Ultrasonic Testing - ETS Course Notes - Issue 17 - PP [PDF]

Zitiervorschau

ULTRASONIC TESTING

IMechE Engineering Training Centre 4 Europa View Sheffield Business Park Sheffield S9 1XH

T: +44 (0)114 399 5720 E: [email protected] W: trainingsolutions.imeche.org

NOTES

TABLE

OF

CONTENTS

INTRODUCTION .......................................................................................... INTRO1-1 BASIC PRINCIPLES .............................................................................................. UT1 Introduction to the basic concept .....................................................................UT1-1 The nature of sound .......................................................................................UT1-2 The acoustic spectrum ....................................................................................UT1-3 THE PROPAGATION OF SOUND ............................................................................ UT2 The ultrasonic beam ......................................................................................UT2-1 Side lobes .....................................................................................................UT2-3 The ultrasonic pulse .......................................................................................UT2-4 Resolution .....................................................................................................UT2-5 Pulse Repetition Frequency (PRF) .....................................................................UT2-5 Modes of propagation .....................................................................................UT2-6 Bulk waves………………………………………………………………………………………………………………………UT2-6 Boundary waves ............................................................................................UT2-7 Factors affecting the propagation of ultrasound ..................................................UT2-8 Acoustic impedance ........................................................................................UT2-8 Couplant .......................................................................................................UT2-9 Attenuation ................................................................................................. UT2-10 The decibel (dB) .......................................................................................... UT2-10 SOUND GENERATION .......................................................................................... UT3 The piezo effect ............................................................................................UT3-1 Reflection, refraction and Snell’s law .................................................................UT3-2 Mode conversion ...........................................................................................UT3-3 Diffraction ....................................................................................................UT3-3 Delta Technique .............................................................................................UT3-4 Critical angles ...............................................................................................UT3-6 EQUIPMENT ......................................................................................................... UT4 Probes ..........................................................................................................UT4-1 Probe frequency, bandwidth and damping .........................................................UT4-5 Probe selection ..............................................................................................UT4-5 Calibration blocks and their uses ......................................................................UT4-7 0° compression probe uses..............................................................................UT4-7 Shear probe uses ...........................................................................................UT4-8 Block no.2, A4, V2, DIN54/122 or kidney block ..................................................UT4-8 Institute of Welding (IOW)/A5 block .................................................................UT4-9 Ultrasonic flaw detectors .................................................................................UT4-9 Equipment checks ........................................................................................ UT4-10 0° PROBE SCANNING .......................................................................................... UT5 Calibration ....................................................................................................UT5-1 To calibrate a 0° probe to a range of 0 to 100 mm ..............................................UT5-2 Accurate measurement using an Analogue Flaw Detector .....................................UT5-2 Multiple back wall method ...............................................................................UT5-3 Defect detection.............................................................................................UT5-4 Sensitivity .....................................................................................................UT5-4

© Institution of Mechanical Engineers Issue 17 2016

NOTES

TABLE

OF

CONTENTS

Graphs and DAC curves ..................................................................................UT5-5 Scanning patterns 0° probe .............................................................................UT5-6 Sizing methods 0° probe .................................................................................UT5-7 ANGLE PROBE SCANNING..................................................................................... UT6 Calibration ....................................................................................................UT6-1 Angle probes test sensitivity ............................................................................UT6-2 Scanning patterns ..........................................................................................UT6-4 Skip factors ...................................................................................................UT6-5 The ratio of the sides of the triangles in the three most common probe angles ........UT6-6 The irradiation factor ......................................................................................UT6-6 Wall thickness ...............................................................................................UT6-7 Plotting systems ............................................................................................UT6-9 Sizing methods angle probes ......................................................................... UT6-10 TESTING TECHNIQUES ......................................................................................... UT7 A, B & C scanning ..........................................................................................UT7-1 Pulse echo systems ........................................................................................UT7-3 Through transmission testing ...........................................................................UT7-4 The tandem technique ....................................................................................UT7-5 Immersion testing ..........................................................................................UT7-5 Time of flight diffraction .................................................................................UT7-6 ULTRASONIC THICKNESS SURVEYING ................................................................. UT8 Using a thickness meter ..................................................................................UT8-1 Using a flaw detector ......................................................................................UT8-1 Peak and flank ...............................................................................................UT8-2 Accept/reject criteria .....................................................................................UT8-4 Reporting ......................................................................................................UT8-5 ULTRASONIC WROUGHT PLATE TESTING ............................................................. UT9 Technique .....................................................................................................UT9-1 Defects in plate material .................................................................................UT9-2 Accept/reject criteria ......................................................................................UT9-6 Reporting ......................................................................................................UT9-6 ULTRASONIC WELD TESTING ............................................................................. UT10 Technique .................................................................................................. UT10-1 Defect signal interpretation ........................................................................... UT10-3 Accept/reject criteria .................................................................................. UT10-10 Reporting .................................................................................................. UT10-10 ULTRASONIC TESTING OF FORGINGS ............................................................... UT11 General ...................................................................................................... UT11-1 Technique ................................................................................................... UT11-1 Defects in forgings ....................................................................................... UT11-3 Accept/reject criteria ................................................................................... UT11-7 Reporting .................................................................................................... UT11-7 ULTRASONIC TESTING OF CASTINGS ................................................................ UT12 General ...................................................................................................... UT12-1

© Institution of Mechanical Engineers Issue 17 2016

NOTES

TABLE

OF

CONTENTS

Technique ................................................................................................... UT12-1 Defects in castings ....................................................................................... UT12-2 Accept/reject criteria .................................................................................... UT12-6 Reporting .................................................................................................... UT12-7 BRITISH STANDARDS ............................................................................. APPENDIX A British Standards relating to ultrasonic testing.................................................. APPA-1 FORMULAE USED IN ULTRASONIC TESTING .......................................... APPENDIX B TABLE OF ACOUSTICAL VELOCITIES ...................................................... APPENDIX C Table of acoustical velocities in different materials ............................................ APPC-1 TABLE OF ACOUSTIC IMPEDANCES ........................................................ APPENDIX D Table of acoustic impedances for different materials .......................................... APPD-1 ATTENUATION FACTOR ......................................................................... APPENDIX E Example method for determining the attenuation factor of a material ................. APPE-1 EXAMPLE CALCULATIONS ...................................................................... APPENDIX F Example calculations used in ultrasonics ......................................................... APPF-1 A6 and A7 BLOCKS ................................................................................ APPENDIX G A6 and A7 Calibration Block drawings ............................................................ APPG-1

© Institution of Mechanical Engineers Issue 17 2016

INTRODUCTION NOTES

HISTORY OF ULTRASONIC TESTING 10

Piezo electric effect discovered in 1880 by Pierre & Jacques Curie in Paris. This was when they noticed that a piece of quartz, when subjected to load, generated a spark (therefore produced electricity). 20

In 1881 Lippman deduced that the opposite effect was also possible and it was proven by experiment that the application of voltage to quartz made it change shape. Following the Titanic disaster in 1912 experimenters such as Alexander Belm used low frequency ultrasound to detect icebergs (which are 90% below water and therefore ideal targets).

30

By 1914 a working sonar system was developed by Reginald Fessenden in Canada. In 1915 Paul Langelin (France) and Constantin Chilowski (Russia) were using higher frequency sonar for detection of submarines.

40

50

Ultrasonic Testing of metals as we know it today had to wait for the development of instruments capable of measuring extremely small time intervals. In the 1920’s a Russian named Sokolov made possibly the first NDT related experiment using through transmission in a bath of mercury and proved that objects in the sound path affected the transmission of sound. These were developed in the early 1940’s by Firestone (USA), Sproule (UK) and Trost (Germany) amongst others.

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

INTRO 1-1

UNIT UT1  BASIC PRINCIPLES BASIC PRINCIPLES

NOTES

INTRODUCTION TO THE BASIC CONCEPT 10

20

The most common technique used in ultrasonic testing is the pulse echo technique. This makes use of the phenomenon that sound waves travel in straight lines and are reflected by an obstacle placed in their path. The mechanism is just the same as audible sound waves bouncing off a brick wall and an echo being received. The strength of the echo is controlled by the size of the wall. Also, if the time lapse between sending and receiving the sound is measured, it is possible to determine the distance to the wall. Examples of the time intervals encountered are below:

30



In air, a reflection from a wall 165m away will take 1 second



In water, a reflection from an iceberg 740m away will take 1 second



In steel, a reflection from a defect 100mm away will take 0.000036 seconds (for a compression wave)

40

50

Given the required instrumentation we can pass sound waves through solid materials and receive echoes from the back wall of the material. If a defect is present in the material then the sound energy would be reflected back from it and give an echo earlier than that from the back wall because the sound has not travelled as far. The strength or amplitude of this echo may be an indication of the size of the defect and the distance travelled by the sound will tell us its depth. This then is the basis of ultrasonic testing.

60

The instrument that produces the sound energy is called the probe and the echoes are shown on a flat screen LCD within a flaw detector (or Cathode Ray Tube (CRT) on older analogue sets).

70

80

Note: The echo at A1 is the result of sound energy reflecting back off the front surface of the specimen together with the ringing of the crystal and the initial pulse all merged into one signal envelope.

90

Sound energy is transmitted from the probe into the test specimen at surface "A" producing an echo at A1. Some of the sound is reflected by the defect at "B" and the resulting echo appears at B1. The remainder of the sound continues through the specimen to be reflected by the back wall "C", the echo from the back wall appearing at C1.

100

If the screen is calibrated from a test block of known thickness then the depth of the defect from the specimen surface (A to B) can be read off the screen. © Institution of Mechanical Engineers Issue 17 2016

UT1-1

UNIT UT1  BASIC PRINCIPLES THE NATURE OF SOUND

NOTES

Sound is caused by mechanical vibrations. 10

In order for sound to pass there must be a medium that will support mechanical vibrations therefore SOUND CANNOT TRAVEL IN A VACUUM. The particles (molecules) within the medium vibrate passing on energy from one to another giving the effect of sound movement through the material.

The density and elasticity 20 of a medium are also the main factors that affect the velocity.

The ability to support sound depends on the elasticity and density of the medium. Since these properties will vary, from one material to another, some materials will pass sound more easily than others. Sound follows a waveform:

30

Minimum detectability is generally accepted as the smallest possible 40 discontinuity a given probe can find. Which is known as half of one wavelength, hence surface waves being the most sensitive waveform refer to UT2-7. 50

60

Minimum detectability

VELOCITY

is the distance moved in unit time

WAVELENGTH is the distance between successive peaks of a ……………………………… wave

70 The term sensitivity in ultrasonic examination references the ability to locate the smallest recordable indication at the further test distance. 80

PERIOD

is the time taken for one complete cycle

FREQUENCY

is the number of cycles per second

1 cycle per second = 1 Hertz (Hz) 1 Kilohertz (KHz)

= 1,000 Hz

1 Megahertz (MHz) = 1,000,000 Hz Wavelength Wavelength is a function of frequency and velocity.

90 Note: Velocity is sometimes denoted by the letter ‘c’.

Wave Length=

Therefore: 100

Velocity Frequency

v = f x



or

and



v f

f =

v



(Ref Appendix F for example calculation)

© Institution of Mechanical Engineers Issue 17 2016

UT1-2

UNIT UT1  BASIC PRINCIPLES THE ACOUSTIC SPECTRUM

NOTES

10 Note: The maximum frequency the human ear can detect reduces with age. It is generally accepted that most people will have heard all the high 20 frequency sounds that they are liable to encounter by the time they reach ten years of age. 30

200KHz

40

The point sound may be focussed into beams used in conventional ultrasonic scanning techniques.

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT1-3

UNIT UT2  THE PROPAGATION OF SOUND THE PROPAGATION OF SOUND

NOTES

THE ULTRASONIC BEAM 10

20

30

40

* Explanation may be taken further with the Huygens Principle. 50

** This can be seen on the flaw detector when testing a thin material where the first back wall echo is a lower amplitude then the repeat echoes.

The dead zone Seen on the screen as an extension of the initial pulse, the dead zone is the RINGING TIME of the crystal and is minimised by the damping medium behind the crystal. Flaws or other reflectors, lying in the dead zone region of the beam will not be detected. The dead zone can be seen at the start of the trace on an A-scan flaw detector, but only with single crystal probes. The dead zone increases when the probe frequency decreases or when the sound velocity increases. The near or fresnel zone In this region of the beam, the sound intensity is variable owing to wave interference*, therefore, reflectors or flaws lying in this zone may appear smaller or larger than their actual size. The signal heights displayed on the screen are unpredictable so it is desirable to keep the near zone length to a minimum**. The near zone length can be calculated using the following formula:

60

Near zone length (mm) 

70

80

D2 x f D2 or 4 4 x v

Where: D = crystal diameter (mm)  = wavelength (mm) f = probe frequency (Hz) v = test material velocity (mm/s) It can be seen from the formula that the near zone can be decreased by decreasing the crystal diameter or decreasing the probe frequency. Reference appendix F for example calculation.

90

The far or fraunhoffer zone Beyond the near zone the far zone exists. In the far zone the beam diverges resulting in a decay in sound intensity as the distance from the crystal is increased, just as a beam of light from a torch gets weaker the further it travels.

100

© Institution of Mechanical Engineers Issue 17 2016

UT2-1

UNIT UT2  THE PROPAGATION OF SOUND The amount of beam divergence depends upon the crystal size and the wavelength as shown in the following formula (assuming a circular transducer/crystal):

NOTES

10

Sin  

20

Where:

K factors:

 K



Extreme Edge (Null point/0% Intensity) = 1.22 50% Edge (-6dB) = 0.56 10% Edge (-20dB) = 1.08 30

K x v K or D D x f = the half angle D = crystal diameter (mm) = a constant f = probe frequency (Hz) = wavelength (mm) v = material velocity (mm/s)

Note: For calculation purposes, unless stated the constant value should always be the extreme edge (1.22) Reference appendix F for example calculation.

40

It may be seen from the above beam spread formula, that the beam divergence can be decreased by increasing the crystal diameter or by increasing the probe frequency. Unfortunately this will extend the length of the near zone. So in probe design there is a compromise to obtain a minimal beam spread and a short near zone.

50

60

70

80

In the far zone of the ultrasonic beam there is no wave interference therefore the sound intensity in this zone is predictable. The sound intensity reduces from 100% in the centre to 0% at the edge of the beam, therefore when the centre of the beam hits a reflector/flaw the amplitude of the signal on the screen will be at its maximum. The sound intensity will also decrease with a greater distance (in the range axis) to a reflector or flaw.

90

100

Inverse and Inverse Square laws In the far zone the amplitudes of reflected sound from large and small reflectors follow different laws. LARGE REFLECTORS (larger than the width of the ultrasonic beam) follow the INVERSE LAW - The amplitude is inversely proportional to the distance, i.e. if the distance is doubled then the signal amplitude is halved (i.e... reduced by 6dB).

© Institution of Mechanical Engineers Issue 17 2016

UT2-2

UNIT UT2  THE PROPAGATION OF SOUND NOTES

10

SMALL REFLECTORS (smaller than the width of the beam) follow the INVERSE SQUARE LAW - The amplitude is inversely proportional to the square of the distance, i.e. if the distance is doubled then the amplitude from the second reflector is one quarter of the amplitude of the nearer (12dB less).

20

30

40

50 * Further information may be gained by reviewing Huygen’s Principle.

60

70

SIDE LOBES* Side lobes are secondary lobes to the primary ultrasonic beam (or main lobe) that are formed at the face of a transducer and radiate away from the main lobe. They represent areas of high and low acoustic intensities and may cause unwanted echoes to be received by the probe, especially on rough surfaces, which may be mistaken for flaws on the screen. For shear wave probes, the minimum refracted beam angle in steel is approximately 33° to 35°, but at these relatively acute angles, side lobes may be formed which, although usually negligible, may cause spurious indications on the screen. For this reason it is usually safer to set the minimum beam angle for shear wave probes in steel at 38°. The narrower the main lobe, i.e. the smaller the half-angle of the beam, the weaker and more numerous the side lobes.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT2-3

UNIT UT2  THE PROPAGATION OF SOUND THE ULTRASONIC PULSE

NOTES

10

20

In a modern ultrasonic pulse echo flaw detector the pulse of ultrasound is created by charging a capacitor in the circuitry then suddenly releasing this charge of electrical energy, about 1Kv to 2Kv, into the probe. This electrical energy is converted into a mechanical vibration by the piezo electric crystal in the probe. The ultrasonic vibrations are formed by the collapse of the crystal after the electrical energy has been removed. The behaviour of the crystal, on collapse, can be likened to the behaviour of a spring when it is stretched then released. The spring will return to its former shape, then shorten, then stretch, etc. until it finally comes to rest in its original shape. This cycle of expansion and contraction is what forms the ultrasonic pulse. This is usually in the order of 15 cycles per second for an undamped crystal.

30

40

50

60

70

Pulse length An undamped pulse of 15 cycles per second would be unacceptable since in order to show separate, clear reflected signals on the CRT, the pulses of sound must be short and sharp. To shorten the pulses the ultrasonic crystal must be damped with a backing medium which absorbs the sound energy (in much same way as a shock absorber fitted to a spring on a motor vehicle dampens the vibration of the suspension). In this way the pulse length can be reduced to between 3 and 5 cycles.

80

90

100

The ideal pulse length would be approximately two cycles but such levels of damping are difficult to achieve with conventional backing mediums and commercially available crystals.

© Institution of Mechanical Engineers Issue 17 2016

UT2-4

UNIT UT2  THE PROPAGATION OF SOUND DAMPING, then controls PULSE LENGTH (number of cycles x wavelength).

NOTES

10

The other factor that controls pulse length is probe frequency. The higher the frequency the shorter the wavelength, i.e. the length of each cycle in the pulse and hence the shorter the pulse length (containing the same number of cycles). PULSE LENGTH controls RESOLUTION.

20

THE HIGHER THE FREQUENCY THE SHORTER THE DEAD ZONE

RESOLUTION 30

40

Resolution is the ability to separate on the timebase two or more reflectors that are close together in terms of beam path length. Consider the two reflectors within the beam with a beam path (length) difference of 3mm. If the pulse length was greater than the beam path length between the reflectors, then the signals from the two reflectors would be contained within the same envelope, Fig. A. The pulse length must be equal to or less than half the beam path length before the signals separate (resolve) Fig. B.

50

60

70

The above therefore demonstrates that the shorter the pulse length, the better the resolution. PULSE REPETITION FREQUENCY (P.R.F)

Note: P.R.F. is sometimes called timebase frequency.

80

90

100

The Pulse Repetition Frequency (P.R.F.) or Pulse Repetition Rate (P.R.R.) is the number of pulses of ultrasonic energy that leave the probe in a given time (usually per second). Each pulse of energy that leaves the probe must return before the next pulse leaves, otherwise they collide causing "ghost" or spurious echoes to appear on the CRT. 

The time taken for the pulse to travel from the probe and return is known as the transit time.



The time between pulses leaving the probe is known as the clock interval.

Therefore it can be stated that the transit time must be shorter than the clock interval or ghosting occurs. Practically speaking the clock interval should be around five times the transit time. Most digital flaw detectors allow the manual selection of the PRF.

© Institution of Mechanical Engineers Issue 17 2016

UT2-5

UNIT UT2  THE PROPAGATION OF SOUND NOTES

TRANSIT TIME (µsec)  10

DISTANCE TRAVELLED (mm) VELOCITY (km/s) CLOCK INTERVAL : Minimum  TRANSIT TIME Practical  5 x TRANSIT TIME

CLOCK INTERVAL (sec) 20

1 P.R.F. (MHz)

Ref Appendix F for example calculation.

