipment and theory useless. Moreover, for the simple or less important cases of failure, it may be unnecessary. Once

criteria have been established, an optimum inspection technique can be selected, or designed and constructed.

The terms accuracy, sensitivity, and reliability are used loosely in NDE. Often, they are discussed as one term to

avoid distinguishing among the specific aspects of these terminologies.

Accuracy is the attribute of an inspection method that describes the correctness of the technique within the limits of its

precision. In other words, the technique is highly accurate if the indications resulting from the technique are correct. This

does not mean that the technique was able to detect all discontinuities present, but rather that those indicated actually

exist.

Sensitivity, on the other hand, refers to the capability of a technique to detect discontinuities that are small or that have

properties only slightly different from the material in which they reside. Figure 15 schematically illustrates the concepts

of accuracy and sensitivity in the context of detection probability. In general, sensitivity is gained at the expense of

accuracy because high sensitivity increases the probability of false alarms.

Fig. 15 Detection p

robability for a true positive indication. High sensitivity increases the likelihood of false

indications. The minimum NDE parameter size required to establish fitness-for-

purpose must lie to the right of

the transition region, or reliability threshold, to achieve satisfactory reliability.

Reliability is a combination of both accuracy and sensitivity. Three factors influence reliability: inspection procedure,

including the instrumentation; human factors (inspector motivation, experience, training, education, and so on); and data

analysis. Uncalibrated equipment, improper application of technique, and inconsistent quality of accessory equipment

(transducers, couplant, film, chemicals, and so on) may affect accuracy and, in some cases, sensitivity. Poor inspector

technique, unfamiliar response, lack of concentration, and other human factors can combine to reduce reliability. Data

analysis, or the lack of it, can influence reliability as well; generally, inspection is performed under conditions in which

detection probability is less than 100% and is not constant with discontinuity severity. Consequently, statistics must be

employed to establish the level of confidence that may be attached to the inspection results.

High sensitivity with low accuracy may be far worse, from the viewpoint of reliability, than low sensitivity with high

accuracy, especially if the sensitivity level is adequate for detecting the weld discontinuities in question. As a general rule,

the transition region of the detection probability curve indicates the degree of reliability. If this region occurs with the

limits encompassed by inspection capabilities, which are smaller than the values required for evaluating the fitness-for-

purpose of the welds being inspected, reliability is satisfactory. If, on the other hand, the region occurs at values higher

than those required, reliability is unsatisfactory. The transition region can be viewed as the reliability threshold. More

detailed information on the reliability of nondestructive test data can be found in the Section "Quantitative Nondestructive

Evaluation" in this Volume.

Visual Inspection

For many noncritical welds, integrity is verified principally by visual inspection. Even when other nondestructive

methods are used, visual inspection still constitutes an important part of practical quality control. Widely used to detect

discontinuities, visual inspection is simple, quick, and relatively inexpensive. The only aids that might be used to

determine the conformity of a weld are a low-power magnifier, a borescope, a dental mirror, or a gage. Visual inspection

can and should be done before, during, and after welding. Although visual inspection is the simplest inspection method to

use, a definite procedure should be established to ensure that it is carried out accurately and uniformly.

Visual inspection is useful for checking the following:

• Dimensional accuracy of weldments

• Conformity of welds to size and contour requirements

• Acceptability of weld appearance with regard to surface roughness, weld spatter, and cleanness

• Presence of surface flaws such as unfilled craters, pockmarks, undercuts, overlaps, and cracks.

Although visual inspection is an invaluable method, it is unreliable for detecting subsurface flaws. Therefore, judgment of

weld quality must be based on information in addition to that afforded by surface indications.

Additional information can be gained by observations before and during welding. For example, if the plate is free of

laminations and properly cleaned and if the welding procedure is followed carefully, the completed weld can be judged on

the basis of visual inspection. Additional information can also be gained by using other NDI methods that detect

subsurface and surface flaws.

Dimensional Accuracy and Conformity. All weldments are fabricated to meet certain specified dimensions. The

fabricator must be aware of the amount of shrinkage that can be expected at each welded joint, the effect of welding

sequence on warpage or distortion, and the effect of subsequent heat treatment used to provide dimensional stability of the

weldment in service. Weldments that require rigid control of final dimensions usually must be machined after welding.

