Kuppan T. Heat Exchanger Design Handbook
Подождите немного. Документ загружается.
500
944
Chapter
I4
Depth of Penetration and Frequency
The depth of penetration of eddy current is a critical factor. For example, in the case of tube
inspection, if the eddy currents do not penetrate the wall thickness, then it will miss the defects.
The depth of penetration of eddy current is given by the following equation [69]:
‘I;=-
6f
where
‘I;
is the standard depth of penetration (mm),
G
the conductivity (m/R-mm’),
p
the
magnetic permeability,
f
the test frequency (Hz), and
p
is the magnetic permeability
(1
for
nonmagnetic materials), and
f
is the test frequency (Hz).
The standard depth of penetration is generally taken to be that depth at which the eddy
current field intensity drops to
37%
of the intensity at the conductor surface. For tube inspec-
tion, the test frequency is often the frequency at which the depth of penetration is equal
to
the
tube wall thickness. This is given by the equation [69]:
where
t
is the tube wall thickness (mm).
When phase analysis is being applied, the
fso
frequency, that is, the frequency at which
there is a 90” phase difference between signals from the inside wall of the tube and signals
from the outside wall, will be used. Thefw frequency is given by the equation [69]:
where
p
is the resistivity (pohm*cm).
Fill Factor and Probe Size Requirements
The probe size requirements for eddy current tube testing are determined by the
“fill
factor,”
given by the expression [69]:
Fill factor
=
E]
where
d,
is the diameter
of
the probe and
d2
the inside diameter
of
the tube.
For an internal or bobbin type coil, the fill factor indicates how well the inspection coil
fills
the inside of the tube being inspected. Ideally the fill factor should be close to
1.0.
A
fill
factor of
1.0
will not allow smooth travel of the probe inside the tube. Therefore, as a thumb
rule, the optimum
fill
factor for tube testing can be about 0.70 [69]. This allows reasonable
sensitivity to be achieved while still maintaining adequate clearance when dirt or dents may
be present in the tube.
Edge Effect
When
an
inspection coil approaches the end or edge of a part being inspected, the eddy currents
are distorted, because they are unable to flow beyond the edge of the part. The distortion of
eddy current results in an indication known as “Edge effect”. In general, it is not advisable to
inspect any closer than
94
in
(3.2
mm) from the edge of a part [70].
Skin Effect
Eddy currents are not uniformly distributed throughout a part being inspected. They are densest
at the surface immediately beneath the coil and become progressively less dense with increas-
QA,
QC,
Inspection,
and
NDT
945
ing distance below the surface. This phenomenon is known as “skin effect.” At some distance
below the surface of a thick wall there will be essentially no current flowing.
13.8 Inspection Method for Tube Interior
The probe, usually an internal bobbin type, is introduced into the tube under test and passed
along the length of the tube to the far end, usually by means of compressed air, but also by
other means. It is then withdrawn from the tube at a constant speed, typically in the range of
0.5
to
1.0
m/s,
using a winch unit [34,69]. During the probe withdrawal, the eddy current
signals are recorded on a chart and analyzed further. The tube inspection method is shown
schematically in Fig. 38.
Inspection of Ferromagnetic Tubes
The highly ferromagnetic nature of carbon steel and ferritic stainless steels severely limits the
application of standard eddy current technique. Nonferromagnetic materials have a relative
permeability of approximately 1, whereas ferromagnetic materials have typical relative perme-
abilities of 500-2000. Since the higher the permeability
is,
the lower is the depth
of
penetra-
tion, eddy currents in ferromagnetic materials are concentrated at the surface, and hence buried
or back-wall defects remain undetected. Also, small variations
in
permeability give rise to
relatively high noise levels [69]. This limitation is overcome by saturating the ferromagnetic
tubes magnetically with a direct current,
so
that the material reacts as nonmagnetic material
when inspecting [34], or by remote field slit eddy current
(RFEC)
technique. The drawbacks
of these two methods are discussed by Bergander [72]. Such drawbacks are as follows:
(1)
Since magnetic saturation uses high amperage current for saturation, there is a need for probe
cooling, the possibility of false indications due to variations in permeability, difficulty
in
detecting gradual tube thinning, and complicated probe design.
(2)
To make the
RFEC
tech-
nique sensitive to gradual wall thinning, the probes has to be complex. To overcome these
difficulties, Bergander [72] suggests using a magnetic flux leakage method to examine ferritic
tubes. The principle of flux leakage as applied to ferrous heat exchanger tubing is illustrated
in Fig. 39.