30

MODES OF PROPAGATION BULK WAVES

Note: Compression waves are produced in steel if the incident angle of the beam in perspex is less than approximately 27.4°. Refer to critical angles page UT3-4

40

Compression or longitudinal waves (Alternative name – Dilatational) Probes that produce compression waves will normally have an incident and refracted angle of, or close to, 0°.

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT2-6

UNIT UT2  THE PROPAGATION OF SOUND Shear or transverse waves (Alternative name – Torsional)

NOTES

Note: Shear waves only are produced in steel if the incident angle of the beam in perspex is between approximately 28° and 56°. Refer to critical angles page UT3-4

10

20

30

40

Shear waves are normally produced by mode conversion at an interface. 50

See also UT3-3. BOUNDARY WAVES

Love waves A type of surface (Rayleigh) wave radiating from an epicentre like that of an earthquake is often described in references as a horizontally polarised surface wave.

60

These forms of propagation can only occur when a solid to gas interface is present. If the objects were immersed, these modes would be fully attenuated.

70

Surface or rayleigh waves Surface waves are formed when shear waves refract to 90°. The whiplike particle vibration of the shear wave is converted into an elliptical motion by the particles changing direction at the interface with the surface.

80

These waves are not often used in industrial N.D.T. although they do have some applications in the aerospace industry. Their mode of propagation is elliptical along the surface of a material, penetrating to a depth of one wavelength. They will follow the contour of a surface and they travel at approximately 90% the velocity of shear waves.

90

Where sharp changes in contour occur, such as a corner edge, reflected energy will return to the probe. 100

© Institution of Mechanical Engineers Issue 17 2016

UT2-7

UNIT UT2  THE PROPAGATION OF SOUND NOTES Lamb wave Velocity varies with frequency, mode, angle of incidence, thickness etc. Usually limited to material 1 – 2 wavelengths thick.

10

Plate or lamb waves Plate waves are formed by the introduction of surface waves into thin plate material. They are a combination of compression and surface or shear and surface waves causing the plate material to flex by totally saturating the material. There are two types of plate waves:

20

30

40

50

FACTORS AFFECTING THE PROPEGATION OF SOUND 60

The propagation of ultrasonic waves in a material is dependant on the density and elastic properties of that material and the type of wave transmitted. The practical considerations which will affect propagation will include:

70

Homogeneous: uniform properties 80 2 Anisotropic: The grains are random in orientation and have different elastic properties in different directions. 1

90 Note: Velocity is sometimes denoted by the letter ‘c’ (constant velocity). See table in App D-1



the test material's grain size



attenuation (absorption and scatter effects)



acoustic impedance of the test material



characteristic impedance of inclusions



diffraction



lack of homogeneity1



anisotropic2 materials

ACOUSTIC IMPEDANCE Acoustic impedance (Z) is the resistance of a material to the passage of ultrasound. It is the product of the material density (ρ) and sound velocity (v). i.e. Z = ρv

100

It is the acoustic impedance difference between two different materials/mediums which governs the intensity of ultrasound reflected

© Institution of Mechanical Engineers Issue 17 2016

UT2-8

UNIT UT2  THE PROPAGATION OF SOUND NOTES

10

* This can be seen when placing the probe (with a perspex shoe) on the perspex insert in the No 1 block. Note: The ideal acoustic impedance of couplant should be in between the acoustic impedance of the probe and the acoustic impedance of the test material. The ideal thickness of the layer of couplant should be one quarter of the wavelength of sound through it.

20

from the interface between them. Conversely, the amount of ultrasound passing from one material to another depends on this difference between the two materials. This difference is expressed as the acoustic impedance ratio. Theoretically if an ultrasonic wave was passed through two materials, with the same acoustic impedance (1:1 ratio), in intimate contact, then no reflection would occur, i.e. 100% transmission of sound would occur*. In practice it is very difficult to achieve intimate contact without a coupling medium (see next section). The couplant would have a different acoustic impedance to the material and so would affect the amount of sound reflected. The amount of energy reflected at an interface can be calculated with the following formula:

30

It can be seen from the formula that: Where Z1 and Z2 are the 2

Some recently developed ultrasonic systems use no 40 couplant, these are known as air coupled systems and they use very powerful amplification and sensitive received circuitry. 50

 Z1 - Z2   x 100 % Reflected energy    Z1  Z2 

respective acoustic impedances of the two materials.

HIGH ACOUSTIC IMPEDANCE RATIO (e.g. 20:1) = MORE REFLECTED ENERGY LOW ACOUSTIC IMPEDANCE RATIO (e.g. 1:1) = MORE TRANSMITTED ENERGY

60

It can also be seen from the formula that the same amount of energy is reflected, regardless of which direction the sound is travelling across the interface. COUPLANT

70

80

90

100

Because of the very high acoustic impedance ratio of air to a solid material, almost 100% of the energy is reflected at an interface between them (the basis of flaw detection). Therefore to enable the sound energy to transmit more readily into the test specimen we have to exclude any air that may be present between the probe and test surface. This is achieved by substituting the air with a material that has a closer acoustic impedance ratio to the probe and test material. This is known as a couplant. Common couplants are: glycerine.

water, oil, grease, polycell, swarfega and

The selection of couplant is sometimes based on the post-test use of the material being tested, e.g. water based couplants may cause rusting or corrosion but are easier to clean off in preparation for painting or coating when compared to oil or grease, which may actually protect the material from corrosion. Viscosity of the couplant may also be a consideration, ideally rough surfaces require a more viscous couplant to effectively fill the air gaps more uniformly. Whatever couplant is used for calibration/setting the search sensitivity, must be used throughout the subsequent inspection. © Institution of Mechanical Engineers Issue 17 2016

UT2-9

UNIT UT2  THE PROPAGATION OF SOUND ATTENUATION

NOTES

10

The two main causes of attenuation are SCATTER and ABSORPTION. It is expressed as energy per unit distance travelled and quantified in dB/ mm or dB/m*.

*Alternative unit is the ‘Neper’ 20

30

40

50

60

The sound intensity that is being compared is an average of the smallest audible sound. This is typically heard when approx 7 – 12 years old.

Attenuation is defined as the loss in intensity of the ultrasonic beam as it passes through a material and is dependant upon the physical properties of the material.

70

Refer to appendix F for example calculation. Scatter This is the major cause of attenuation and is the redirection of the sound waves reflecting off grain boundaries, porosity and non-metallic inclusions, etc. and becomes more apparent on the inspection when the size of grains become 1/10th of the wavelength of the search unit being employed. Absorption As the sound travels through a material a small amount of the energy is used up by the interaction of the particles, as they vibrate, causing friction which is dissipated as heat. As the frequency of the sound is increased the attenuation increases due to more particle vibration (absorption) and increased sensitivity to small reflectors (scatter from grain boundaries, porosity and inclusions) which are related to the wavelength of the sound. Materials such as castings and austenitic stainless steel are highly attenuative due to their coarse grain structures, etc. The attenuation factor/coefficient of a material can be measured and is expressed in dB/mm (see the appendices for an example). Natural attenuation also occurs due to the divergence of the beam in the far zone, i.e. assuming compression probe use, the amplitude of the backwall echo will be halved (-6dB) every time the distance from the probe is doubled.

THE DECIBEL (dB) The decibel is a logarithmic base unit used to compare sound intensities. Because we do not know the actual energy being transmitted by a probe, we can only compare sound intensities being received and express them as a ratio, e.g. twice as much, ten times as much etc.

80

A change in sound intensity, expressed in dB, can be measured by comparing signal heights on a calibrated screen. The change in dB is given by the formula:

dB  20 log10

H1 H2

Where H1 and H2 are the respective signal heights.

90

By transposing the formula it is possible to determine the ratio of the signal heights when the dB difference is known.

100

The gain/attenuator controls on a conventional ultrasonic flaw detector are calibrated in decibels, i.e. if we reduce the intensity of ultrasound by 6dB, any signal on the screen will drop to half its original height.

© Institution of Mechanical Engineers Issue 17 2016

UT2-10

UNIT UT2  THE PROPAGATION OF SOUND If we reduce or increase the intensity by 20dB, the signal will reduce to a tenth or increase by ten times its original height respectively.

NOTES

10

It is important to note that on certain flaw detectors, if reject or suppression is used to remove small unwanted signals from the display then the linearity of the amplifier, and hence the other signals, will be adversely affected, i.e. a 6dB drop will not reduce the signal by 50%. Table of approximate dB drops:

20 Often the inertia noise level in cars are compared so a 6dB difference between cars means the 30 loudest level is double the quietest.

dB

H2

Drop

H1:H2 ratio

20 14 12 10 6 2

10% 20% 25% 33% 50% 80%

90% 80% 75% 67% 50% 20%

10:1 5:1 4:1 3:1 2:1 5:4

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT2-11

UNIT UT3  SOUND GENERATION SOUND GENERATION

NOTES

10 The selection of a material suitable for producing ultrasound and receiving the resultant pulse back is based on three parameters: i. sensitivity ii. resolution iii. efficiency

Electrical pulse from flaw detector is very short negative going (usually) square wave.

This is defined as the property of certain crystals to convert electrical energy into mechanical energy and vice versa. These crystals may be naturally occurring, artificially manufactured or grown in solution.

20

30

-200v

100^S Typical

THE PIEZO EFFECT

40

Piezo electric crystals These crystals may be X-cut or Y-cut depending on which orientation they are sliced from the crystal material. The crystals used in ultrasonic testing are X-cut due to the mode of vibration they produce (compressional). This means that the crystal is sliced with its major plane (the crystal face) perpendicular to the X axis of the crystal material.

See also UT 4-4 band width. 50

60

Fundamental frequency is also known as the resonance frequency and is the lowest frequency the body/material/object will resonate at.

70

80

*May be re-arranged so the thickness is equal to ½ the Wavelength.

The frequency of the crystal is determined by its thickness and its acoustical velocity and can be calculated with the formula:

V* Ff  2t

Where Ff = Fundamental frequency V = Crystal material velocity t = Crystal thickness

From the formula it can be seen that the thinner the crystal, the higher the frequency.

Piezo electric crystal materials 90

Natural Quartz Tourmaline

Artificially grown Lithium Sulphate (LiSO4)

Manufactured ceramics Barium Titanate (BaTiO3) Lead Zirconate (PbZrO3) Lead Zirconate Titanate (PZT) Lead Metaniobate (PbNb2O6)

100

© Institution of Mechanical Engineers Issue 17 2016

UT3-1

UNIT UT3  SOUND GENERATION Properties of piezo electric materials

NOTES

10 The limitations of modern ceramic crystal materials are that they have low mechanical strength, i.e. they are brittle, and they have a tendency to age. 20 The advantage however is that they are excellent generators of ultrasound.

Crystal material Quartz Lithium sulphate

Barium Titanate

Lead Zirconate 30

* Polarisation to align crystals in a uniform direction 40 The Curie temperature for Barium Titanate is around 100°C to 120°C, although the piezo electric properties of Barium 50 Titanate will start to degrade at temperatures of 70°C and above. The primary reason standard probes are not usually used on materials above 50°C is because of 60 the possibility of degradation of the crystal. The secondary reason is due to the probe shoe characteristics beginning to change, altering velocity and therefore the beam 70 angle on shear wave probes. 1

Specular: Mirror-like.

80

Lead Zirconate Titanate

Advantages Stable Good wear resistance Best received and easily damped Best transmitter and good piezo electric properties May be performed to focus beam Good piezo electric properties Good transmitter and all round properties

Limitations Poor piezo properties

electric

Soluble in water

Temperature critical

Poor silvering

The *polarisation of ceramics In their natural state the polycrystalline ceramic material's crystals are randomly orientated and the piezo electric properties cancel each other out. To polarise these ceramics they are heated up to their Curie temperature and subjected to an electrostatic field. The crystals align themselves with the direction of the field, which is maintained during cooling. This polarised ceramic material then behaves as a piezo electric transducer until heated again to its Curie temperature. The most common crystal materials in use are Barium Titanate and Lead Zirconate Titanate. REFLECTION, REFRACTION AND SNELL’S LAW Reflection Ultrasonic waves are reflected by objects or interfaces placed in their path. When striking a specular1 reflector the angle at which this reflection takes place is governed by the law of reflection, which states: Angle of incidence = Angle of reflection Refraction This describes what happens to an ultrasonic beam when it passes from one medium to another where the two media have different acoustical velocities, e.g. from perspex to steel. The beam changes direction or angle in the vertical plane.

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT3-2

UNIT UT3  SOUND GENERATION NOTES

Snell’s Law is taken from 10 the laws of optics/light. A change of velocity from one medium to another is required to allow refraction to occur. Note: If V1 remains constant as V2 increases, the larger the resultant refracted angle will be.

Snell's law The relationship between the incident angle and refracted angles is governed by Snell's law that states: Sin  V1  Sin  V2

Where: 

= = V1 = V2 =



incident angle refracted angle velocity in medium 1 velocity in medium 2

20

MODE CONVERSION Mode conversion can be thought of as an acoustic mirage, an example can be detailed during the critical root scan detailed below. 30

40

50

At the critical root stand-off position Probe A has the leading edge of the beam spread deflected by the lack of side wall fusion on to the bottom face of the plate. If the orientation of the lack of side wall fusion compliments the leading edge of the ultrasonic beam it will strike the bottom face normal (at 90 ͦ) and reflect back towards the lack of side wall fusion and hence back to the probe. As the echo is reflected from a smooth reflector, i.e. the bottom surface of the plate, then the echo dynamic pattern may appear to be lack of root fusion on the ultrasonic screen. However the echo will plot on to the far side of the weld which will indicate that it is from some form of geometrical anomaly. The nature of the true position of the indicator can be proved by either using a different angle probe (different refracted angle striking the reflector), or preferably go onto full skip (position B), to strike the sidewall at 90° to eliminate the chance of mode conversion completely.

60

70

Echo from Probe A 80

The X to Y ratio 90 1:1.84 is a function of the difference in the respective shear wave velocity to compression velocity. 100

Echo from Probe B

© Institution of Mechanical Engineers Issue 17 2016

The beam path length of the leading edge reflected signal to the bottom plate (Path Y) is plotted as if it is travelling to the root (Path X) i.e. X = Y

UT3-3

UNIT UT3  SOUND GENERATION NOTES

10

20

Mode conversion requires the confluence of many factors, the initial sound wave may be mode converted into a compression, shear or surface wave. As with pulse echo ultrasonic it is hard to predict where any echo is coming from after the wave front has hit its first major reflector due to mode conversion altering sound velocities from the prescribed calibration velocity. The difficulty occurs when the 1st major reflector is orientated at an obtuse angle to the major axis of the sound beam and little or no sound energy is directly reflected back to the probe hence always use a plotting system and look along the beam to see if it was passing through the sidewall at an obtuse angle to get to the plotted point where ever possible. DIFFRACTION

30

This occurs when sound waves pass the tip of a narrow reflector. Some of the sound scatters off the tip causing waves in different directions that reinforce or cancel out the original waves. This results in a series of high and low intensity waves radiating out from the tips, giving the impression of sound bending around the edges of the defect.

40

50

* Delta Technique: a transmitting angle probe with a receiving compression probe. 60

70

It is this effect that is used in both Time of Flight Diffraction (TOFD) and Delta Techniques in ultrasonic testing. Reflectors should be larger than 1 wavelength for significant diffraction to occur. Delta Technique Smooth shaped vertical linear indications can be very difficult to detect by ultrasonic as most of the sound will be deflected away from the transmitting probe. The Delta techniques uses a shear wave probe to transmit the sound and an alternative probe, usually a compression probe, to receive. In effect it is used as a twin crystal probe with the crystals being separated and independently positional.

80

90

Whilst accurate depth and positioning is not possible the indication can be approximately located by manipulating probe B until the signal is maximised. Manipulation of probes A & B can give an estimation of the flaw extent. If the Delta Technique is used for detecting an indication at a known depth, such as lack of root penetration in a double ‘V’ weld prep, then using an existing stand off from a trigonometry set up would give better results. The screen capture from the clear section shows no received signal on the ultrasonic trace whilst the second screen capture shows sound diffracted from lack of root penetration located at the centre of a double ‘ V ‘ preparation.

100

© Institution of Mechanical Engineers Issue 17 2016

UT3-4

UNIT UT3  SOUND GENERATION NOTES

10

20

30

40

Delta Technique from clear section in a double ‘V’ weld.

50

60

70

80

90

Delta Technique from lack of penetration in a double ‘V’ weld.

An alternative and complimentary technique that can be used in conjunction with the delta technique is the pitch and catch system. 100

© Institution of Mechanical Engineers Issue 17 2016

UT3-5

UNIT UT3  SOUND GENERATION CRITICAL ANGLES

NOTES

These are the incident angles in the first medium at which the refracted angles in the second medium change over from one waveform to the next. The first critical angle is where the refracted compressional wave is just about to disappear, leaving only shear waves in the second medium. The second critical angle is where the refracted shear wave has changed to a surface wave.

10

The critical angles can be calculated using Snell's law. 20

α

0°

30

V1

V1

V2

V2

C C

40

S Dia 1

β

At 0° the energy is partially reflected and partially transmitted across the boundary.

50

Dia 2

Increasing angle α produces a compression wave in V2 (C) and also produces a weak shear wave.

1st critical angle 60

1st C

2nd critical angle

α Reflected

70

C

90°

V1

V1

V2 Shear

V2 Shear

80

Dia 3

90

When α reaches the 1st critical angle the compression wave refracts through 90° this leaves shear waves in V2. Note the compression wave does not form a surface wave. ≃ 27° Perspex to Steel.

Dia 4 Between the 1st and 2nd critical angles we have shear waves in V2 and a reflected compression wave in V1.

100

© Institution of Mechanical Engineers Issue 17 2016

UT3-6

UNIT UT3  SOUND GENERATION NOTES

1st α

10

Reflected

Surface 20

30

40

2nd

2nd critical angle

V1

V1

V2

V2

Dia 5

Dia 6

At the second critical angle the shear wave refracts through 90° and becomes a surface wave. Surface waves travel in an elliptical motion at a velocity marginally less than the shear (approx 90%). As the surface wave has the smallest velocity (all other factors being equal) it has the shortest wavelength and therefore it is the most sensitive wave form. ≃ 56° Perspex to Steel.

When the 2nd critical angle is exceeded all conventional modes of propagation are reflected internally.

50

R L 60 S

S

Combination of both Long and Shear Waves 70

Only Shear Waves present

Only Surface Waves present

SHEAR

LONGITUDINAL

SURFACE

80

5°

10

15

20

25

30

35

40

45

50

55

60

65

70

75

INCIDENT ANGLE° 1ST CRITICAL ANGLE

2ND CRITICAL ANGLE

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT3-7

80

UNIT UT3  SOUND GENERATION Calculation of the critical angles for a Perspex to Steel interface.

NOTES

10

 = incident angle β= refracted angle v1 = compressional velocity in Perspex = 2740 m/s

Sin α 20 V1

v2 = velocity in steel, compressional = 5960 m/s shear = 3240 m/s

V2 Sin β

30

Sin α =

V

Sin β

V2

1st critical angle:

v1 x Sin  v2c

Sin  

1

40

Sin  

2740 m/s x Sin 90 5960 m/s

Sin   0.459731543 x 1

  27.4 2nd critical angle: 50

Sin  

v1 x Sin  v2s

Sin  

2740 m/s x Sin 90 3240 m/s

Sin   0.845679012 x 1 60

70

  57.7 At the first critical angle, compression and shear waves co-exist, so the lowest angle for shear waves (only in practical use) is just beyond the first critical angle, at an incident angle of 29°, which gives a refracted shear angle of 35°.

Sin  

v2s x Sin  v1

Sin  

3240 m/s x Sin 29 2740 m/s

Sin   1.182481752 x 0.4848096 80

  35

90

At the second critical angle surface waves exist so the highest incident angle we use for shear waves is 56° that gives an 80° shear wave.