Dimensional tolerances for as-welded components depend on the thickness of the material, the alloy being welded, the

overall size of the product, and the particular welding process used.

The dimensional accuracy of weldments is determined by conventional measuring methods, such as rules, scales, calipers,

micrometers, and gages. The conformity of welds with regard to size and contour can be determined by a weld gage. The

weld gage shown in Fig. 16 is used when visually inspecting fillet welds at 90° intersections. The size of the fillet weld,

which is defined by the length of the leg, is stamped on the gage. The weld gage determines whether or not the size of the

fillet weld is within allowable limits and whether there is excessive concavity or convexity. This gage is designed for use

on joints between surfaces that are perpendicular. Special weld gages are used when the surfaces are at angles other than

90°. For groove welds, the width of the finished welds must be in accordance with the required groove angle, root face,

and root opening. The height of reinforcement of the face and root must be consistent with specified requirements and can

be measured by a weld gage.

Minimum allowable length of leg

Maximum allowable length of leg

1.414 times maximum allowable throat size (specifies maximum allowable convexity)

Maximum allowable length of leg when maximum allowable concavity is present

A plus B plus nominal weld size (or nominal length of leg)

Minimum allowable throat size (specifies maximum allowable concavity)

Additional tolerance for clearance of gage when placed in the fillet

Fig. 16

Gage for visual inspection of a fillet weld at a 90° intersection. Similar gages can be made for other

angles. Dimension given in inches

Appearance Standards. The acceptance of welds with regard to appearance implies the use of a visual standard, such

as a sample weldment or a workmanship standard. Requirements as to surface appearance differ widely, depending on the

application. For example, when aesthetics are important, a smooth weld that is uniform in size and contour may be

required.

The inspection of multiple-pass welds is often based on a workmanship standard. Figure 17 indicates how such standards

are prepared for use in the visual inspection of groove and fillet welds. The workmanship standard is a section of a joint

similar to the one in manufacture, except that portions of each weld pass are shown. Each pass of the production weld is

compared with corresponding passes of the workmanship standard.

Fig. 17 Workmanship standard for visual comparison during inspection of single-V-

groove welds and fillet

welds. Dimensions given in inches

Discontinuities. Before a weld is visually inspected for discontinuities such as unfilled craters, surface holes,

undercuts, overlaps, surface cracks, and incomplete joint penetration, the surface of the weld should be cleaned of oxides

and slag. Cleaning must be done carefully. For example, a chipping hammer used to remove slag could leave hammer

marks that can hide fine cracks. Shotblasting can peen the surface of relatively soft metals and hide flaws. A stiff wire

brush and sandblasting have been found to be satisfactory for cleaning surfaces of slag and oxides without marring.

Additional information on the uses and equipment associated with the visual examination of parts and assemblies is

available in the article "Visual Inspection" in this Volume.

Magnetic Particle Inspection

Magnetic particle inspection is a nondestructive method of detecting surface and near-surface flaws in ferromagnetic

materials. It consists of three basic operations:

• Establishing a suitable magnetic field in the material being inspected

• Applying magnetic particles to the surface of the material

•

Examining the surface of the material for accumulations of the particles (indications) and evaluating the

serviceability of the material

Capabilities and Limitations. Magnetic particle inspection is particularly suitable for the detection of surface flaws

in highly ferromagnetic metals. Under favorable conditions, those discontinuities that lie immediately under the surface

are also detectable. Nonferromagnetic and weakly ferromagnetic metals, which cannot be strongly magnetized, cannot be

inspected by this method. With suitable ferromagnetic metals, magnetic particle inspection is highly sensitive and

produces readily discernible indications at flaws in the surface of the material being inspected. Details on the application

of this method are available in the article "Magnetic Particle Inspection" in this Volume.

The types of weld discontinuities normally detected by magnetic particle inspection include cracks, LOP, LOF, and

porosity open to the surface. Linear porosity, slag inclusions, and gas pockets can be detected if large or extensive or if

smaller and near the surface. The recognition of patterns that indicate deep-lying flaws requires more experience than that

required to detect surface flaws.