100

© Institution of Mechanical Engineers Issue 17 2016

UT3-8

UNIT UT3  SOUND GENERATION Therefore the range of shear wave probe angles in steel (for practical purposes) are 35° to 80°, produced from incident angles of 29° to 56° in perspex.

NOTES

α

10

20

β 30

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT3-9

UNIT UT4  EQUIPMENT EQUIPMENT

NOTES

10

20

* May also be referred to as twin crystal probe.

PROBES The angle of a probe used in ultrasonic testing is measured from a line drawn perpendicular to the test surface. This line is known as the normal. A 0° probe is one which transmits sound at 90° to the test surface. Also known as a normal probe, this probe usually transmits compressional or longitudinal waves. A 60° angle probe would transmit sound at 60° to the normal, i.e. 30° from the surface. The most common angle probes transmit shear waves (although angled compression probes do exist for special applications) and the manufacturers quote the angle of the probe for use on mild steel. 0° combined double* probe

30 Both compression and angle probes can be either single or twin crystal probes. 40 On combined double probes crystals would normally be angled towards each other (roof angle) to ensure beam returns to receiver crystal. These probes thus have a ‘focal distance’ and are limited to thinner sections.

50

60

Double probes have two crystals, one transmits and the other receives ultrasound. The cork separator in between the shoes prevents "crosstalk" or "chatter" between the crystals. Using oil as a couplant may eventually break down the acoustic barrier and produce spurious standing echoes on the display. Having separate crystals eliminates the dead zone1 on the display, enabling the detection of near-surface defects. These probes are therefore useful for testing thin sections, e.g. thickness gauging and examining for near surface flaws. The crystals may be focused to give a focal point at the ideal beam path range to be examined.

70

Single crystal angle probe

80 1 Dead zone: Ringing time of crystal.

Some probes use an acoustic lens which shifts the location of the focal point, commonly used in immersion testing.

90

100

Single crystal probes have one crystal that transmits and receives ultrasound. The flaw detector controls the process by transmitting a pulse of energy then switching the circuit to receive, listening for any returning sound, in between pulses. The circuitry can be switched quicker than the crystal can be damped. So the receiver picks up the

© Institution of Mechanical Engineers Issue 17 2016

UT4-1

UNIT UT4  EQUIPMENT last few vibrations of the crystal, as it switches on, and displays them on the screen as the dead zone. This eliminates the possibility of detecting near-surface defects.

NOTES

10

Angle probes have a perspex shoe, on which the crystal sits, that can be machined to any angle. The angle of the wedge determines the angle that the ultrasound strikes the interface (incident angle). This in turn, according to Snell's law, controls the angle that the sound will propagate through the test material (refracted angle).

20

Damping material on the back of the crystal (also known as a backing slug) controls the length of the ultrasonic pulses by absorbing the sound energy, producing short sharp pulses. The length of the pulse is the main factor in determining the resolution of the equipment. The most common damping/backing medium is Tungsten Araldite.

30

SHORT PULSE LENGTH/WIDTH/DURATION MEANS GOOD RESOLUTION. Soft nosed probe

40

50

60

70

These are often sub-divided into two divisions: 1)

Bubbler with a small water gap flowing under low pressure.

2)

Squirters with larger water gaps under high pressure.

This has a soft diaphragm mounted on the front of the crystal, clamped in place by a threaded ring, the space in between the diaphragm and the crystal being filled with couplant to expel any air. The soft diaphragm follows the contour of the surface under test, making this probe ideal for rough or uneven surfaces, e.g. castings or rough machined components. Water gap or gap scanning probe

80

90

100

This consists of a water jacket with a nozzle at the end and a probe inside. Water is fed into the jacket and flows out through the nozzle, forming a column of water, to the test surface, through which the © Institution of Mechanical Engineers Issue 17 2016

UT4-2

UNIT UT4  EQUIPMENT NOTES

10

sound can travel. Because of the flexibility of the coupling medium (water) the probe can be used on rough or uneven surfaces. These probes are usually used in automated ultrasonic scanning systems and can be set up, using a guide wheel to follow the contour of a component. They can also be used in arrays to scan a wider area. Often used for ‘through transmission’ testing of composites. Wheel type probe

20

30

40

50

60

In this probe the crystal is within the axle of the wheel and the sound travels through the soft tyre into the test material. The spring loaded joint allows the probe to follow the contour of the surface so it can be used on rough or uneven surfaces. It is used in a similar way to the water gap probe. The main advantage of this type of probe is that it removes the requirement of externally applied couplant, mainly used in aerospace industries. Delay line probe

70

80

90

100

The delay line probe is very similar in construction to the soft nosed probe. The difference is that it has a long perspex shoe clamped in instead of a diaphragm. The length of the shoe extends the time taken for the echo from the front surface, of the material under test, to return to the crystal. This places the Front Surface Echo (FSE) further along the timebase, i.e. beyond the dead zone. It enables near surface defects to be located or thin plate to be tested using a single crystal

© Institution of Mechanical Engineers Issue 17 2016

UT4-3

UNIT UT4  EQUIPMENT NOTES

10

probe. These probes are usually high frequency probes (which means they have a small dead zone), but high frequency = long near zone. Therefore, to use them for near surface flaw detection/sizing the long shoe is used to contain the near zone in the probe not in the test material. Length of the delay line is important to avoid internal reflections interfering with the back wall signal. Magnetostrictive transducers

20

30 Other styles of transducers include paint brush/ phased array/ mosaic/ EMA, but are not covered here due to specialist usage.

40

50

60

Used for detecting defective bar stock, the transducer coil has a magnetic field that is switching at ultrasonic frequency. This field causes the bar stock to vibrate at an ultrasonic frequency and the vibrations travel along the length of the bar. When the vibrations reach the other end of the bar, they reflect back and are picked up by the transducer (in receive mode) and registered on the detector. The equipment is calibrated off a defect free piece of bar stock to register a specific value on the detector and defective bar stock is recognised by a change in this value. PROBE FREQUENCY, BANDWIDTH AND DAMPING

The frequency stated on the probe is known as the central operating frequency. This is the frequency of the highest output of sound from the probe.

70

80

90

An ultrasonic probe transmits sound at a range of frequencies, not just at the stated frequency, known as the bandwidth. For example a 5MHz probe may produce a frequency range of 4 to 6MHz. The bandwidth is also an indication of the damping factor.

Broad Band Probes (low Q)

Narrow Band Probes (high Q)

They are highly damped Have a short pulse length (typically 1 to 2 cycles) A short ringing time (dead zone) Better resolving power Poor penetration

They have low damping A longer pulse length (typically 3 or 4 cycles) A long ringing time (dead zone) Poor resolution Good penetration

100

© Institution of Mechanical Engineers Issue 17 2016

UT4-4

UNIT UT4  EQUIPMENT NOTES

Ff

Peak Q = Quality factor Q

=

FL

10

Ff

FH

Usually 70% of peak

FH - FL MHz Ff = fundamental frequency

2

3

20

4

5

3

6

4

Narrow bandwidth

5

Broad bandwidth

PROBE SELECTION 30

40

The selection of probes for ultrasonic inspection is influenced by various aspects of the test and the particular material under test. These may include: 

The type and size of defect being sought



The type of material under test



The distance the sound has to travel through the material

Probe angle is another consideration when searching for defects at different orientations throughout the material. 50

Below is a table of properties of probes using the two criteria that we can select i.e. frequency and diameter. Effect of frequency

60

70

80

90

100

Low Frequency

High Frequency

Long wavelength More beam spread Shorter near zone Better penetration Less attenuation Longer dead zone Less sensitivity Less sensitive to orientation of reflector

Short wavelength Less beam spread Longer near zone Less penetration More attenuation Shorter dead zone Higher sensitivity More sensitive to orientation of reflector

Effects of diameter Large Diameter

Small Diameter

Less beam spread Longer near zone Better penetration Less attenuation (due to beam spread) Difficult coupling on curved surfaces More coverage on flat surfaces

More beam spread Shorter near zone Less penetration

© Institution of Mechanical Engineers Issue 17 2016

More attenuation Easier coupling on curved surfaces Less coverage on flat surfaces

UT4-5

UNIT UT4  EQUIPMENT Another consideration is whether to use a single crystal or a combined double crystal probe.

NOTES

10

20

30

40 Note: some blocks are available in metric or inch sizes and it should be verified before use which version is at hand. 50

The current British Standard for this ultrasonic calibration block is BS EN ISO 2400 (which replaces BS EN 12223) which refers to it as Calibration Block No.1. And the removal of the Perspex insert although the 50mm hole remains

60

The advantages of a single crystal probe are; better penetration, for the same size probe as a double, because the effective transmitter crystal diameter is larger, no focal point, i.e. it works effectively over a longer range and cost (cheaper). The main advantage of a double crystal probe, is that there is no dead zone on the screen, this means better near surface resolution can be achieved. It can be seen from the tables that higher frequency probes have a higher sensitivity. In this context, sensitivity refers to the ability to detect small defects. The higher the probe frequency the smaller the wavelength and the smaller the size of reflector the probe can detect. It is generally accepted that the smallest reflector a probe can detect is half the probe's wavelength. So a probe with a long wavelength (low frequency) will not detect small reflectors, such as small defects or grain boundaries, therefore the sound will penetrate further through the material because it is not reflected at these small interfaces. CALIBRATION BLOCKS AND THEIR USES Tolerances: Wherever practical the limits on dimensions should be ±0.1mm. Materials: Steel blocks are made from low or medium carbon ferritic steel (killed), normalised to produce a fine grained homogenous structure throughout. The International Institute of Welding (I. I.W.) block Also referred to as Block No.1, A2, V1, DIN54/120 or Dutch block.

70

80

90

100

0° COMPRESSION PROBES Calibration 0° probe calibration can be set using Back Wall Echoes (BWE) off the various thicknesses available, i.e. 5, 10, 25, 100 and 200mm. It can also be checked (rough) on the 23mm thick perspex insert which gives a reading of 50mm when calibrated on steel (the ratio of sound

© Institution of Mechanical Engineers Issue 17 2016

UT4-6

UNIT UT4  EQUIPMENT NOTES

A6 block: Determination of the dead zone using an A6 block will give a more accurate result. See App G for drawing (ref BS 2704 obsolete). A Block should be at same temperature as test item as sound velocity changes with temperature. Perspex having much bigger changes than steel.

A7 block: An A7 block will give a more quantifiable value for resolution regardless being a 0° or angle probe. See App G for drawing (ref BS 2704 obsolete).

10

20

30

40

50

60

70

*An A5 block (IOW) with 1.5mm side drilled holes over various depths will give a far more accurate probe angle or beam spread profile.

80

90

velocity in steel to the velocity in perspex is 5960m/s to 2740m/s = 50:23). A minimum of two echoes are required for calibration with 0° probes. The 91mm step in the block serves to calibrate the screen for use with shear wave probes by using a compression probe. If a 0° probe is placed over the 91mm and the echoes placed at 5 and 10 on the graticule then the screen is calibrated for a range of 0 to 182mm compressional. This is equivalent to 0 to 100mm shear, the ratio of the velocities of compression to shear waves is 1.82:1 (5960m/s:3240m/s). Dead zone measurement (single crystal probe) Place the probe over the 5mm section. If the signal is visible outside the dead zone then the dead zone is less than 5mm. If the signal is not visible then place the probe on the 10mm section. If the signal is now visible then the dead zone is greater than 5mm but less than 10mm. If the signal is still not visible then go on to the 15mm deep hole. This procedure can be carried out with an uncalibrated screen. An alternative method would be to calibrate the screen and read the length of the dead zone off the flaw detector graticule. Resolution The resolution of a 0° probe can be checked by using the three different thickness sections around the slot below the centre of the 100mm radius. Place the probe above the slot and with a calibrated screen note the separation between the 85, 91 and 100mm signals. SHEAR PROBE USES Index or sound exit point Place the probe on the top of the block over the centre of the 100mm radius, with the beam travelling toward the radius. Maximise the signal by moving the probe back and forth, stopping at the point where the signal is highest. Mark the position of the small slot in the block, onto the probe, to represent the point where the centre of the sound beam is leaving the probe. The engraved lines either side of the small slot (and the ones on the probe) can be used to measure the movement of the index point as the probe shoe wears down. Shear probe calibration This can be carried out using the 100mm radius, repeat signals being secured by the small slot used for indexing. Shear probe angle check Maximise the reflected signal from the 50mm diameter (side) of the Perspex, insert and note the position of the probe index in relation to the engraved graduations on the block, to read off the approximate angle. A more accurate check can be made using the reflection from the 1.5mm diameter hole in the same way*. Shear probe output Maximise the signal from the 100mm radius and adjust to full screen height using the gain and note the dB figure indicated on the controls. This figure can be used to compare different probes or to check the probe in use (daily) for deterioration.

100

© Institution of Mechanical Engineers Issue 17 2016

UT4-7

UNIT UT4  EQUIPMENT BLOCK NO.2, A4, V2, DIN54/122 KIDNEY BLOCK

NOTES

The current British Standard for this calibration block is BS EN ISO 7963 (2010).

10

20

12.5mm, 20mm or 25mm

30

40

COMPRESSION PROBE USES

50

Calibration This block can be obtained in various thicknesses, although the current standards in use for ultrasonic calibration blocks may only mention 12.5, 20 or 25 mm. The repeat signals secured from this through thickness can be used to calibrate the 0° probe. SHEAR PROBE USES

60

70

Dead zone measurements, resolution and shear probe angle checks can only be approximated on the A2/A4 blocks. Specific blocks such as the A5, A6 and A7 should be used for more accurate reproducible results as quoted in associated standards.

80

90

Probe calibration With the probe aiming towards the 25mm radius, signals occur at; 25mm, 100mm, 175mm, 250mm, 325mm, 400mm, etc. With the probe facing the other way, toward the 50mm radius, the signals occur at; 50mm, 125mm, 200mm, 275mm, 350mm, 425mm, etc. To calibrate; the radius which gives the easiest signals, within the range selected, to align on the graticule should be selected. Index or sound exit point Using the 25mm or the 50mm radius, maximise the reflected signal and mark the position of the central graduation (the centre of the radii) onto the probe. (It is recommended however that block no.1 is more accurate for this check). Probe angle check Maximise the echo from the drilled hole and check the angle from the position of the index point.

100

© Institution of Mechanical Engineers Issue 17 2016

UT4-8

UNIT UT4  EQUIPMENT INSTITUTE OF WELDING (I.O.W) / A5 BLOCK

NOTES Terminology 10 Calibration block: manufactured to known sizes and used to set the time base range. Reference block: Ideally manufactured from the same material as to be tested with known size 20 and type of reflectors used to set sensitivity. Reference standard block: a combination of calibration and reference blocks (i.e. A5 block)

30

*Due to this block having SDH (side drilled holes) it40 is also capable of being used to produce a DAC curve.

50

This block can be used as a calibration block with a compression probe, however, its main use is as a reference block with either compression or shear wave probes. Its two most common uses are for plotting the beam profile and for setting test sensitivity, using the various individual side drilled holes as reference reflectors*. The five side drilled holes on one side of the block that are drilled close together may be useful to check the resolution capabilities of angle probes and to confirm accuracy of 20dB beam profiles against known ‘size’ reflectors (not shown on diagram above). ULTRASONIC FLAW DETECTORS

60

Modern ultrasonic flaw detectors are small, portable, microprocessorbased instruments suitable for both shop and field use. They generate and display an ultrasonic waveform that is interpreted by a trained operator, often with the aid of analysis software, to locate and categorize flaws in test pieces. They will typically include an ultrasonic pulser/receiver, hardware and software for signal capture and analysis, a waveform display, and a data logging module.

70

80

Thickness meter

Digital Flaw Detector

90

100

Analogue Flaw Detector © Institution of Mechanical Engineers Issue 17 2016

UT4-9

UNIT UT4  EQUIPMENT While some analog-based flaw detectors are still manufactured, most contemporary instruments use digital signal processing for improved stability and precision.

NOTES

10

EQUIPMENT CHECKS Periodically ultrasonic flaw detection equipment must be checked to ensure performance characteristics have not deteriorated. 20

Some of the checks include: 

30 The current British Standard for equipment checks is BS EN 12668: Part 3. 1

Timebase linearity



Amplifier linearity



Timebase range calibration



Signal to noise ratio



Angle probe index point



Probe angle check



Beam profile determination



Resolution check

40

This is not a comprehensive list, the recommended checks to be carried out can usually be found in the relevant current standards1.

The position can change due to the echo width changing with height. To ensure the reading is consistent maintain the same height for each reading

The tolerance for timebase linearity in BS EN 12668-3 is ± 2% of the whole timebase.

50

60

Timebase linearity Carried out over the ranges to be used, this is performed by placing a compression probe on a calibration block to obtain multiple echoes. Calibrate the screen by placing the first and last echoes, within the required range, in their correct respective positions on the timebase and check that the intermediate echoes are in their correct respective positions. The tolerance2 on linearity can be found in the current standard1.

2

70

Note: To BS EN 12668-1 it is ± 1% of the whole timebase.

80

Amplifier linearity Position a probe on a calibration block to obtain a signal from a 1.5 or 5 mm SDH. Using the gain control, adjust the signal to 80% FSH and note the dB level. Increase the gain by 2 dB and the signal should rise to 101% FSH (not less than 97%). Reduce the signal by 2dB to return to the 80% reference level. Reduce the gain by 6dB and the signal should fall to 40% FSH (37 – 43%) reduce the gain by another 6 dB and the signal should fall to 20% FSH (17 – 23%). Reduce again by 6dB and the signal falls to 10% FSH (8 – 12%) and again by a final 6dB so the signal falls to 5% FSH (visible below 8%).

90

If the signals are within the tolerance the amplifier is said to be linear to BS EN 12668-3.

100

© Institution of Mechanical Engineers Issue 17 2016

UT4-10

UNIT UT4  EQUIPMENT NOTES

dB 10

20

Digital flaw detectors have a self check capability which automatically runs when the set is initially turned on.

30

40

50

Expected echo height % FSH

Tolerance echo height % FSH

Ref

80

80

+2

100

Not less than 95

-2

80

80

-6

40

37 – 43

-6

20

17 – 23

-6

10

8 – 12

-6

5

Visible below 8

Timebase range calibration Check the ability of the equipment to be calibrated to the ranges required. Signal to noise ratio Place the probe on a calibration block to obtain a reflected signal from a transverse drilled hole. Using the gain adjust the signal to 20% of full screen height and note the gain setting (dB). Increase the gain until the grass (noise) level reaches 20% screen height at the same timebase position and note the new gain setting (dB). The difference in the two dB gain settings is the signal to noise ratio and can be used to compare different equipment or to monitor the equipment in use. For example, a high signal to noise ratio would give less noise on screen.

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT4-11

UNIT UT4  EQUIPMENT Angle probe index point This check is covered in the calibration block section of the notes.

NOTES

10

20 This will be covered in more depth during practical work

30

40

50

60

Probe angle check Also covered in the calibration block section, the more accurate check being when plotting the beam profile using the A5 block (see next paragraph). Beam spread Example: 20 dB drop beam spread (vertical) Although the beam spread can be calculated, it is usually plotted out practically using the A5 block and a range of different depths of reference holes. Before plotting the beam profile the probe index point should be checked. The probe is placed above one of the holes, then by moving the probe back and forth, the signal from the hole is maximised and the gain adjusted to give a signal at 100% full screen height. The position of the index point is then marked onto the block. The probe is then moved forward until the signal falls to 10% screen height and again the position of the index point is marked onto the block. The hole is now in the 10% (-20dB) intensity trailing edge of the beam and the distance between the two marks on the block represents the distance from the centre to the 20dB trailing edge of the beam at the depth of the hole. The procedure is then repeated in the opposite direction (backwards) to find the leading edge of the beam. This is repeated on several (a minimum of three over the intended working range to be used) different depths of hole to find the profile of the beam. The marks on the block can be transferred to a graph to give a pictorial representation of the beam and/or transferred to a plotting system for use in plotting and sizing defects.

70

80

90

100

Resolution This check can be found in the calibration block section of the notes. (See UT 2-5).

© Institution of Mechanical Engineers Issue 17 2016

UT4-12

UNIT UT5  0° PROBE SCANNING 0° PROBE SCANNING

NOTES

10

CALIBRATION The initial pulse, or main bang, is a test signal that the flaw detector creates and has no significance for calibration. It usually lies just off to the left of a calibrated screen.

20

30

40

50

60

70 * Dead zone – although with twin crystal probes no visible indication exists on the display screen, the term dead zone correctly means an area where indications cannot be located and is still present near surface due to the angle of the crystals used to focus the beam at the optimum usage distance.

When an ultrasonic probe is placed on to a piece of steel, some of the ultrasound in the probe reflects off the interface between the probe shoe and the steel and some is transmitted through into the steel. When the transmitted energy strikes the back surface of the steel it virtually all reflects off the steel to air interface and returns to the steel to perspex interface. Here some energy transmits into the probe and creates the first signal (1) and the rest reflects back inside the steel and the process repeats itself, creating the repeat signals, (2 etc.) until the energy decays away. The spacing between the echoes represents the thickness of the steel, so if we place the probe on an A2 block, on the 25mm thickness, then the echoes are 25mm apart. Note: If we are using a single crystal probe then the initial energy that reflected back into the probe will create a signal at the start of the screen (F) which will be very close to the initial pulse and there will also be a dead zone visible on the display screen. If we are using a double crystal probe (separate transmit and receive crystals) then there will be no signal from the front surface and no dead zone* visible.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT5-1

UNIT UT5  0° PROBE SCANNING TO CALIBRATE A 0° PROBE TO A RANGE OF: 0 TO 100 MM

NOTES

10

20

USING A DIGITAL FLAW DETECTOR In the calibration menu select the correct velocity for the calibration block (i.e. 5960m-s), apply couplant to the No 1 calibration block and obtain a signal at an appropriate depth (i.e. 25mm), using either the zero or probe delay control (depending on manufacturer) adjust the signal and position it at the correct place on the timebase. This can be set visually or for more accurate calibration use the gate control to obtain a digital read out of the signal depth. Always confirm the calibration on a second thickness. USING AN ANALOGUE FLAW DETECTOR

30

40

50

Apply couplant to the A2 block and place the probe on the 25mm thickness to obtain multiple echoes. We require a range of 100mm on the screen so four echoes would fit in to this range, we adjust the coarse range control to give us about four echoes on the screen. We then adjust the delay control to position the first backwall echo a quarter of the way along the screen and adjust the fine range control to position the fourth echo at the end of the screen. This procedure is repeated until all four echoes take up their respective positions (see sketch). The same basic procedure applies to different ranges using different thicknesses. By dividing the range by the thickness we can obtain the number of echoes required and by evenly spacing the echoes on the screen the desired range is achieved.

60

70

Digital flaw detectors need only be calibrated at the beginning of the inspection as the range control is linear and holds its calibration as the range is changed. However it is always good practice to check against a calibration block each time the range is altered.

80

ACCURATE MEASUREMENT USING AN ANALOGUE FLAW DETECTOR

90

100

For accurate beam path measurement, such as thickness surveying using a flaw detector, the achievable accuracy is determined by the range selection. For example if the range is set to 100 mm full screen, then each large graticule division is 10 mm and each small division is 2 mm. This means that the most accurate you could read the screen, by judging the halfway distance between the divisions, would be 1 mm. However, the manufacturers of analogue flaw detectors using a C.R.T. can only guarantee the horizontal (time base) linearity of the display to be within 2% of the whole time base. Meaning that an echo could be one small division (or 2 mm on the 100 mm range scale) out

© Institution of Mechanical Engineers Issue 17 2016

UT5-2

UNIT UT5  0° PROBE SCANNING of position so the guaranteed accuracy would normally have a tolerance of ±2% of the range (the same size as one small division).

NOTES

10

20

Note: Signal width 30 varies as echo height changes. Calibrate and read signal at same height. Although this may be negated using peak to 40 peak readings instead of the ‘flank’ position of the signal.

Timebase range 500 mm 200 mm 100 mm 50 mm 20 mm 10 mm

Large division 50 mm 20 mm 10 mm 5 mm 2 mm 1 mm

Small division 10 mm 4 mm 2 mm 1 mm 0.4 mm 0.2 mm

Read accuracy 5 mm 2 mm 1 mm 0.5 mm 0.2 mm 0.1 mm

MULTIPLE BACK WALL METHOD Another method of reading accurate thickness measurements is to use the multiple backwall method. This involves calibrating the screen to a larger range, reading the nth repeat signal from the thickness and dividing the reading by n (where n is the clearest signal that you can read the furthest along the screen).

50

60

70

In this example the 6th signal can be quite easily read off the screen at a beam path of 42 mm. Note: Can also measure 80 between any two echoes (echo – echo mode) in digital flaw detectors.

90

100

The thickness can be calculated by:

42  7mm 6 Different materials (digital and analogue) When testing materials other than steel, the velocity of the sound wave may be different. If this is the case, the difference in velocity between the material under test and the calibration block must be taken into consideration and used to compensate for the difference in readings obtained. Alternatively, a calibration block made of the same material as the test material must be obtained.

© Institution of Mechanical Engineers Issue 17 2016

UT5-3

UNIT UT5  0° PROBE SCANNING The following formula can be used to compensate when the flaw detector is calibrated using a steel calibration block:

NOTES

*Take note of the units of measurement used e.g. if units of distance are mm’s then velocity is mm/sec.

Actual Thickness 

10

20

If the sound velocity in a material is not known but the actual thickness can be physically measured, the velocity can be calculated by transposing the above formula thus:

Velocity 

30

Material Velocity x Timbase Reading Calibration Block Velocity

Actual Thickness x Calibration Block Velocity Timebase Reading

DEFECT DETECTION When using a 0° probe to search for defects we must consider the following. 

Which range should be used, for accuracy and through thickness coverage?



Probe selection, taking into account material attenuation and defect size.



What level of test sensitivity to use - ensuring defects which are considered harmful to the product (not necessarily all flaws are considered harmful) are located and to assure that reproducible test results can be obtained, by different operators, using different manufacturers equipment.

40

50

1

Note: Probe and range selection have been covered in previous sections1 of these notes, setting sensitivity is as follows.

Section UT4 - Equipment 60

SENSITIVITY There are various methods of setting the test sensitivity, these include: 70

80

90



Back wall echo level method (0° probes only)



Grass level



Using a reference reflector



Using a graph or curve plotted from reference reflectors (DAC/DGS)

Back Wall Echo (BWE) The backwall echo method involves coupling the probe to the test material and increasing the gain until the back wall echo is at the predetermined level. The level can be varied in several ways - if the second back wall echo (BWE) is set to full screen height (fsh) this would be more sensitive than setting the first BWE to fsh. Another way is to set the BWE to a lower level (less sensitive) or to set it to a percentage of fsh and add a pre-determined number of dBs to the gain (increase sensitivity).

100

The BWE method can obviously only be used with 0° probes since reflections off the back surface, when using angle probes, do not

© Institution of Mechanical Engineers Issue 17 2016

UT5-4

UNIT UT5  0° PROBE SCANNING NOTES *More information on attenuation and 10 transfer correction in section Appendix E.

20

30

40

50

return to the probe. It does have an advantage of automatically taking into consideration attenuation and transfer correction requirements for the test material*. Grass The grass or grain interference method involves coupling the probe to the test surface and increasing the gain until the reflections from the grain structure of the material reach a pre-determined level. This is often quoted as 2 mm - 3 mm in height at the maximum test depth but ideally should be referenced as a percentage of full screen height as not all flaw detectors use the same dimension screen. The sensitivity can be adjusted by increasing or decreasing the level, or by adding or subtracting dBs to or from the gain. This has the advantage of incorporating attenuation and transfer correction factors. Being more sensitive means you can monitor the grain structure for inconsistencies as you scan along. Reference reflectors A common method of setting sensitivity is to set a maximised signal from a reference reflector, at target depth, to a predetermined level, for example full screen height. The reference reflector could take the form of a known reflector e.g. a transverse side drilled hole, a flat bottom hole, a slot or a vee notch. Or it could be a real, or simulated, defect of known size and type. This technique allows easier consistency between tests/operators using a known manufactured reflector. However, the expense of the manufacture of the blocks and attenuations/transfer correction will have to be separately considered and adjusted for. GRAPHS AND D.A.C CURVES

60

70

Another common method of setting sensitivity involves plotting a graph or curve on paper or the flaw detector screen using transverse or flat bottom holes. One of these is known as the Distance Amplitude Correction (DAC) curve. This is a curve plotted on the screen using transverse or flat bottom holes, of the same size but at different depths, in a block of the same or structurally similar material as the material under examination. The screen is calibrated to the required range and the probe is placed over the hole that gives the best signal response. The response is maximised and set to a pre-determined level using the gain.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT5-5

UNIT UT5  0° PROBE SCANNING NOTES

10

20

30

The peak of the signal is then marked onto the screen (usually on a transparent inlay) and the probe is moved along to a deeper hole. The signal from the deeper hole is then maximised and with the gain setting unaltered the peak of the signal is marked onto the screen and the probe is moved to the next hole down. The procedure is repeated until the end of the range is reached. The marks on the screen are then joined up with a line drawn through and forms the DAC curve. The range, gain setting and probe identification should all be recorded (on the screen usually) along with the curve. The curve shape is a probe characteristic, the gain setting is dependant on the flaw detector, i.e. if the flaw detector is changed for another one the gain setting will be different and if the probe is changed for another one, another curve should be plotted. SCANNING PATTERNS 0° PROBE When scanning for defects the scanning pattern to be used is sometimes dependant on the size of defect sought. The two main factors to consider are: 

40

50

60



The pitch (distance between scans) or The overlap (the amount, if any, that the each scan overlays the next) and the pattern or direction of scanning.

If the pitch is less than the size of the probe the scans will overlap. If the pitch is greater than the size of the probe there will be a gap between the scans. Whether there is a gap between the scans or not may depend on the size of defect sought and the size of the test piece. For example, on a large test piece looking for defects over 100mm the pitch may be 75mm between scans, regardless of the probe size, because scanning every 75mm will locate defects over 100mm in size. The pattern may require scanning in one direction or in two directions at 90° to each other.

70

80

SIZING METHODS 0° PROBE* There are five main sizing techniques used with 0° probes: * ‘Sizing’ refers to the 90 reflective surface of the reflector, which is always smaller than the actual true dimension.

100

1) 2) 3) 4) 5)

6dB drop Equalisation Maximum amplitude Multi back wall echo (flood fill) Distance Gain Sizing (DGS)

Note: Another technique is 20dB drop - not as common with 0°.

© Institution of Mechanical Engineers Issue 17 2016

UT5-6

UNIT UT5  0° PROBE SCANNING NOTES

** Ideal due to: a) No separate beam profile has to be plotted prior to usage

10

b) Small probe movements for the signal drop off mean more 20 consistent coupling conditions can be maintained.

6dB drop technique** Used to size large defects i.e. defects that are bigger than the beam spread, such as laminations. This is where the probe is moved off the edge of the reflector until the signal amplitude is reduced by 50% (6dB). The position of the centre of the probe is marked onto the material surface. The probe is now in a position where the beam is half on and half off the defect. If this is repeated along the edge of the reflector, the reflector's size and shape will be marked out onto the material's surface.

c) Works on large reflectors at any depth within the test piece. 30

40

50

60

70

Equalisation technique The equalisation technique is very similar in operation to the 6dB drop except that the probe is moved off the edge of the reflector until its signal is equal in amplitude to the rising BWE. At this position the centre of the probe is marked onto the surface, again continuing along the edge of the reflector to map out the shape and size.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT5-7

UNIT UT5  0° PROBE SCANNING NOTES

10

20

30

Both the 6dB drop and the equalisation methods only work accurately on large reflectors and will grossly oversize small ones. Note: The flaw must also be along the centre line of the plate for this technique, or again sizing accuracy will be adversely affected. Maximum amplitude (max. amp) technique This is used to size areas of small defects, such as inclusions, or to size multi-faceted defects, such as cracks. The technique involves moving the probe off the defect area until the signals disappear, then slowly bringing the probe back, watching the whole signal group, to the first position where one of the signals maximises. The probe position is marked, as in the other methods, to mark out the edge of the defect area. This technique will pick out the last individual inclusion of a group or the last facet of a crack giving the overall size of the defect or area.

40

50

Multi back wall technique 60

Multi back wall echo technique is very similar to equalisation except it employs a pattern of back wall echoes rather than the first back wall. The main advantage is that it is excellent on thin wall materials as other techniques would fall in the near zone of the sound wave.

70

*This is only able to ‘size’ small reflectors. Once the reflective surface equals or is greater than 80 the beam profile all the sound is reflected and the signal response no longer increases in height, regardless of the reflectors increase in dimension. 90

Distance Gain Sizing (D.G.S.) DGS uses the reflections from flat bottom holes or disc reflectors, of different sizes and over a range of depths, plotted on a graph. Signal amplitudes from defects* are compared to the graph to give the minimum size or, more correctly, the minimum reflective size to the defect. These graphs are provided by the manufacturer of the probe and are illustrated in some reference standards or can be plotted. The DGS diagram must correspond to the probe being used (supplied by manufacturer). Maximise the backwall signal from a defect free area of the work piece and record the dB reading when set at 80% FSH. Maximise the signal from the defect to the same height (80%) and note the difference in dB’s.

100

Using the diagram read vertically from the back wall depth up to the back wall line (∞) read horizontally across to the dB value and record add the dB difference (back wall – defect signal) to this value and read horizontally until it reaches the depth of the defect.

© Institution of Mechanical Engineers Issue 17 2016

UT5-8

UNIT UT5  0° PROBE SCANNING Follow the diagonal line intersecting this line down to the defect size. (This gives a flat bottom hole equivalent size).

NOTES

10

20

30

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT5-9

UNIT UT6  ANGLE PROBE SCANNING ANGLE PROBE SCANNING

N OT E S

10

The initial pulse and the dead zone, mentioned in the previous section, also occur with angle probes and should be regarded in the same way. DIGITAL FLAW DETECTORS

20

The same principles apply to shear waves as 0° probes (see UT 5-1) by setting the required velocity (i.e. 3240 m/s in steel) and adjusting the signal using the zero or probe delay. The signal may be obtained off the No 1 calibration block (100mm radius) or the No 2 calibration block 25 or 50mm radius) CALIBRATION FOR ANALOGUE FLAW DETECTORS

30

40

With angle probes, the reflector must be perpendicular to the beam and there must be some method of capturing repeat signals. Both the V1 (A2) and the V2 (A4) blocks fulfil this criteria. The V1 block has a 100mm radius to reflect the sound and a slot cut at the centre to capture repeat signals. Signals occur every 100mm, therefore can be used to calibrate the screen e.g. to calibrate the screen for a 0 to 200mm range we would place the first echo on 5 on the graticule and the second on 10. To calibrate for a 0 to 100mm range we would place the first echo on 0 and the second on 10 (which gives a 100 to 200mm range), then delay the first echo across the screen to 10.

50

60

70

80

90

The V2 block has a 25mm radius and a 50mm radius both irradiating from the same centre. This has the effect of bouncing the sound from one radius to the other, via the centre, creating repeat echoes. After the first echo, which occurs at a distance representative of the radius that the probe is facing, the echoes occur every 75mm (the sum total of the two radii), this feature can be used to calibrate the screen. The direction the probe faces varies with the range required because it is easier to align more of the echoes on the graticule, when facing a particular radius, for a particular range, than on the other radius.

100

For example, a 0 to 100mm range, the probe would face the 25mm radius where the 25mm signal and the 100mm signal can be easily aligned. For a 0 to 200mm range the probe would face the 50mm

© Institution of Mechanical Engineers Issue 17 2016

UT6-1

UNIT UT6  ANGLE PROBE SCANNING radius where the 50mm and 200mm signals can be easily aligned (the 125mm signal falling somewhere in between).

N OT E S

10

20

30

40

50

Advantages of V2 Block: It is easily employed to check the final calibration by turning the probe to the other radii e.g. calibrated for 0100mm off the R25 turn around to the R50 and you will see 1 signal mid-way across the timebase. ANGLE PROBE TEST SENSITIVITY The following methods can be used for setting test sensitivity.

60

70

80

90

100



Grass level



Reference reflectors



DAC curves

Grass The grass or grain interference method involves coupling the probe to the test surface and increasing the gain until the reflections from the grain structure of the material reach a pre-determined level. The level is often quoted as 2 - 3 mm in height at the maximum test depth but ideally should be referenced to as a percentage of full screen height as not all flaw detectors use the same dimension screen. The sensitivity can be adjusted by increasing or decreasing the level, or by adding or subtracting dBs to or from the gain. Advantages: No additional cost of reference blocks, the attenuation and transfer correction factors are automatically built in, high sensitivity allows constant monitoring of the uniformity of the grain structure of the test piece. Note: care must be taken when setting, it is the grain structure and not the surface profile noise used. Reference reflectors A common method of setting sensitivity is to set a maximised signal from a reference reflector, at target depth, to a predetermined level, for example full screen height. The reference reflector could take the form of a known reflector e.g. a transverse side drilled hole, a flat bottom hole (drilled at the appropriate angle for the probe), a slot or © Institution of Mechanical Engineers Issue 17 2016

UT6-2

UNIT UT6  ANGLE PROBE SCANNING a vee notch. Or it could be a real, or simulated, defect of known size and type.

N OT E S

10

20

*If the surface profile between the reference block and test piece 30 differs even though the structure is the same, then transfer correction factors must be compensated for. 40

Advantages: Easily reproducible between operators from known reflectors, but attenuation and transfer correction factors may have to be separately addressed. D.A.C. Curves Another common method of setting sensitivity involves plotting a graph or curve on paper or the flaw detector screen using transverse holes. One of these is known as the Distance Amplitude Correction (DAC) curve. This is a curve plotted on the screen using transverse holes, of the same size but at different depths, in a block of the same or structurally similar material as the material under examination*. The screen is calibrated to the required range and the probe is placed over the hole that gives the best signal response. The response is maximised and set to a pre-determined level using the gain. The peak of the signal is marked onto the screen (usually on a transparent inlay) and the probe is moved along to a deeper hole. The signal from the deeper hole is then maximised and with the gain setting unaltered the peak of the signal is marked onto the screen and the probe is moved to the next hole down. The procedure is repeated until the end of the range is reached. The marks on the screen are then joined up with a line drawn through and forms the DAC curve.

50

60

70

80

*i.e. as used for DGS blocks/ curves. 90

The range, gain setting and probe identification should all be recorded (on the screen usually) along with the curve. The curve shape is a probe characteristic, the gain setting is dependant on the flaw detector, i.e. if the flaw detector is changed for another one the gain setting will be different and if the probe is changed for another one, another curve should be plotted. Sensitivity methods involving flat bottom holes* are rarely used with angle probes (particularly in the UK) due to the fact that the holes have to be drilled at an angle to suit the probe in use, i.e. the flat reflector at the bottom of the hole has to be perpendicular to the beam. This is difficult to achieve because probe angles can vary by one or two degrees. It also means that you would need a separate block for each probe in use.