Nonrelevant indications that have no bearing on the quality of the weldment may be produced. These indications are

magnetic particle patterns held by conditions caused by leakage fields. Some of these conditions are:

• Particles held mechanically or by gravity in surface irregularities

• Adherent scale or slag

• Indications at a sharp change in material direction, such as sharp fillets and threads

• Grain boundaries. Large grain size in the weld metal or the base metal may produce indications

•

Boundary zones in welds, such as indications produced at the junction of the weld metal and the base

metal. This condition occurs in fillet welds at T-joints, or in double-V-

groove joints, where 100%

penetration is not required

• Flow lines in forgings and formed parts

•

Brazed joints. Two parts made of a ferromagnetic material joined by a nonferromagnetic material will

produce an indication

• Diff

erent degrees of hardness in a material, which will usually have different permeabilities that may

create a leakage field, forming indications

Operational Requirements. The magnetic particle inspection of weldments requires that the weld bead be free of

scale, slag, and moisture. For maximum sensitivity, the weld bead should be machined flush with the surface; however,

wire brushing, sandblasting, or gritblasting usually produces a satisfactory bead surface. If the weld bead is rough,

grinding will remove the high spots.

Weldments are often inspected using the dry particle method. A powder or paste of a color that gives the best possible

contrast to the surface being inspected should be used.

The type of magnetizing current used depends on whether there are surface or subsurface discontinuities. Alternating

current is satisfactory for surface cracks, but if the deepest possible penetration is essential, direct current, direct current

with surge, or half-wave rectified alternating current is used.

The voltage should be as low as practical to reduce the possibility of damage to the surface of the part from overheating

or arcing at contacts. Another advantage of low voltage is freedom from arc flashes if a prod slips or is withdrawn before

the current is turned off.

The field strength and flux density used must be determined for each type of weldment. An overly strong field will cause

the magnetic particles to adhere too tightly to the surface and hinder their mobility, preventing them from moving to the

sites of the flaws. Low field strengths result in nondiscernible patterns and failure to detect indications.

Inspections can be made using the continuous-field and residual-field methods. In the continuous-field method, magnetic

particles are placed on the weldment while the current is flowing. In the residual-field method, the particles are placed on

the weldment after the magnetizing current is turned off. Residual magnetic fields are weaker than continuous fields.

Consequently, inspections using the residual-field method are less sensitive.

The need for the demagnetization of weldments after magnetic particle inspection must be given serious consideration.

Where subsequent welding or machining operations are required, it is good practice to demagnetize. Residual magnetism

may also hinder cleaning operations and interfere with the performance of instruments used near the weldment.

Liquid Penetrant Inspection

Liquid penetrant inspection is capable of detecting discontinuities open to the surface in weldments made of either

ferromagnetic or nonferromagnetic alloys, even when the flaws are generally not visible to the unaided eye. Liquid

penetrant is applied to the surface of the part, where it remains for a period of time and penetrates into the flaws. For the

correct usage of liquid penetrant inspection, it is essential that the surface of the part be thoroughly clean, leaving the

openings free to receive the penetrant. Operating temperatures of 20 to 30 °C (70 to 90 °F) produce optimum results. If

the part is cold, the penetrant may become chilled and thickened so that it cannot enter very fine openings. If the part or

the penetrant is too hot, the volatile components of the penetrant may evaporate, reducing the sensitivity.

After the penetrating period, the excess penetrant remaining on the surface is removed. An absorbent, light-colored

developer is then applied to the surface. This developer acts as the blotter, drawing out a portion of the penetrant that had

previously seeped into the surface openings. As the penetrant is drawn out, it diffuses into the developer, forming

indications that are wider than the surface openings. The inspector looks for these colored or fluorescent indications

against the background of the developer. Details of the mechanics of this method are given in the article "Liquid Penetrant

Inspection" in this Volume.