100

© Institution of Mechanical Engineers Issue 17 2016

UT6-3

UNIT UT6  ANGLE PROBE SCANNING SCANNING PATTERNS

N OT E S

10

For angle probes the scanning patterns describe the way the probe is manipulated as well as the way it is moved. The most common patterns referred to in some standards and application procedures are:

20

Orbital scan Where the probe is manipulated through an arc movement whilst maintaining the beam focused on a fixed reflector. Used often to identify porosity where the signal can be maintained on an orbital scan.

30

40

Swivel scan Where the probe around it. Used and to ensure transverse scan present.

is rotated on the spot, effectively scanning the beam to identify multi-faceted, planar or multiple defects complete coverage when performing a limited on a weld where the weld reinforcement is still

Lateral scan The probe is moved sideways along a fixed line. Used in the critical root scan of a single vee weld or for sizing the length of a defect longitudinally. Depth scan This is where the probe is moved back and forth in the direction of the beam. As in locating the position of a defect when plotting or when maximising the signal off a transverse hole to set sensitivity.

50

Other "scans" referred to such as "root scan", "transverse scan" etc. are scans for a particular type of defect or in a particular area (root scan - in the root area; transverse scan - for transverse defects).

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT6-4

UNIT UT6  ANGLE PROBE SCANNING N OT E S

10

Basic Trig. Equations:

S= O

SKIP FACTORS In angle probe scanning plotting systems are used for projecting defect depths and positions in relation to the probe index by applying the beam path, read from the screen, and the stand off or surface distance from a reference datum on the test surface. The system works on a series of right angled triangles, so the depths and positions can also be calculated, with trigonometry, using the probe angle and the beam path reading on the CRT timebase.

20

H C= A H

30

T= O A 40

To calculate the expected beam path to a reflector, when the depth and the probe angle are known, we transpose the Cosine formula:

bp 

50

d Cos 

To calculate the depth of a reflector, when the beam path and probe angle are known, we transpose the Cosine formula again:

d  bp x Cos 

60

To calculate the surface distance, when the beam path and probe angle are known, we transpose the Sine formula:

sd  bp x Sin  70

To calculate the beam angle when the depth and surface distance to a reflector are known we use the Tangent formula:

Tan   80

sd d

Finally if we wish to calculate the surface distance (stand off) where the depth of the signal and probe angle is known the above formula becomes:

sd  Tan  x d

90 *Be careful to input the actual beam angle, not the nominal 45/60/70° as stated on the probe case. 100

Digital flaw detectors often have trigonometry functions built in to calculate the values required*.

© Institution of Mechanical Engineers Issue 17 2016

UT6-5

UNIT UT6  ANGLE PROBE SCANNING THE RATIO OF THE SIDES OF THE TRIANGLES IN THE FOUR MOST COMMON PROBE ANGLES

N OT E S

10

38° 1.29 1 0.788

20

1 30

THE IRRADIATION FACTOR 40

When testing tubular materials around the circumference with angle probes it is possible that due to the curvature, wall thickness and probe angle, that the beam will not strike the inside surface of the material. We can calculate the minimum probe angle that will strike the inside surface (at a tangent). This is known as the irradiation factor.

50

60

70

Sin  

opposite IR  hypotenuse OR

By trigonometry:

 = probe angle IR = inside radius OR = outside radius 80

90

Since tubular materials are usually measured by diameter, we can convert the equation to:

Sin  

ID OD

100

© Institution of Mechanical Engineers Issue 17 2016

UT6-6

UNIT UT6  ANGLE PROBE SCANNING Wall Thickness

N OT E S

10

The ultrasonic inspection of tubular products and longitudinal welds by shear waves will involve scanning circumferentially of the item under inspection. The calculation of beam path length and skip distance can be complicated as it is affected by geometrical factors related to the outer diameter to inner diameter ratios.

20

30

40

50

60

Where no reference tube section is available an ad-hoc method is to increase the dB settings until grass level appears on the time base at the half skip and full skip position. The outside diameter full skip position can be confirmed by damping the position with couplant and your finger. This method will not work if laminations etc. are present in the material under inspection.

70

80

90

The actual wall thickness in relation to the outer diameter (ø) is important since it will affect the choice of shear wave probe that will impinge on the bore surface.

100

© Institution of Mechanical Engineers Issue 17 2016

UT6-7

UNIT UT6  ANGLE PROBE SCANNING N OT E S

10

20

30

In the example given above 45° shear wave angle probe only extends to about 50% of the through wall thickness. 40

50

The following formula can be used to select the correct shear wave angle probe to use for the given wall thickness and outer tube diameter that will enable the centre beam to reach the bore of the tube:

t=

Where



t = Maximum possible wall thickness d = pipe outside diameter = Probe angle

This formula can be transposed so that you can calculate the angle given for a wall thickness, for example: 60

Sin

= 1-

See examples in Appendix F (Example Calculations) 70

80

90

In this example you would look to select the closest ordinary angle for a probe, which would be 60° for this pipe schedule. For angles ≤35° and below no shear wave angle probes are available because of the confusion that can arise from the false compression wave signal. There is a maximum wall thickness that can be tested by shear wave angle probes for any given outside diameter that can be tested by the half skip technique. The maximum wall thickness for shear wave angle probes is where the ratio of Outer ø to Inner ø exceeds 4.5 to 1. There is no such limitation for 0° compression probes. For convenience, the formula t =



has been modified

to t = dxf

100

© Institution of Mechanical Engineers Issue 17 2016

UT6-8

UNIT UT6  ANGLE PROBE SCANNING Where f is a factor for standard angle probes that has been calculated from:

N OT E S

These factors are shown in the table below.

10

20

Probe angle (

45

60

70

Probe factor (f)

0.146

0.067

0.030

Digital ultrasonic sets: 30

40

50

Digital ultrasonic sets have a function already in their software menus that will perform these calculations for a prescribed probe angle this will locate on the ultrasonic screen time base the point at which the beam path will hit the bore and reflect onto the outer surface. By inputting the outside diameter or pipe radius, wall thickness, activating the skip function, thickness function and the surface convex / concave function a geometry corrected position for the maximum ½ and full skip positions, depth and surface stand-off can be activated. To deduce the maximum wall thickness that can be tangentially touched for a prescribed probe angle and pipe radius simply increase the wall thickness function incrementally in 1mm steps until the maximum half skip and full skip markers on the ultrasonic screen time base disappear. The functions detailed above are usually located in the measurement menu section under the trigonometry sub-menu in most digital sets.

60

Rather than calculate the position of a reflector in relation to the probe index, using trigonometry, we can draw the probe angle onto a card (or transparent film) and by overlaying onto a cross-sectional diagram of the test piece we can plot the reflector's position.

Alternative names for plotting systems include: i. ultrasonic calculators ii. slide rules iii. plotting cards

PLOTTING SYSTEMS

70

They are predominately used in weld testing.

The following illustrations show two examples of plotting systems, one for use on a flat surface and one for a curved surface.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT6-9

UNIT UT6  ANGLE PROBE SCANNING SIZING METHODS AND ANGLE PROBES

N OT E S

There are three main sizing techniques used with angle probes: 10

20

30

1)

6dB drop

2)

20dB drop

3)

Maximum amplitude

6dB Drop technique Used to size defect dimensions which are larger than the beam, such as the length of a lack of sidewall fusion in a weld. The probe is moved off the end of the defect until the signal amplitude is reduced by 50% (6dB). The position of the centre of the probe is then marked onto the material surface. The probe is now in a position where the beam is half on and half off the defect. If this is repeated at the other end of the defect then the distance between the marks represents its length. Advantages: No need for accurate beam profile plotting prior to usage. Disadvantages: Only suitable for large reflectors.

40

20dB Drop technique This technique is used for defects that are less than the width of the beam, such as the cross-sectional size of a lack of sidewall fusion in a weld*. It requires the use of a 20dB beam profile, plotted out for the probe in use, drawn onto a plotting system.

50

The signal from the defect is first maximised and the position of the defect plotted down the main beam on the plotter as in fig.1. The probe is then moved forwards, off the defect, until the signal drops to 10% of its original height.

*(However use on defects larger than the Ultrasonic Beam is also possible).

60

70

As the probe has moved forward the defect is now in the trailing edge of the beam, so we now plot the signal down the trailing edge on the plotter, see fig.2. This should give a point plotted just above the previous plot and this represents the top edge of the defect. If we now move the probe backwards, past the maximum, to a position where the signal is again 10% of the maxim, then plot the signal down the leading edge of the beam, as in fig.3, this should give us the bottom edge of the defect and thus the overall size. Advantages: Works for both large and small reflectors.

80

Disadvantages: Larger probe measurements mean it is more difficult to maintain constant coupling conditions. Additional reference block are required to determine the beam spread.

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT6-10

UNIT UT6  ANGLE PROBE SCANNING Maximum amplitude (max. amp) technique This is used to size areas of small defects, such as inclusions or porosity, or to size multi-faceted defects, such as cracks.

N OT E S

10

20

30

The technique involves moving the probe off the defect area until the signals disappear, then slowly bringing the probe back, watching the whole signal group, to the first position where one of the signals maximises. The defect is then plotted using the main beam on the plotting system. If this is carried out in both directions the crosssectional extremities of the defect can be plotted out. The technique is repeated moving the probe laterally to size the length of the defect by marking the position of the centre of the probe. This technique will pick out the last individual inclusion of a group, or the last facet of a crack giving the overall size of the defect or area. It can also be used to plot the shape of a defect and for condition monitoring where critical sizing is required by plotting each individual signal in the group as it maximises. Advantages: Small probe movements for consistent conditions. Accurate sizing for small or large reflectors.

40

coupling

Disadvantages: Only suitable for multifaceted reflectors.

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT6-11

UNIT UT7  TESTING TECHNIQUES TESTING TECHNIQUES

NOTES

10

20

A, B & C SCANNING SYSTEMS A-scan This is one of the most common systems in use for manual ultrasonic inspection. It displays the reflected energy as signals on a flaw detector screen. The horizontal axis on the flaw detector screen represents elapsed time or distance and the vertical axis represents signal amplitude or sound energy returning to the probe. This system can provide an indication on the size of a defect from signal amplitude, the defect location, from the position of the signal on the timebase, and the signal shape and behaviour, on movement of the probe, can indicate defect type.

30

The disadvantages of this system are that the signals require interpretation, which means that more skill is required for operation. The advantages of this system are its portability and less time involved in setting up. Plus it will give more information for skilled users.

40

50

60

A-scan display modes

70

80

90

Half Wave Rectified (HWR) either positive or negative signals (rarely used).

100

© Institution of Mechanical Engineers Issue 17 2016

UT7-1

UNIT UT7  TESTING TECHNIQUES NOTES

10

20

30

FWR (full wave rectified) negative signals inverted to be shown with positive pulses (most common).

40

50

Note: There are many variations in Auto UT where combinations of A/B/C scans give a 3 dimensional image of the object reflectors i.e. P/S/T. These are not covered here due to specialist use. Unrectified – The display can be presented as Radio Frequency (RF) alternating positive/ negative signals (rarely used) except in EchoEcho.

60

70

80

90 1 In the case of a defect in steel plate, the defect forms an air interface so the through thickness of the defect is not shown on the display. 100

B-scan system The B-scan system provides us with a cross-sectional view of the material under test by scanning the probe across the surface (sometimes at high speed). The image is retained using digital recording, giving a permanent record of the ultrasonic data. The amplitude of the received signal is represented by the brightness of the image and the synchronisation of the movement of the probe and the display can give a true representation of the size1 of the defect.

© Institution of Mechanical Engineers Issue 17 2016

UT7-2

UNIT UT7  TESTING TECHNIQUES NOTES

10

20

30

C-scan system This system gives us a plan view of the scanned area, showing defects as contrasting areas, on a printout or plotting system that is synchronised with the probe's movement as it traverses over the material. The big advantage of the system is an instant permanent record. The disadvantages would be no indication of defect depth or orientation and setting up the system can be time consuming.

40

50

60

PULSE ECHO SYSTEMS

70

A system that sends out pulses of ultrasonic energy then listens out for the returning echoes is called a pulse echo system. The probes used can be in the single or double crystal format. The single crystal probe transmits pulses of energy, typically at a rate of anything between 150 to 1000Hz, in between pulses the circuitry switches to receive mode to listen for any returning echoes.

80

90

The double crystal probe has separate transmit and receive crystals, the transmitter still sends out pulses, (at the same rates as above) but rests in between, whilst the receiver is in "listening" mode permanently. Note: See the "Introduction to the basic concept" (UT1-1) section of the notes for further information and the "Propagation of sound" (UT2) section for details on pulse repetition frequencies. The advantages of the pulse echo system are that defect positions can be located with accuracy and access to only one side of the test material is necessary.

100

The disadvantage is that the sound has to travel through the material twice (there and back) so there is more attenuation.

© Institution of Mechanical Engineers Issue 17 2016

UT7-3

UNIT UT7  TESTING TECHNIQUES THROUGH TRANSMISSION TESTING

NOTES

10

Mainly used in automated systems, this technique uses two probes, one either side of the test material - one transmitting pulses of energy the other receiving the energy. Any technique using separate probes are referred to as ‘Pitch and Catch’. The received energy signal is set to a pre-determined level and the presence of a defect is indicated by a reduction in amplitude or loss of signal.

A gate is a marker on the display showing the area of interest and the required signal amplitude or level to trigger the alarm. 1

20

The parameters required to set up a gate are: 30 i. Start position. ii. End position or gate length. iii. Level. iv. It can be set to work off 40 positive or negative trigger.

In automated systems the signal may be set to reach or exceed a negative gate1. This means that a portion of the screen in the area of the signal will have an alarm sound if the signal does not reach the pre-set amplitude. This may be coupled to an automatic marking system, such as a paint sprayer, that marks the material when the signal falls short. The marked areas then being inspected later, manually and in more detail. The advantages of this technique are based on the fact that the sound only has to travel one way through the material, i.e. Materials with higher attenuative properties can be tested, thicker materials can be tested and higher frequency probes can be used. The disadvantages are; there is no indication of defect depth, there must be access to both sides of the material to place the probes, the probes must be correctly aligned and a change in coupling conditions (causing a loss of signal amplitude) could be mistaken for a defect.

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT7-4

UNIT UT7  TESTING TECHNIQUES THE TANDEM TECHNIQUE

NOTES

*A separate but similar 10 technique is the Delta Technique (Refer to UT3-4)

20

This employs two probes, one transmitting sound and one receiving with both the probes on the same surface of the test material*. The probes are set at a fixed distance from each other so that the pulses from the transmitter, if reflected from a defect, will be directed to the receiver probe and thus create a signal. The distance between the probes is dependant on the probe angle, the material thickness and the depth of expected defects. The technique is used when looking for defects at a pre-determined depth such as in the root of a double sided weld. The advantage of this technique is, that vertical defects, which would normally be extremely difficult to locate ultrasonically by 0° or angle probes, would be easily found.

30

The disadvantage is, that only defects at the pre-determined depth would be located.

40

50

60

IMMERSION TESTING This is an automatic ultrasonic inspection technique that is carried out in laboratories or specialised factory inspection areas. 70

* Surface waves are extremely limited in immersion systems (tanks) since it is a boundary wave and requires a solid to gas interface to propagate without massive attenuation.

80

The system uses a compression probe mounted in a manipulator that is carried on a bridge over a tank of water in which the test material sits. To prevent the formation of air bubbles on the test piece surface, the water is heated to ambient temperature to de-aerate it. The manipulator allows the probe to be tilted at any angle. By varying the angle beyond the critical angles, various shear wave refracted* angles can be produced in the test material as required. The bridge allows the probe to be moved over the test material. The test material is sometimes placed on a rotating table in the tank and is rotated as it is scanned. Probe frequencies up to 25MHz are not uncommon in immersion testing.

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT7-5

UNIT UT7  TESTING TECHNIQUES NOTES

10

20

30

40

50

Calibration is done with the same set-up which is to be used for testing, i.e. same probe and water path. The water path between the probe and the test material front surface is then delayed off the screen so that the zero end of the screen represents the front surface of the test material.

60

70

80 1 In practice, the ‘rule of thumb’ is that the water gap should be at least a ¼ of the material thickness (assuming Fe or Al + 90 ¼” (6 mm))

The velocity of ultrasound in steel is four times the velocity in water. So when testing steel the water gap should be greater than one quarter the thickness of the steel1. Otherwise, the repeat signals from the front surface will start to occur before the bwe and a front surface echo will occur within the test area on the screen, thus masking any defects within the test piece at this depth. TIME OF FLIGHT DIFFRACTION (TOFD) Unlike normal pulse echo ultrasonics which use reflected and refracted signals, TOFD uses diffracted signals which are much weaker and radiate from the tips of flaws in all directions.

100

Using a pitch and catch arrangement, TOFD uses two transducers; one which transmits and one which receives. When using TOFD it is

© Institution of Mechanical Engineers Issue 17 2016

UT7-6

UNIT UT7  TESTING TECHNIQUES NOTES

10

20

generally presented as a B-scan because when in A-scan form it is very difficult to interpret. The main ultrasonic beam used in TOFD is a compression wave and this travels through the weld volume at an angle determined by the wedges on which the transducer sits. As well as the main compression wave, TOFD also has a lateral wave which runs between the transducers near the surface of the weld and is used to determine the velocity of the material and if an indication is surface breaking or not.. The B-scan image presented by TOFD is known as a grey scale. This is due to the amplitude and phase changes of the received diffracted waves. Unlike a normal A-scan, TOFD A-scan uses an unrectified signal, i.e. RF signal which appears at about 50% screen height on a normal A-scan.

30

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT7-7

UNIT UT8  ULTRASONIC THICKNESS SURVEYING ULTRASONIC THICKNESS SURVEYING

NOTES

10

20

Thickness surveys usually take the form of multiple measurements at pre-determined positions on a component, e.g. a boiler wall or a ship's hull, using a dedicated thickness meter (D-meter or T-gauge), or an A-scan flaw detector and a 0° probe. The probe selection is dependant on the material thickness and attenuation properties. Using a thickness meter Dedicated thickness meters are either pre-calibrated at the factory for a particular material, with a supplied probe unit, or may use a calibration block and a calibration routine is carried out prior to use. A typical calibration routine on a digital thickness meter would be: 1.

Switch the unit on, clean the probe shoe and press the "zero" function button to zero the probe.

2.

Select "calibrate" and place the probe on a thin section of the calibration block, press "zero" again and enter the actual thickness into the unit.

3.

Place the probe on a thick section of the calibration block, press "vel" and enter the actual thickness.

4.

The meter will then automatically calibrate and is ready for use on the same material as the calibration block.

5.

The meter may have other features such as digital storage for the readings, adjustments for accuracy of the readings or minimum thickness recording.

30

40

50

60

80

Using a flaw detector The A-scan flaw detector can be used to obtain thickness readings to a good degree of accuracy by calibrating to a small range or using the multiple back wall method as explained in the "0° probe scanning" section of the notes. The advantage of a flaw detector over a thickness meter is that a representation of the signal shape can be seen on the display which indicates whether the reading is off a back wall or off a defect within the material.

90

When using a flaw detector, if the surface is coated or painted then the reading should be taken between the repeat signals and not from 0 to the first signal. When using a thickness meter any coatings or paint on the test surface does not affect the reading-because it automatically reads the repeat distance.

70

Some digital flaw detectors have an echo – echo feature in the measure mode which allows a true metal path reading eliminating the need to remove paint or coatings etc.

Defects within the material can give rise to incorrect thickness readings when using a D-meter that has no A-scan display. When using a D-meter, which reads the beam path between the first and second or the second and third repeat signals, corroded back wall may cause loss of readings due to the attenuation of the sound. If the test material velocity is not the same as the velocity of the calibration block, or the material the D-meter is set up for at the factory, then false readings will occur.

100

© Institution of Mechanical Engineers Issue 17 2016

UT8-1

UNIT UT8  ULTRASONIC THICKNESS SURVEYING If using the flank of the signal, signal amplitudes must be at similar heights when checking the position on the timebase, both when calibrating and when taking readings.

NOTES

10

The timebase on an A-scan flaw detector must be linear to attain accurate readings, a check for this is explained in the equipment section (UT4) of the notes.

20

Flank

30

PEAK AND FLANK 40

There are 2 main methods of recording the thickness of a sample that relate to the received and displayed echoes. Flank (Edge mode) Flank measures the first point at which the received and displayed echo crosses the electronic gate, regardless of the amplitude response height of the echo and /or echoes within the gate.

50

Peak Peak measures the distance to the echo with the highest amplitude response (highest peak) within the gate, regardless of any echo that precedes it with a lower amplitude response.

Note for Flank Mode: Always measure the echo height at the same height you calibrate from. Extra power added will cause the echo peak to be taller and the echo at the baseline to broaden, creeping along the gate (as in picture b). Note for Peak Mode: The power does not affect accurate measurement as the measurement is always taken when the highest peak crosses the electronic gate.

60

(a)

(c)

70

80

(b)

(d)

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT8-2

UNIT UT8  ULTRASONIC THICKNESS SURVEYING Accuracy on thickness readings

NOTES

10

If the flank detection mode is used for accurate thickness measurement, careful attention must be paid to the received echo heights. The test block echo height used for calibration and the echo height received from the sample under test should be of the same echo height response. The apparent arrival time of the leading edge of a signal will vary with the amplitude height of the signal.

20

When using the peak mode for thickness measurement the start and width of the gate should be positioned in such a way that it only encompasses the desired echo under consideration (generally the first echo on the displayed screen). Care should be taken when using a twin crystal transducer as this type of ultrasonic probe will display a dual peak.

30

Whichever mode is chosen for calibration must be used for any following thickness measurement subsequently taken. Note: These methods are not interchangeable and therefore recalibration is needed if a change in method is undertaken.

40

Regardless of method of thickness mode measurement, the received signal response height should not be above 100% FSH.

50

Flank Mode, blue X indicates the measuring point. 60

70

Peak Mode – INCORRECT set up, blue X indicates the measuring point. If measuring in peak mode, the gate needs to be positioned over the echo to be considered (generally the first), ensuring the gate is not too wide.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT8-3

UNIT UT8  ULTRASONIC THICKNESS SURVEYING NOTES

Peak Mode – CORRECT placement of the gate, blue X indicates the measuring point.

10

20

30

Peak Mode – Noting peak bifurcated signal change from 1st BWE to repeat BWE signals (the highest peak can appear on the left or right)

40

50

60

Time Corrected Gain (TCG)/Automatic Gain Control (AGC)

70

80

As mentioned previously the flank or edge mode requires that the shape and height of the echo signal received be consistent for accurate and reproducible thickness measurement. The TCG/AGC function on a modern ultrasonic digital flaw detector can improve the reliability of thickness surveying by increasing the echo amplitude to a pre-set % FSH that corresponds to the height used in calibration. This will maintain the correct detection threshold on the leading edge and will equalize echo height from various section thickness and profiles. ACCEPT/REJECT CRITERIA

90

100

When thickness surveying you may be asked to evaluate the measurements taken instead of, or as well as, recording them. This may be by using the accept/reject criteria from a national standard or a written procedure for the job in hand. Acceptance tolerances may be given in the form of maximum and minimum thicknesses or given as a percentage tolerance of a nominal thickness, e.g. Minimum 13.5mm, maximum 16.5mm or 15mm ± 1.5mm or 15mm ± 10%. The first two examples are quite easy to

© Institution of Mechanical Engineers Issue 17 2016

UT8-4

UNIT UT8  ULTRASONIC THICKNESS SURVEYING follow but the percentage tolerances are not always simple figures like 10%.

NOTES

10

To calculate the value of the tolerance from the stated percentage and hence the maximum and minimum thicknesses we use the formula:

Tolerance 

t x n 100

Where t is the plate thickness and n is the percentage. 20

So:

Maximum thickness  t 

t x n 100

30

Minimum thickness  t 

40

50

t x n 100

REPORTING When reporting the results of a thickness survey the readings may be electronically stored, in memory on some thickness meters or digital flaw detectors, or written down. In each case the location of the reading must be stored along with the thickness for use as a reference in further checks or for mapping out the test surface. The electronically stored readings may be downloaded into a database application or directly into a graphics program that will give a visual representation of the test area.

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT8-5

UNIT UT9  ULTRASONIC WROUGHT PLATE TESTING ULTRASONIC WROUGHT PLATE MATERIAL

NOTES

10

20

30

40

TECHNIQUE When searching for defects in wrought plate you should have, as a minimum, the following information, which is usually written on a technique or instruction sheet (see the appendices for an example). 

The test component identification and area to test.



Actions to be taken when defects are found.



The purpose of the test (defects sought and acceptance criteria).



Equipment required.



What method and level of test sensitivity to use.



The method of scanning.

The instruction sheet would also contain sections giving details of any relevant safety procedures and post test procedures such as the cleaning of the test area afterwards. It would also have the company name, a unique technical reference number, the originator's name and signature and an authorising signature. Test area The test may involve testing the whole of a component, or just parts, but this must be specified.

60

Actions to be taken When defects are found it may be required that the defects are reported, e.g. on a diagram or as a written description, or the wrought plate, may be accepted or rejected based on the defects found. If defects are to be reported then the defect information that needs reporting would be contained in this section, i.e. Defect type, size, lateral and longitudinal position in relation to datums, etc.

70

Purpose of the test This section tells us the accept/reject criteria for particular defects, i.e. what size and type of defects to report, which defects render the component rejectable, or which defects to assess for grading of the material.

50

Equipment The type of flaw detector, type, size and frequency of probes, type of couplant and calibration blocks to use, should be stated. 80

90

Sensitivity Method of setting and level of sensitivity need to be quoted for each scan, e.g. 2nd BWE. F.S.H. off clean parent material. Scanning method The method of scanning the material is either a written step by step instruction or technique sheet, or involves following the steps laid out in the relevant national standard.

100

© Institution of Mechanical Engineers Issue 17 2016

UT9-1

UNIT UT9  ULTRASONIC WROUGHT PLATE TESTING An example written step by step could be:

NOTES

10

1. Prepare the material surface by removing any loose scale, rust, dirt or other debris and visually inspect for surface defects or damage. 2. Calibrate the screen on the flaw detector, using a 0° probe and the A2 calibration block, for a range of 0 to 50 mm. 3. Set the sensitivity (as quoted in the relevant section above) and apply couplant to the test area.

20

30

*Ref to UT5-7 for usage/ selection criteria of sizing methods. 40

50

60

4. Scan the designated test area with a probe overlap, between scans, of at least 20% of the probe's diameter at a maximum probe movement rate of 150mm/sec. 5. When defects meeting the criteria in the "Purpose of the test" section are found, record the relevant defect data as in the "Actions to be taken" section. 6. Defects larger than the ultrasonic beam, i.e. where there is no B.W.E. present, should be sized using the 6 dB drop or equalisation methods. Defects that are smaller than the ultrasonic beam should be sized and positioned using the maximum amplitude technique*. Where there are found to be a number of small defects together they should be grouped and sized as an area, using the maximum amplitude technique on the defects that are at the edge of the area. 7. Prepare a neat concise report giving details of the component identification. Test area, equipment used, sensitivity settings and a drawing with the defect details as recorded in section 5 above. Sign and date the report and state your relevant qualifications. Post test procedures This would involve cleaning any remaining couplant and dirt from the test area and covering the surface with protective coatings according to client's requirements. DEFECTS IN PLATE MATERIAL

70

80

The interpretation of defects in plate material involves knowledge or experience of the expected types of defect and the possible signals from them. In some situations it is a case of reading the signal, evaluating which defects do not give this type of signal, then choosing from the remaining possibilities as to which type of defect is most likely. Here are a few of the more common defects found in plate: Laminations A lamination is a defect that is larger than the ultrasonic beam and lies parallel to the plate surface, normally midway through the plate depth. It is formed from the rolling out of secondary pipe in cast ingots. The air and the slag, which were originally on the ingot surface, are trapped within the defect forming an acoustic barrier (interface). This means that sound is totally reflected off the defect, so there is no B.W.E.

90

The defect echoes all behave in a similar fashion e.g. a change in coupling conditions causes the whole group of repeat echoes to fluctuate.

100

© Institution of Mechanical Engineers Issue 17 2016

UT9-2

UNIT UT9  ULTRASONIC WROUGHT PLATE TESTING NOTES

10

20

30

40

50

Inclusions Inclusions, in plate material, are formed from lumps of trapped solid non-metallic material in the cast ingot. These lumps are crushed, flattened and broken up during the rolling process and end up as smaller flatter shapes. Small inclusions are easily differentiated from laminations because B.W.E. signals are still present on the screen among the defect signals and they may be found at any depth. The two most common types of inclusions are linear and scattered inclusions and can be differentiated by the signal pattern on the screen. Linear inclusions This defect is formed from a single inclusion or a closely grouped cluster of inclusions in the cast ingot.

60

70

This results in the rolled out defects ending up at similar depths within the plate. The signal pattern consists of a set of defect repeat signals and a set of back wall echoes. The centre of the ultrasonic beam has the most intense energy and as the probe is moved across the material surface the beam centre is sometimes on the back wall (as it passes between the small defects) and sometimes on the individual defects. This has the effect of high B.W.E.'s and small defect signals or high defect signals and small B.W.E.'s alternating as the probe is moved.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT9-3

UNIT UT9  ULTRASONIC WROUGHT PLATE TESTING Scattered inclusions

NOTES

10

20

30

These are formed from various sized inclusions throughout the cast ingot and when rolled out the shapes, sizes, orientations and depths of the defects in the plate vary. The varying orientation and shape has the effect of scattering the sound beam as it passes through the plate and if the sound reaches the back wall and reflects back, it scatters again on the return journey. This causes significant attenuation in the amplitude of the B.W.E. compared to a defect free area. The amplitudes of the signals from the defects also vary because of the differences in sizes and orientations. The signals we see then on the screen are a low B.W.E. and a cluster of signals varying in amplitudes and depth from the defects. The cluster of signals from the defects has a constantly changing pattern when you move the probe across the surface.

40

50

60

Stringers These are formed from non-metallic inclusions in the cast ingot. 70

The inclusions are rolled out into long thin string-like shapes (as the name implies). The signal response from a stringer is very much like a linear inclusion signal when scanning across the rolling direction of the late. In the rolling direction the B.W.E. is still present, but the signal can be maintained along the defect's length.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT9-4

UNIT UT9  ULTRASONIC WROUGHT PLATE TESTING A Rolling lap This defect occurs in the rolling process when too great a reduction in section is attempted in one rolling pass.

NOTES

10

The material folds over onto itself and is flattened into the surface by the rolls leaving a visible seam on one side of the plate. The signal response from the opposite side of the plate is the same as with a lamination on one edge of the defect (Probe Position A).

20

30

At the other edge, (Probe Position B) the signal from the defect drops very low, or disappears, before the B.W.E. appears i.e. before the probe reaches the edge of the defect. This is because the defect surface slopes down toward the bottom surface of the plate, this causes the sound to deflect away from the probe. As the probe is moved off the edge of the defect (Probe Position C), the B.W.E. comes up. The sloped end of the defect therefore has to be sized by performing a 6 dB drop on the B.W.E. (If the sloped area has a degree of irregularity, then the maximum amplitude technique would be an alternative).

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT9-5

UNIT UT9  ULTRASONIC WROUGHT PLATE TESTING NOTES

10

ACCEPT/REJECT CRITERIA When defects are found it may be required that the defects are reported. Or the material may be accepted, rejected or graded according to the defects found. The accept/reject criteria tell us what size and type of defects to report, which defects render the component rejectable, or which defects to assess for grading of the material. The criteria can be found in a procedure, a written instruction sheet or in a national standard.

20

30

REPORTING A report should give details of the component identification, test area, surface condition, equipment used, sensitivity settings and a drawing showing the defects and details such as; defect type, size, lateral and longitudinal positions in relation to datums, etc. The report should be signed and dated and there may be a requirement to state your relevant qualifications.

40

Alternatively, if grading material, instead of a drawing you may be asked to give a written statement of conformity to the relevant grade, or acceptance level, of the national standard employed.

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT9-6

UNIT UT10  ULTRASONIC WELD TESTING ULTRASONIC WELD TESTING

NOTES

10

20

30

40

50

60

70

80

90

TECHNIQUE When searching for defects in welds you should have, as a minimum, the following information which is usually written on a technique or instruction sheet (see the appendices for an example). 

The test component identification and area to test.



Actions to be taken when defects are found.



The purpose of the test (defects sought and acceptance criteria).



Equipment required.



What method and level of test sensitivity to use (preparations).



The method of scanning.

The instruction sheet would also contain sections giving details of any relevant safety procedures and post test procedures such as the cleaning of the test area afterwards. It would also have the company name, a unique technical reference number, the originator's name and signature and an authorising signature. Test area The test may involve testing the whole of a component, or just parts, but this must be specified. Actions to be taken When defects are found it may be required that the defects are reported, e.g. on a diagram or as a written description, or the weld, may be accepted or rejected based on the defects found. If defects are to be reported then the defect information that needs reporting would be contained in this section, i.e. Defect type, size, lateral and longitudinal position in relation to datums, etc. Purpose of the test This section tells us the accept/reject criteria for particular defects, i.e. what size and type of defects to report, or which defects render the weld, or parent metal, rejectable. Note: Defects in the parent metal, adjacent to the weld, could limit the weld scans with the angle probes. Equipment The type of flaw detector, types, sizes, angles and frequencies of probes, type of couplant and calibration or reference blocks to be used, should be stated. Sensitivity The method of setting and level of sensitivity needs to be quoted for each scan, e.g. using an 80% F.S.H. DAC curve, plotted from 3 mm diameter side drilled holes, add 14 dB to the gain. This information may be contained in a section on preparation for the test, along with things like; lighting conditions, surface cleanliness etc.

100

© Institution of Mechanical Engineers Issue 17 2016

UT10-1

UNIT UT10  ULTRASONIC WELD TESTING NOTES * Generally the initial scans are to locate the worst case defects i.e. surface breaking (most likely to propagate due to being in the most highly stressed area). The next sequence is to view for the next most serious defect i.e. internal planar reflections (such as lack of fusion). The last scans are designed to find volumetric/rounded defects which are less likely to propagate as rapidly.

10

An example written step by step instruction for a single vee butt weld could be: 1. Visually inspect the parent metal and weld surfaces, reporting the surface condition and the presence of any weld cap defects. 20

30

2. Scan the parent metal with a 0° probe, check and report the thickness and any defects. Where the cap is dressed flat, scan the weld metal with the 0° probe for defects and record on a rough report. 3. Draw up full size working diagrams and cursors (plotting systems), noting surface distances and beam paths for each angle probe on half skip and full skip positions. 4. Mark the centreline of the weld and the surface distance for each probe onto the scanning surface.

40

50

60

70

*Take care plotting to ensure these indications are not mode conversions from discontinuities already located on the full skip examination scan.

Scanning method The method of scanning the material is either a written step by step instruction or technique sheet, or involves following the steps laid out in the relevant national standard*.

5. Using a guide strip behind the probe, perform a critical root scan by scanning laterally on a fixed line parallel with the weld axis, with the probe index point at the half skip surface distance, with each probe (access permitting). Make a note on a rough diagram of any suspected defective areas of the root, as they are located with each probe. Assess each suspect area individually to ascertain whether the area is a defect, whether the defect is in the root and if so, what type of defect and its size and position. Record the defects on the rough report. 6. Scan the weld body on full skip, with each angle probe in turn (access permitting), by moving the probe back and forth between the half and full skip surface distances whilst gradually traversing the length of the weld. Assess each signal that falls within the half skip to full skip beam path range as it is located. Record the defects on the rough report. 7. Scan the weld body on half skip, with each angle probe in turn (access permitting), by moving the probe back and forth between the half skip surface distance and the weld cap, or past the weld centre line if the cap has been removed, whilst gradually traversing the length of the weld. Assess each signal that falls within the zero to half skip beam path range (except the dead zone), as it is located*. Record the defects on the rough report.

80

8. Scan the weld for transverse defects by scanning down the axis of the weld, where the cap is removed, using sufficient scans and different angle probes to ensure full coverage of the weld body, on half and full skip where necessary. Assess and record the defects on the rough report.

90

9. Transfer the noted defects from the rough report to a pro-forma report sheet and make a note of signal amplitudes in comparison to the DAC curve at the test sensitivity level.

100

All noted indications should be assessed, using the plotting system and changing probes as necessary, as to whether they are in fact defects, not spurious indications. If they are defects, the type, size and position in relation to the datum and the centre line of the weld

© Institution of Mechanical Engineers Issue 17 2016

UT10-2

UNIT UT10  ULTRASONIC WELD TESTING should be assessed. (The sizing of defects to be carried out as in the "0° scanning" or "angle probe scanning" section of these notes, as appropriate, or as in a relevant national standard)

NOTES

10

Definition: Defect is normally used to indicate a discontinuity (unintentional addition) which is outside the acceptance criteria. Flaw would be more commonly referred to as an acceptable discontinuity.

Note: The finished report should be signed and dated by a level two operator. DEFECT SIGNAL INTERPRETATION

20

Once it has been established that a signal is an indication of a defect, the next stage is to try and establish what type of defect it is. This is done by interpretation of the signal shape, size and response to movement of the probe, the position of the defect in the weld and knowledge of the types of defects expected.

30

40

50

Signal shape and size High amplitude sharp signals are indicative of specular (mirror-like) reflectors, such as large flat defects, that are perpendicular to the ultrasonic beam. When using angle probes in weld testing, a high amplitude sharp signal that drops in amplitude significantly as the probe is swivelled slightly (so that the beam is not perpendicular to the defect) would indicate a lack of fusion. If the defect plots at a position where the bevel on the parent metal was, prior to welding, this may indicate a lack of side wall fusion. If the signal plotted in the bottom corner of the root face then it may indicate a lack of root fusion. If it plotted in the bottom corner of the root face on both sides of the weld it may be incomplete penetration of the root run. This illustrates that defect interpretation is not only dependant on signal characteristics, but also on its plotted position.

60

Specular Reflection

70

Digital flaw detectors have pixelated displays which may mask certain defect characteristics.

80

*As a general rule 90 this does NOT relate to cracks as they are multiple planar reflectors in all directions, hence some reflectors are always 100 perpendicular to the beam.

Low amplitude signals are indicative of poor or highly attenuative reflectors, or defects or reflectors that are not perpendicular to the ultrasonic beam. Examples of poor reflectors are interfaces where the ratio of the acoustic impedances of the interface materials is low, such as cladding materials applied to improve surface qualities of some components (load bearing or anti-corrosion materials). Highly attenuative reflectors are ones with rough surfaces, such as small multiple defects such as porosity or inclusions*, which scatter the sound in different directions (away from the probe). Specular reflectors that are not perpendicular to the beam (even by only a few degrees) redirect the sound away from the probe so less, or none of it, returns.

© Institution of Mechanical Engineers Issue 17 2016

UT10-3

UNIT UT10  ULTRASONIC WELD TESTING NOTES

10

20

30

Multiple signals are often obtained from multi-faceted or multiple defects such as cracks, porosity or slag inclusions. Cracks usually give a higher signal response than porosity or slag (size for size, at the same sensitivity and beam path). The signals from a crack, or porosity, will rise and fall as the probe is swivelled. The signal pattern from a crack will decrease in amplitude if the probe is orbited around it. Whereas the signal pattern from porosity, or a slag inclusion, can be maintained when the probe is orbited because the porosity or slag inclusion is volumetric. Notes: See "Angle Probe Scanning" (UT6) section of the notes for details on swivel and orbital scans.