Radiographic Inspection

Radiography is a nondestructive method that uses a beam of penetrating radiation such as x-rays and -rays (see the

article "Radiographic Inspection" in this Volume). When the beam passes through a weldment, some of the radiation

energy is absorbed, and the intensity of the beam is reduced. Variations in beam intensity are recorded on film or on a

screen when a fluoroscope or an image intensifier is used. The variations are seen as differences in shading that are

typical of the types and sizes of any discontinuities present. Radiography is used to detect subsurface discontinuities as

well as those that are open to the surface.

Penetrameters. Most U.S. codes and specifications for radiographic inspection require that the sensitivity to a specific

flaw size be indicated on each radiograph. Sensitivity is determined by placing a penetrameter (image-quality indicator)

that is made of substantially the same material as the specimen directly on the specimen, or on a block of identical

material of the same thickness as the specimen, prior to each exposure.

The penetrameters are usually placed on the side of the specimen nearest the source, parallel and adjacent to the weld at

one end of the exposed length, with the small holes in the penetrameter toward the outer end. Penetrameter thickness is

usually specified to be 2% of the specimen thickness (through the weld zone), although 1 and 4% penetrameter

thicknesses are also common. American Society for Testing and Materials and American Society of Mechanical

Engineers plaque-type penetrameters have three small drilled holes, whose diameters equal one, two, and four times the

penetrameter thickness. Image quality is determined by the ability to distinguish both the outline of the penetrameter and

one or more of the drilled holes on the processed radiograph; for example, 2-2T sensitivity is achieved when a 2%

penetrameter and the hole that has a diameter of twice the penetrameter thickness are visible. Also, the image of raised

numbers that identify the actual thickness (not the percentage thickness) of a penetrameter should appear clearly,

superimposed on the image of the penetrameter.

Surface discontinuities that are detectable by radiography include undercuts (Fig. 18 and 19), longitudinal grooves,

concavity at the weld root (Fig. 20 and 21), incomplete filling of grooves, excessive penetration (Fig. 22), offset or

mismatch (Fig. 23 and 24), burn-through (Fig. 25), irregularities at electrode-change points, grinding marks, and electrode

spatter. Surface irregularities may cause density variations on a radiograph. When possible, they should be removed

before a weld is radiographed. When impossible to remove, they must be considered during interpretation.

Fig. 18

External undercut, which is a gouging out of the piece to be welded, alongside the edge of the top or

external surface of the weld. Radiographic image: An irregular darker density along the edge of the weld image.

The densit

y will always be darker than the density of the pieces being welded. Welding process: SMAW. Source:

E.I. Du Pont de Nemours & Company, Inc.

Fig. 19

Internal (root) undercut, which is a gouging out of the parent metal, alongside the edge of the bottom

or internal surface of the weld. Radiographic image: An irregular darker density near the center of the width of

the weld image an

d along the edge of the root pass image. Welding process: SMAW. Source: E.I. Du Pont de

Nemours & Company, Inc.

Fig. 20 Exter

nal concavity or insufficient fill, which is a depression in the top of the weld, or cover pass,

indicating a thinner-than-

normal section thickness. Radiographic image: A weld density darker than the density

of the pieces being welded and extending across

the full width of the weld image. Welding process: SMAW.

Source: E.I. Du Pont de Nemours & Company, Inc.

Fig. 21 Internal con

cavity (suck back), which is a depression in the center of the surface of the root pass.

Radiographic image: An elongated irregular darker density with fuzzy edges, in the center of the width of the

weld image. Welding process: GTAW-SMAW. Source: E.I. Du Pont de Nemours & Company, Inc.

Fig. 22 Excessive penetration (icicles, drop-through), which is extra metal at the bottom (roo

t) of the weld.

Radiographic image: A lighter density in the center of the width of the weld image, either extended along the

weld or in isolated circular drops. Welding process: SMAW. Source: E.I. Du Pont de Nemours & Company, Inc.

Fig. 23

Offset or mismatch with LOP, which is a misalignment of the pieces to be welded and insufficient filling

of the bottom of the weld or root a

rea. Radiographic image: An abrupt density change across the width of the

weld image with a straight longitudinal darker-

density line at the center of the width of the weld image along

the edge of the density change. Welding process: SMAW. Source: E.I. Du Pont de Nemours & Company, Inc.