40

50

Signal responses from weld defects As previously mentioned, the position where the signal plots plays a significant role in determining defect type and here are a few examples: Root defects Lack of penetration

60

High amplitude corner signals both sides of the weld, rapidly decreasing in amplitude on rotational scan. Plotting at plate thickness depth, the width of the root gap apart, with no cross-over.

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT10-4

UNIT UT10  ULTRASONIC WELD TESTING Lack of root fusion (LORF)

NOTES

10

High amplitude signal plotting on the defect side of the weld, rapidly decreasing in amplitude on the swivel scan and plotting at plate bottom beam path. (There may also be a signal from the root bead as well, particularly if using a steep angle probe, e.g. 45° - see sketch).

20

30

40

50

60

70

From the opposite side, a signal from the root bead should be observed which could vary in signal amplitude on probe movement. The beam path plotting slightly longer than the pre-determined BP from plate bottom. The tip of the LORF is unlikely to be monitored at all from this side because of its vertical orientation. Root crack

80

Subject to orientation and crack irregularity it would be normal to expect a high amplitude, multi-faceted reflector probably from both sides of the weld. If the vertical height of the crack was substantial, a characteristic running signal on the time base would be noted on a depth scan with the angle probe. The response would rise and fall on the swivel and lateral probe movements due to crack irregularity.

90

Toe crack plotting at the toe of the weld root and centre line cracking plotting at the root centre.

100

© Institution of Mechanical Engineers Issue 17 2016

UT10-5

UNIT UT10  ULTRASONIC WELD TESTING NOTES

10

20

30

40

Root undercut 50

60

The severity of the undercut will determine the type of amplitude received, e.g. it could be a relatively low amplitude response or on the other hand, it can give high amplitude responses. However, associated with the undercut echo will be a signal from the root bead as well (see sketch). If the undercut is on one side of the weld only (as shown in the sketch) when the root area is being examined from the opposite side it is likely that a normal root bead response will be observed only.

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT10-6

UNIT UT10  ULTRASONIC WELD TESTING Excess root penetration (over penetration)

NOTES

Definition: 10 Excess penetration, where the weld has a greater depth than the original plate thickness. IF it protrudes beyond a specified limit and 20 becomes a defect, it is normally referred to as excessive penetration.

Root bead type signals, both sides of the weld, plotting beyond expected beam path length to the bead and crossing over. Steeper angled probes (e.g. 38° or 45°), access permitting, giving good results. Note: If weld cap is flush, a 0° probe would give best results and possibly the actual amount of surplus weld material. Whereas the angle probes cannot determine the amount of penetration since they do not allow the sound to reflect from bottom of the root bead.

30

40

50

60

Root concavity/Root Suck Back 70

Low amplitude signals, both sides of the weld, plotting short of plate thickness, no cross-over. If only slight concavity it is likely that it will not be observed ultrasonically.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT10-7

UNIT UT10  ULTRASONIC WELD TESTING Defects in the weld region (sidewall/body)

NOTES

Lack of sidewall fusion 10

High amplitude signal from "A" on full skip and "C" on half skip (access permitting), plotting on the bevel (as shown). Low amplitude signals or no response (dependant on slag entrapment) from "B" and "D" (lower amplitude signals from "A" and "C" when the probe angle is such that the beam is not perpendicular to the defect).

20

30

The signal will normally be clean with a high amplitude response (as previously described) and on swivel and orbital scanning the echo will fall quickly. When sizing the defect's length using a lateral scan, the amplitude response should remain constant.

40

50

60

Lack of inter-run fusion (between weld runs) would give similar signal responses to the above, but plotting anywhere in the body of the weld, the angle probe with a beam perpendicular to the major plane of the defect giving the best response. Slag inclusion

70

80

Detectable from all accessible positions and directions, due to volumetric nature. Signal contains numerous half-cycles and has a rounded peak. Signal appears to roll on movement of probe (the front edge of the signal appears to fall as the back edge rises and vice versa). Should be able to be detected, within reason, with any angle probe.

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT10-8

UNIT UT10  ULTRASONIC WELD TESTING Cluster porosity or multiple small inclusions

NOTES

10

Detectable from all accessible positions and directions, due to volumetric nature. Very low amplitude response due to signal attenuation giving multiple signals with a wide time base. Signal can be maintained on an orbital scan.

20

30

40

Cracks 50

60

Cracks can appear at the toes in the heat affected zone or in the centre of a weld as well as in the root area. The signal response from a crack in these locations is much the same as in the root. (See previous explanation root crack). The orientation of the crack has an effect on the amplitude and width of the signal. If the major plane of the crack is perpendicular to the beam then a high amplitude, narrow, group of signals is seen. If the major plane is at an oblique angle to the beam then a lower amplitude, broad based, group of signals is seen (very similar in shape to the signal from cluster porosity). The signals will rise and fall on a swivel scan and the signals will diminish on an orbital scan.

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT10-9

UNIT UT10  ULTRASONIC WELD TESTING Sizing The sizing methods are explained in the "0° probe scanning" and "angle probe scanning" sections of the notes.

NOTES

10

20

30

40

50

60

There are various sizing methods available and normally it is left up to the ultrasonic technician as to which one he/she prefers. However, certain projects/contracts may refer to detailed ultrasonic procedures which dictate the sizing method to be applied. In general, whether applying maximum amplitude, 6 dB drop or 20 dB drop sizing techniques, providing they are all used correctly, they will all give similar results. But remember these techniques can only size the reflective surface to the beam, not the whole defect.

ACCEPT/REJECT CRITERIA When defects are found it may be required that the defects are reported, or the weld may be accepted or rejected according to the defects found. The accept/reject criteria tell us what size and type of defects to report or which defects render the weld rejectable. The criteria can be found in a procedure, a written instruction sheet or in a national standard.

REPORTING A report should give details of the component identification, test area, surface condition, equipment used, sensitivity settings and a drawing showing the defects and details such as; defect type, size, lateral and longitudinal positions in relation to datums, etc. making a note of signal amplitudes in comparison to the DAC curve at the test sensitivity level. The report should be signed and dated and there may be a requirement to state your current relevant qualifications. Alternatively, if accepting or rejecting the weld, instead of a drawing you may be asked to give a written statement of conformity to the relevant acceptance level, of the procedure used or national standard employed.

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT10-10

UNIT UT11  ULTRASONIC TESTING OF FORGINGS ULTRASONIC TESTING OF FORGINGS

NOTES

10

20

GENERAL With the ultrasonic testing of forgings of simple geometry, such as bar and billet, there are few limitations. When testing general forgings, such as crankshafts, etc. then the most limiting factor is the shape. On complex shapes the surface curvatures may not allow good contact or coupling, the angles of the surfaces may prevent back wall echoes with 0° probes and some forgings, simple or complex, may be anisotropic in grain structure (different grain sizes in different directions). When to test

30

If possible, forged products should be tested after final heat treatment but before final machining of keyways recess etc which could limit the test. Any further heat treatment may consequently produce defects due to thermal shock and/ or changes to the grain structure. TECHNIQUE

40

50

60

70

80

90

100

When searching for defects in forgings you should have, as a minimum, the following information which is usually written on a technique or instruction sheet (see the appendices for an example). 

The test component identification and area to test.



Actions to be taken when defects are found.



The purpose of the test (defects sought and acceptance criteria).



Equipment required.



What method and level of test sensitivity to use.



The method of scanning.

The instruction sheet would also contain sections giving details of any relevant safety procedures and post test procedures such as the cleaning of the test area afterwards. It would also have the company name, a unique technical reference number, the originator's name and signature and an authorising signature. Test area The test may involve testing the whole of a component, or just parts, but this must be specified. Actions to be taken When defects are found it may be required that the defects are reported, e.g. on a diagram or as a written description, or the component, or material, may be accepted or rejected according to the defects found. If defects are to be reported then the defect information that needs reporting would be contained in this section, i.e. defect type, size, lateral and longitudinal position in relation to datums, etc. Purpose of the test This section tells us the accept/reject criteria for particular defects, i.e. what size and type of defects to report or which defects render the component rejectable.

© Institution of Mechanical Engineers Issue 17 2016

UT11-1

UNIT UT11  ULTRASONIC TESTING OF FORGINGS NOTES

10

20

30

40

Equipment This section should give information on; the type of flaw detector, type, size and frequency of probes, type of couplant, calibration blocks and reference blocks to use. Sensitivity Method of setting and level of sensitivity need to be quoted for each scan, e.g. set the BWE from the DGS block to 80% FSH and note the gain setting. Still on the DGS block, maximise the signal from the flat bottom hole at target depth (test material thickness) and set that to 80% FSH noting the difference in dBs between the new gain setting and the previous one. Set the BWE from the test material to 80% FSH, add the difference noted in the first two gain settings to the present gain and scan at this level. Scanning method The method of scanning the material is either a written, step by step, instruction or technique sheet, or involves following the steps laid out in the relevant national standard. An example written step by step could be: 1. Prepare the material surface by removing any loose scale, rust, dirt or other debris and visually inspect for surface defects or damage. 2. Calibrate the screen on the flaw detector, using a 0° probe and the A2 calibration block, for a range of 0 to 200 mm.

50

3. Set the sensitivity (as quoted in the relevant section above) and apply couplant to the test area. 4. Scan the designated test area, with a probe overlap between scans of at least 20% of the probe's diameter and at a maximum probe movement rate of 150mm/sec.

60

*Refer to UT5-7 for usage/selection criteria of sizing methods.

70

80

90

100

5. When defects meeting the criteria in the "Purpose of the test" section are found, record the relevant defect data as in the "Actions to be taken" section. 6. Defects larger than the ultrasonic beam, i.e. where there is no BWE present, should be sized using the 6 dB drop or equalisation methods*. Defects that are smaller than the ultrasonic beam should be sized and positioned using the DGS diagram for the probe in use. With a calibrated screen, maximise the signal from the defect and set the amplitude to 20% FSH and record the gain setting. Move the probe to an area of the material where the back wall is the same distance as the previously recorded defect, there are no defects and the surface condition and curvatures are the same as the located defect area. Set the BWE to 20% fsh and note the difference between the previous and the new gain settings. Using the DGS diagram look on the infinity line at the BWE distance for the dB figure, add this figure to the previously noted dB difference. Read the total dB figure (the two just added), at the reflector beam path (depth), off the graph to give the equivalent size of the reflector. 7. Prepare a neat concise report giving details of the component identification, test area, equipment used, sensitivity method and settings and a drawing with the defect details as recorded in section 5 above. Sign and date the report and state your relevant qualifications.

© Institution of Mechanical Engineers Issue 17 2016

UT11-2

UNIT UT11  ULTRASONIC TESTING OF FORGINGS NOTES

10

Post test procedures This would involve cleaning any remaining couplant and dirt from the test area and covering the surface with protective coatings according to client's requirements. DEFECTS IN FORGINGS

20

30

40

50

The interpretation of defects in forgings involves knowledge or experience of the expected types of defect and the possible signals from them. In some situations it is a case of reading the signal, evaluating which defects do not give this type of signal, then choosing from the remaining possibilities as to which type of defect is most likely. Here are a few of the types of defects found in forgings: Inclusions Inclusions in forgings are formed from lumps of trapped solid nonmetallic material in the original cast ingot and when forged out the shapes, sizes, orientations and depths of the defects vary. The varying orientation and shape have the effect of scattering the sound beam, as it passes through. When using a 0° probe on parallel sided forgings, if the sound reaches the back wall and reflects back, it scatters again on the return journey. This causes a significant drop in the amplitude of the BWE, compared to a defect free area. The amplitudes of the signals from the defects also vary because of the differences in sizes and orientations. The signals we see on the screen are a low, or no BWE and a cluster of signals, varying in amplitudes and depth from the defects. The cluster of signals from the defects has a constantly changing pattern when you move the probe across the surface.

60

70

80

90

100

Banding When alloys are added to the molten material in a cast ingot, some of them may not mix thoroughly and get left as segregated material in the centre of the ingot after solidification. These segregations get elongated and reduced in section in the rolling and forging processes, this is known as banding. If the acoustic impedances of the alloys and the base metal were different enough, ultrasonic reflections may occur. In steel forging they generally have an acoustic impedance that

© Institution of Mechanical Engineers Issue 17 2016

UT11-3

UNIT UT11  ULTRASONIC TESTING OF FORGINGS is similar to the steel, so they are not usually found ultrasonically unless the sensitivity of the equipment is high.

NOTES

10

20

Axial Looseness (grain structure) When material is reduced in size from ingot down to bar material it is common for the central area or the bar not to receive enough work to deform the grain size (forging reduction ratio). The grain structure can be observed ultrasonically and mistaken for axial bursts which would result in scrapping of the material. The ‘tell tale’ signs of axial looseness is that a backwall signal will still be observed, although reduced in height it is common to report the percentage loss of BWE.

30

40

50

60

A calculation of a ‘Bore to Clear’ (BTC) size is helpful to save rejection of the material. BTC can be calculated as nearest response to surface multiplied by 2 subtracted from the overall diameter.

70

80

90

A forging lap This defect occurs in the forging process by the material folding over onto itself and it is flattened, but not fused onto the surface. This usually leaves a visible seam on the surface of the forging. Using a 0° probe, scanning from the opposite side of the forging shows a signal appearing just before the BWE (2). On the defect side of the forging this defect is very easily missed because it is very near the surface and if using a single crystal probe the signals will be in the dead zone (3). However, evidence of this problem would be a total loss of back wall echo, providing the surface area of the lap is larger than the beam.

100

© Institution of Mechanical Engineers Issue 17 2016

UT11-4

UNIT UT11  ULTRASONIC TESTING OF FORGINGS NOTES

10

20

30

40

50

60

70

80

If a single crystal 0° probe is used, in (3) the defect signal will be in the near zone and dead zone on the screen. In both cases, if the defect is larger than the beam then the BWE will not be present. Slugs These are pieces of foreign material that have been pressed into the surface and give signal responses much the same as a lap.

90

100

Bursts Internal or surface ruptures of the material, caused by processing at too low a temperature or excessive working during forging. The signal response from this defect varies according to the shape, size and orientation of the defect. The normal rules of ultrasonic testing apply to the signals received, i.e. perpendicular orientation and large defect area give a good signal, oblique orientation and/or small defect area gives poor signals and © Institution of Mechanical Engineers Issue 17 2016

UT11-5

UNIT UT11  ULTRASONIC TESTING OF FORGINGS larger defect area than the beam causes a loss of BWE, etc. Deciding whether the defect is a burst or not requires careful plotting of the responses received to determine the shape and position.

NOTES

10

20

30

Hydrogen Flake (Hairline Cracking) What is hydrogen flaking? Hydrogen flakes are short, discontinuous internal fissures caused by stresses produced by localised transformation and decreased solubility of hydrogen during cooling. Hydrogen flaking is also referred to as internal hairline cracking, snow flakes, and shatter cracking. The primary source of hydrogen is water vapour which is in the atmosphere, furnace charge materials, slag ingredients and alloy additions, refractory linings, and ingot moulds. The water vapour reacts with the liquid metal at high temperatures to form hydrogen. Hydrogen solubility is much higher in molten steel (5-12 ppm) than in solid steel at room temperature (0.1 ppm). Therefore, as the steel cools the hydrogen precipitates in molecular form at imperfections such as inclusions, grain boundaries, or micro voids. The high pressures of this gaseous hydrogen cause localised cracking. Formation of flakes generally occurs at temperatures below 200°C.

40

50

60

Magnetic Particle Inspection of Flakes in Bar Stock 70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT11-6

UNIT UT11  ULTRASONIC TESTING OF FORGINGS ACCEPT/REJECT CRITERIA

NOTES

10

20

30

40

When defects are found it may be required that the defects are reported. Or the material may be accepted, rejected or graded according to the defects found. The accept/reject criteria tell us what size and type of defects to report, which defects render the component rejectable, or which defects to assess for grading of the material. The criteria can be found in a procedure, a written instruction sheet or in a national standard. REPORTING A report should give details of the component identification, test area, surface condition, equipment used, sensitivity settings and a drawing showing the defects and details such as; defect type, size, lateral and longitudinal positions in relation to datums, etc.. The report should be signed and dated and there may be a requirement to state your relevant qualifications. Alternatively, if accepting or rejecting the component or material, instead of a drawing you may be asked to give a written statement of conformity to the relevant acceptance level, or reasons for rejection, to the standard employed.

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT11-7

UNIT UT12  ULTRASONIC TESTING OF CASTINGS ULTRASONIC TESTING OF CASTINGS

NOTES

10

GENERAL The ultrasonic testing of cast products is limited, to some degree, by the scattering effects of the coarse grain structure and the rough surfaces produced on most casting processes. This scattering effect can be overcome by using lower frequency probes, but this results in a reduced sensitivity.

20

TECHNIQUE When searching for defects in castings you should have, as a minimum, the following information which is usually written on a technique or instruction sheet (see the appendices for an example). 30

40

50

60

70

80



The test component identification and area to test.



Actions to be taken when defects are found.



The purpose of the test (defects sought and acceptance criteria).



Equipment required.



What method and level of test sensitivity to use.



The method of scanning.

The instruction sheet would also contain sections giving details of any relevant safety procedures and post test procedures such as the cleaning of the test area afterwards. It would also have the company name, a unique technical reference number, the originator's name and signature and an authorising signature. Test area The test may involve testing the whole of a casting, or just sections of it, but this should be specified. Actions to be taken When defects are found it may be required that the defects are reported, e.g. on a diagram or as a written description, or the casting may be accepted or rejected according to the defects found. If defects are to be reported then the defect information that needs reporting would be contained in this section, i.e. defect type, size, lateral and longitudinal position in relation to datums, etc. Purpose of the test This section tells us the accept/reject criteria for particular defects, i.e. what size and type of defects to report or which defects render the casting rejectable.

90

Equipment This section should give information on; the type of flaw detector, type, size and frequency of probes, type of couplant, calibration blocks and reference blocks to use.

100

Sensitivity Method of setting and level of sensitivity need to be quoted for each scan, e.g. for the 0° probe; set the response from the 3 mm flat bottom hole reference reflector to 80% fsh and scan at this level. For the angle probes; use an 80% DAC curve from 3mm dia SDH’s.

© Institution of Mechanical Engineers Issue 17 2016

UT12-1

UNIT UT12  ULTRASONIC TESTING OF CASTINGS Most specifications for castings ask for an attenuation level to be met prior to test and is generally expressed in dB/mm or dB/m. (See example written instruction).

NOTES

10

See Appendix F for example calculation.

20

Scanning method The method of scanning the material is either a written step by step, instruction or technique sheet, or involves following the steps laid out in the relevant national standard. An example written step by step could be: 1. Prepare the material surface by removing any loose sand, rust, dirt or other debris and visually inspect for surface defects or damage.

30

40

2. Calibrate the screen on the flaw detector, using a 0° probe and the A2 calibration block, for a range of 0 to 200 mm. 3. Set the sensitivity (as quoted in the relevant section above) and apply couplant to the test area. 4. Scan the designated test area, with a probe overlap between scans of at least 20% of the probe's diameter and at a maximum probe movement rate of 150mm/sec. 5. When defects meeting the criteria in the "Purpose of the test" section are found, record the relevant defect data as in the "Actions to be taken" section.

50

*Refer to UT5-7 for usage/selection criteria of sizing methods

60

70

6. Defects larger than the ultrasonic beam, i.e. where there is no BWE present, should be sized using the 6 dB drop. Defects that are smaller than the ultrasonic beam should be sized and positioned using the 20 dB drop method. Multiple or multi-faceted defects should be sized using the maximum amplitude technique*. 7. Prepare a neat concise report giving details of the casting's identification, test area, equipment used, sensitivity method and settings and a drawing with the defect details as recorded in section 5 above. Sign and date the report and state your relevant qualifications. Post test procedures This would involve cleaning any remaining couplant and dirt from the test area and covering the surface with protective coatings according to client's requirements. DEFECTS IN CASTINGS

80

90

The interpretation of defects in castings involves knowledge or experience of the expected types of defect and the possible signals from them. In castings, the most common situation is no response from the defect and BWE due to the highly attenuative nature of the material and defect present. In other situations it is a case of reading the signal, evaluating which defects do not give this type of signal, then choosing from the remaining possibilities as to which type of defect is most likely. Here are a few of the types of defects found in castings:

100

© Institution of Mechanical Engineers Issue 17 2016

UT12-2

UNIT UT12  ULTRASONIC TESTING OF CASTINGS NOTES

10

20

30

Inclusions Inclusions are formed from lumps of trapped solid non-metallic material in the casting, of various shapes, sizes, orientations and depths. In large groups of small inclusions, the variation, in orientation and shape, has the effect of scattering the sound beam, as it passes through. When using a 0° probe on parallel sided castings, if the sound reaches the back wall and reflects back, it scatters again on the return journey. This causes a significant drop in the amplitude of the BWE, compared to a defect free area. The amplitudes of the signals from the defects also vary because of the differences in sizes and orientations. The signals we see from multiple inclusions are a cluster of signals, of various amplitudes and depth, from the defects and a low BWE, or no BWE. The cluster of signals from the defects has a constantly changing pattern when you move the probe across the surface. Larger inclusions will give stronger signals dependant on the shape, size and orientation.

40

50

60

70

Segregation When alloys are added to the molten material in a cast ingot, some of them may not mix thoroughly and get left as segregated material in the centre of the ingot after solidification. If the acoustic impedances of the alloys and the base metal were different enough, ultrasonic reflections may occur. In steel casting they generally have an acoustic impedance that is similar to the steel, so they are not usually found ultrasonically unless the sensitivity of the equipment is high.

80

90

A cold flake This defect occurs in the casting process by the material splashing up the sides of the mould, this defect is on but not fused to the surface. This usually leaves a visible "flake" of material on the surface of the casting. Using a 0° probe, scanning from the opposite side of the casting shows a signal appearing just before the BWE. On the defect side of the casting, this defect is very easily missed because it is very near the surface and if using a single crystal probe the signals will be in the dead zone.

100

© Institution of Mechanical Engineers Issue 17 2016

UT12-3

UNIT UT12  ULTRASONIC TESTING OF CASTINGS NOTES

10

20

30

40

50

60

70

If a double crystal 0° probe is used, in "B" the defect signal will be near zero on the CRT. In both cases, if the defect is larger than the beam then the BWE will not be present.

Scabs These are pieces of foreign material from the inside of the mould that have stuck to the surface of the casting and give signal responses similar to a flake if smooth, or may just scatter the beam if rough.

Cold shuts A lack of fusion resulting from splashing (a flake), surging, interrupted pouring or the meeting of two streams of molten metal coming from different directions. This defect gives a good signal response ultrasonically when favourably orientated to the beam.it is one of the few casting echo dynamic patterns that can be described as smooth/planar.

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT12-4

UNIT UT12  ULTRASONIC TESTING OF CASTINGS NOTES

10

Pipe or Shrinkage cavities Internal or surface voids in the material, caused by shrinkage during solidification or insufficient filling of the mould. The signal response from this defect varies according to the shape, size and orientation of the defect. The normal rules of ultrasonic testing apply to the signals received, i.e. perpendicular orientation and large defect area give a good signal, oblique orientation and/or small defect area gives poor signals and larger defect area than the beam causes a loss of BWE, etc.

20

30

40

50

Hot tears Surface or near surface cracks in the material due to different cooling rates at changes in section in a casting. Ultrasonic testing gives low amplitude multiple signals from multiple cracks or may give a high amplitude "ragged" signal from a large crack with the orientation of its major plane favourable to the beam.

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT12-5

UNIT UT12  ULTRASONIC TESTING OF CASTINGS NOTES

10

Porosity This volumetric defect gives a multiple low amplitude signal from all directions, access permitting. If porosity is severe then no response is possible due to the scattering of the sound beam. Using a 0° probe the BWE may reduce or totally disappear.

20

30

40

50

60

Blowholes These are small holes in the surface of a casting caused by the gas evolving from decomposing grease, moisture, etc. This defect is not readily found ultrasonically because it can be confused with rough surface signals normally obtained on some castings.

Airlocks Air trapped in the mould during pouring can be located ultrasonically and gives signal responses dependant on its shape, size and orientation.

70

80

90

100

ACCEPT/REJECT CRITERIA When defects are found it may be required that the defects are reported or the casting may be accepted or rejected according to the defects found. The accept/reject criteria tell us what size and type of defects to report or which defects render the component rejectable. The criteria can be found in a procedure, a written instruction sheet or in a national standard.

© Institution of Mechanical Engineers Issue 17 2016

UT12-6

UNIT UT12  ULTRASONIC TESTING OF CASTINGS REPORTING

NOTES

10

A report should give details of the casting identification, test area, surface condition, equipment used, sensitivity settings and a drawing showing the defects and details such as; defect type, size, lateral and longitudinal positions in relation to datums, etc. The report should be signed and dated and there may be a requirement to state your relevant qualifications.

20

Alternatively, if accepting or rejecting the casting, instead of a drawing you may be asked to give a written statement of conformity to the relevant acceptance level, or reasons for rejection, to the standard employed.

30

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

UT12-7

APPENDIX A  BRITISH STANDARDS BRITISH STANDARDS RELATING TO ULTRASONIC TESTING

NOTES

10

BS EN ISO 17640: 2010 Non-destructive Testing of Welds, Ultrasonic Testing. Techniques, testing levels and assessment. BS EN ISO 11666: 2010 Non-destructive testing of welds. Ultrasonic Testing. Acceptance levels. (Supersedes BS EN 1712: 1997)

20

30

BS EN ISO 23279: 2010 Non-destructive testing of welds. Ultrasonic Testing. Characterisation of indications in welds (Superseded BS EN 1713: 1998) BS EN ISO 16810-2014 Non-destructive testing. Ultrasonic (Supersedes BS EN 583-1-1999)

testing.

General

Principles.

BS EN ISO 16811:2014 Sensitivity and range setting

40

BS EN ISO 16823:2014 Transmission technique BS EN ISO 16826:2014 Ultrasonic examination for imperfections perpendicular to the surface BS EN ISO 16827:2014 Characterisation of sizing of imperfections

50

60

BS EN ISO 16828:2014 Time of flight diffraction technique as a method for detection and sizing of discontinuities BS EN 10160: 1999 Ultrasonic testing of steel flat product of thicknesses of > 6 mm (reflection method) (superseding BS 5996) BS EN 10079: 2007 Definition of steel products BS EN 10228: Part 3: 2016

70

Non-destructive testing of steel forgings. Ultrasonic testing of ferritic or martensitic steel forgings BS EN 10228: Part 4: 2016 Non-destructive testing of steel forgings. Ultrasonic testing of austenitic and austenitic-ferritic stainless steel forgings

80

90

BS EN 12668: Part 3: 2013 Characterisation and Verification equipment

of

UT

equipment:

Combined

BS EN ISO 2400: 2012 Non-destructive-testing - Ultrasonic examination - Specification for calibration block no. 1 BS EN ISO 7963: 2010 Ultrasonic Testing: Specification for Calibration Block No. 2

100

BS EN 1330: Part 1: 2014 Non-destructive-testing terminology: List of general terms

© Institution of Mechanical Engineers Issue 17 2016

APPA-1

APPENDIX A  BRITISH STANDARDS BS EN 1330: Part 2: 1998 Non-destructive-testing terminology: Terms common to NDT methods

NOTES

10

BS EN 1330: Part 4: 2010 Non-destructive-testing terminology: Terms used in Ultrasonic testing BS EN 12680-1: 2003 Ultrasonic Examination: Steel Castings for general purposes.

20

BS EN 10308: 2002 Ultrasonic Testing of Steel Bars BS EN 14127: 2011

30

Ultrasonic Thickness Measurement BS 2704:1978

40

Calibration blocks for use in ultrasonic flaw detection (Obsolete, replaced with BS EN ISO 2400:2012 – however, no drawings of A6 or A7 block included)

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

APPA-2

APPENDIX B  FORMULAE NOTES

FORMULAE USED IN ULTRASONIC TESTING 10

Wavelength

v f

 

Where:  = Sound wavelength (mm) v = Material sound velocity (mm/s) f = Sound frequency (Hz)

20

Near zone

D2 N  4

D2f or 4v



f v

30

40

N D

Half beam angle

Sin  

K Kv or D Df

 D v f



K

= = = = =

Near zone (mm) Crystal diameter (mm) Sound wavelength (mm) Sound frequency (Hz) Material velocity (mm/s)

= = = = = =

Half beam angle (degrees) Crystal diameter (mm) Material velocity (mm/s) Sound frequency (Hz) Sound wavelength (mm) Constant: 1.22 for 0% edge 1.08 for 10% edge 0.56 for 50% edge of the beam

50

60

Snell’s law

Sin  V1  Sin  V2

70

Decibel

dB  20 log10

 

= = V1 = V2 =

H1 H2

Incident (wedge) angle (degrees) Refracted (probe) angle (degrees) Velocity in medium 1 (m/s) Velocity in medium 2 (m/s)

dB = Decibel H1 = 1st signal height (100%) H2 = 2nd signal height (% of H1)

80

Crystal thickness

t 

V 2Ff

t V Ff

90

Material velocity

V 

T x CV TB

100

© Institution of Mechanical Engineers Issue 17 2016

V T TB CV

= Crystal thickness (mm) = Sound velocity in crystal material (mm/s) = Fundamental frequency the crystal vibrates at (Hz)

= = = =

Unknown velocity (m/s) Material actual thickness (mm) Time base reading (mm) Calibration block velocity (m/s)

APPB-1

APPENDIX C  TABLE OF ACOUSTICAL VELOCITIES NOTES

TABLE OF ACOUSTIC VELOCITIES IN DIFFERENT MATERIALS 10

20

30

40

50

Material

Compressional or longitudinal wave velocity (m/s)

Shear or transverse wave velocity (m/s)

Aluminium Brass Cast iron Copper Gold Iron Lead Oil Perspex Mild steel Stainless steel Water Tungsten Zinc Zirconium

6,400 4,372 3,500 4,769 3,240 5,957 2,400 1,440 2,740 5,960 5,740 1,480 5,174 4,170 4,650

3,130 2,100 2,200 2,325 1,200 3,224 790 1,320 3,240 3,130 2,880 2,480 2,300

The velocity in a medium depends upon the medium’s density and elasticity.

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

APPC-1

APPENDIX D  TABLE OF ACOUSTIC IMPEDANCES NOTES

TABLE OF ACOUSTIC IMPENDANCES FOR DIFFERENT MATERIALS 10

Medium

20

30

40

50

60

70

80

Air Aluminium Barium titanate Beryllium Brass Cast iron Copper Glass (plate) Gold Iron Lead Lithium sulphate Magnesium Mercury Molybdenum Nickel Oil Perspex Platinum Quartz Steel Stainless steel Silver Tin Titanium Tungsten Tungsten araldite Tungsten carbide Uranium Water Zinc

Compression velocity (m/s) 330 6,400 5,260 1,289 4,370 3,500 4,760 5,770 3,240 5,960 2,160 5,450 5,790 1,450 6,250 5,480 1,440 2,740 3,960 5,730 5,960 5,740 3,700 3,380 5,990 5,170

Shear velocity (m/s) 3,130 888 2,100 2,200 2,330 1,200 3,220 700 3,100 3,350 2,990 1,320 1,670 3,240 3,130 1,700 1,610 3,120 2,880

2,060

Density (g/cm2)

Acoustic impedance

0 2.7 5.7 1.8 8.45 7.2 8.93 2.5 19.3 7.85 11.4 2.1 1.74 13.55 10.2 8.85 0.9 1.2 21.4 2.65 7.8 7.8 10.5 7.3 4.5 19.3

0 17.2 30 23.2 37 25 42.5 14.5 63 46.8 24.6 11.2 10.1 19.6 63.7 48.5 1.3 3.2 85 15.2 46.5 44.8 36.9 24.7 27 100

-

10.5

21.65

6,650

3,980

10

66.5

3,370 1,480 4,170

2,020 2,480

18.7 1 7.1

63 1.48 29.6

90

100

© Institution of Mechanical Engineers Issue 17 2016

APPD-1

APPENDIX E  ATTENUATION FACTOR EXAMPLE METHOD FOR DETERMINING THE ATTENUATION FACTOR OF A MATERIAL

NOTES

10

20

0° probe method Using a calibrated timebase, place the probe on the material to be measured. Select two back wall echoes at a distance ratio of 2:1, the first one being at least three near zone distances from zero. Measure the difference in amplitude, in dB’s, of the two signals and record their range difference. The back wall echo decreases in amplitude by 6 dB for every doubling of the range. The attenuation factor can be determined by subtracting 6 dB from the amplitude difference (in dB) and dividing this by twice the range difference (return journey of the sound).

30

1

BS EN 583-2: 2001

40

Note: when referring to ‘transfer correction’ this normally refers to the difference in the surface profile between 50 a reference block and the test material, which can cause attenuation losses.

This method becomes less accurate as the number of multiple echoes used increases, due to the fact that about 1 dB of sound re-enters the probe, on each bounce of the sound, at the probe to material interface. Correction for attenuation and transfer loss When reference or calibration blocks are used, there may be attenuation differences between the block and test object (in surface condition or material). Methods for determining these attenuation differences are given in the current standards1.

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

APPE-1

APPENDIX F  EXAMPLE CALCULATIONS NOTES

EXAMPLE CALCULATIONS USED IN ULTRASONICS When using these formulas write down the formula and the values for each part prior to starting.

10

Wavelength (  ) To calculate the wavelength of a 5.0 MHz, 0° probe when used on steel. Frequency (f) = 5.0 MHz

20

30

Velocity of compression waves in steel (v) = 5,960 m/s

 

v f

 

5,960 m/s 5,960,000 mm/s  5 MHz 5,000,000 Hz

  1.19 mm

Near zone (N) 40

To calculate the near zone of a 20 mm diameter, 5.0 MHZ, 0° probe used on aluminium. Frequency (f) = 5.0 MHz Velocity of compression waves in aluminium (v) = 6,400 m/s Crystal diameter (D) = 20 mm

50

N

60

70

D2 D2 f or x 4 4 v

N 

20 mm 2 5 MHz x 4 6400 m/s

N 

5,000,000 Hz 400 mm x 6,400,000 mm/s 4

N  78.125 mm

Half beam angle (θ) 80

To calculate the half beam spread from a 10 mm diameter, 5.0 MHz, 0° probe used on steel. Frequency (f) = 5.0 MHz Velocity of compression waves in steel (v) = 5,960 m/s

90

Crystal diameter (D) = 10 mm Constant (K) = 1.22 (assume extreme edge of beam K-factor unless otherwise stated)

100

© Institution of Mechanical Engineers Issue 17 2016

APPF-1

APPENDIX F  EXAMPLE CALCULATIONS NOTES

10

20

30

40

Sin

1 K K v or x   2 D D f

Sin

5,960 m/s 1 1.22   x 5 MHz 2 10

Sin

5,960,000 mm/s 1 1.22   x 5,000,000 Hz 2 10

Sin

1   0.145424 2

1 Angle   8 22' 2 Snell’s law To calculate the incident (wedge) angle, in perspex, required to produce a 60° refracted (probe) angle in steel (   = incident angle). Refracted angle (  ) = 60° Velocity of compression waves in perspex = 2,740 m/s

50

Velocity of shear waves in steel = 3,240 m/s

Sin  V1 V1  transposes to Sin   x Sin  Sin  V2 V2 60

70

Sin  

2,740 m/s x Sin 60 3,240 m/s

Sin  

2,740 x 0.866025403 3,240

Sin   0.732379508 Angle   47 05'

80

Transit time To calculate the time taken for a longitudinal wave to travel through a piece of steel 20 mm thick and return to the probe. Distance travelled (D) = 40 mm (2 x thickness) Velocity of longitudinal waves in steel (v) = 5,960 m/s

90

Transit time (μ sec)  Transit time 100

D (mm) v (km/s)

40 mm 5.96 km/s

Transit time  6.7 μsec © Institution of Mechanical Engineers Issue 17 2016

APPF-2

APPENDIX F  EXAMPLE CALCULATIONS Clock interval

NOTES

To calculate the time between pulses of energy when the pulse repetition frequency (prf) is set at 4 KHz (4000 Hz). 10

Clock interval (µsec) 

20

Clock interval 

1 prf (MHz)

1 0.004 MHz

Clock interval  250 µsec 30

40

Maximum testable thickness To calculate the maximum thickness of steel we can test with a set prf, we calculate the clock interval, then calculate distance travelled by the sound in that time and divide by two (for return journey). Using the above example prf, the maximum thickness would be: Velocity of compression waves in steel = 5,960 m/s Time (clock interval) = 250 µsec

Maximum thickness (mm)  50

Maximum thickness 

Time (µsec) x Velocity (km/s) 2

250 x 5.96 2

Maximum thickness  745 mm 60

Attenuation

70

80

To calculate the attenuation of a material (either in dB/mm or dB/m) note the difference in dB’s between the first and second backwall. Subtract 6dB from this figure to account for natural loss (inverse law) then divide the figure by the distance the sound beam travels, i.e.: there and back. If there was a 7dB difference in 2 BWE’s and the material is 50mm thick Attenuation dB/mm

7dB – 6 dB (natural loss) (50 mm x 2) distance travelled

= 0.01 dB/mm

90

or

=

1dB 100mm

10dB/m

100

© Institution of Mechanical Engineers Issue 17 2016

APPF-3

APPENDIX F  EXAMPLE CALCULATIONS Pipe Wall Thickness

NOTES

10

20

30

Calculate the beam angle that just grazes the bore of a 100mm outside diameter pipe having a 20mm wall thickness. Sin

= 1-

Sin

= 1-

Sin

= 1- 0.4

Sin

= 0.6 = 37°

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

APPF-4

APPENDIX G  A6 AND A7 BLOCKS A6 CALIBRATION BLOCK

NOTES

10

Reference: BS 2704:1978 Calibration blocks for use in ultrasonic flaw detection (Obsolete, replaced with BS EN ISO 2400:2012 – no drawings of A6 or A7 block included)

20

30

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

APPG-1

APPENDIX G  A6 AND A7 BLOCKS A7 CALIBRATION BLOCK

NOTES

10

Reference: BS 2704:1978 Calibration blocks for use in ultrasonic flaw detection (Obsolete, replaced with BS EN ISO 2400:2012 – no drawings of A6 or A7 block included)

20

30

40

50

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

APPG-2

APPENDIX H  PIPE WALL THICKNESS For example purposes only.

NOTES

Maximum pipe wall thickness for probe angles 10

20

30

40

50

Pipe OD 38°

45°

60°

70°

4” (100mm)

19mm

14mm

7mm

3mm

6” (150mm)

29mm

22mm

10mm

4mm

8” (200mm)

38mm

29mm

13mm

6mm

10” (250mm)

48mm

36mm

17mm

7mm

12” (300mm)

58mm

44mm

20mm

9mm

14” (350mm)

67mm

51mm

23mm

10mm

16” (400mm)

77mm

58mm

27mm

12mm

18” (450mm)

86mm

66mm

30mm

13mm

20” (500mm)

96mm

73mm

33mm

15mm

Pipe Schedule Charts

60

70

80

90

100

© Institution of Mechanical Engineers Issue 17 2016

APPE-1