Figure
38
Tube
inspection
by eddy
current testing
method.
(From
Ref.
69.)
946
Chapter
14
WICNETIC CIRCUIT
-
L
1
/
a
ra
X2La
CRT
~~ ~
L
I
PWGNETIC TAPE
STRIP CHFIRT
Figure
39
Principle
of
flux
leakage inspection
of
ferrous tubes.
(From
Ref.
62.)
13.9
Testing
of
Weldments
Because of the circumferential orientation of eddy currents flow, this method is effective
in
detecting most types of longitudinal weld discontinuities such as open welds, and weld cracks.
On the other hand, difficulty arises for the detection of a thin planar discontinuity that is
oriented substantially perpendicular to the axis of the cylinder.
13.1
0
Calibration
In order to provide accurate and repeatable results, artificial defects of known size are intro-
duced into a specimen piece of tube of the same material and dimensions as the tubes to
be
inspected. Reference defects can be of various types. The most commonly used reference
standard consists of a number
of
through holes in a tube as shown in Fig.
40,
along with their
signal pattern
[73].
13.1
1
Remote Field Slit
Eddy
Current Testing
The remote field slit eddy current
(RFEC)
inspection technique
is
similar to the
ET
method,
which uses low-frequency alternating current and through-wall transmission to inspect pipes
and tubes from the inside. The method is most preferred to apply to ferromagnetic materials
QA,
QC,
Inspection, and NDT
94
7
Figure
40
Reference
standard
for eddy current testing. (From Ref.
73.)
because conventional eddy current testing techniques are not suitable for detecting opposite-
wall defects in such materials unless the material is saturated [74].
Principle. The principle
of
RFEC
is
explained in Ref. 74 and by Atherton et al.
[75].
A
typical
RFEC
probe, shown schematically
in
Fig.
41,
uses a relatively large internal exciter
coil, which is driven with a low-frequency alternating current.
A
detector coil is placed near
the inside of the pipe wall, but axially displaced from the exciter coil by about two pipe
diameters. Two distinct coupling paths exist between the exciter and the detector coils:
(1)
the
direct path inside the tube, which
is
attenuated rapidly by circumferential eddy currents induced
in
the tube wall, and
(2)
the indirect coupling path that originates in fields that diffuse outward
through the pipe wall
in
the vicinity of the exciter. Anomalies anywhere
in
this indirect path
cause changes
in
the magnitude and phase of the received signal, and can therefore be used
to
indicate defects. This method has been used primarily for inspecting well casing, small-diame-
ter boiler tubes, and heat exchanger tubes.
Merits and Demerits.
An
important advantage
of
this method is that the method has higher
sensitivity to axially and circumferentially oriented flaws in ferromagnetic materials. This al-
lows detecting corrosion damage due to erosion-corrosion, tube wall thinning, pitting, and
Figure
41
Remote field eddy current inspection technique.
(From
Ref.
75.)
948
Chapter
I4
stress corrosion cracking [74]. The major disadvantage is that when applied to nonferromag-
netic material,
it
is not generally as sensitive or accurate as conventional eddy current testing
method.
13.12 Merits of ET and Comparison with Other Methods
ET is attractive because it offers both very high sensitivity and relatively high scanning speeds.
Direct contact with the test material is not required and a coupling medium is not necessary;
the process is repeatable, the test method is relatively fast, and
it
can be easily automated.
13.13 Limitations
of
Eddy Current Testing
The major limitations of eddy current testing are:
1.
Skin effects usually restrict the depth of inspection. It is generally limited to
0.25
in
(6.4
mm) for nonferromagnetic materials and 0.010 in
(0.25
mm) for ferromagnetic materials
[28]. The limitation due to high permeabilities of ferromagnetic materials may be signifi-
cantly overcome by magnetic saturation of the area being inspected or by
flux
leakage
method.
2.
Since many variables, namely, permeability, conductivity, probe position, and weld
con-
tour, can affect an eddy current signal, an accurate measurement of one property requires
the ability to eliminate or at least considerably reduce interference from other properties
[76]. Multifrequency techniques are used in practice to increase the information content of
signals [73]. A classical solution, due to recent developments
in
ET, which can reduce but
not eliminate this problem, is the application of different
ET
probes [71].
Other limitations include [69]:
3.
The signal is more closely related to volume of material lost than to wall thickness lost.
4.
Defects at the tube sheet, and at the tube and baffle plate interface, can be difficult to
detect.
5.
For critical applications, results may need to be verified by an alternative technique.
13.14 Recent Advances in
Eddy
Current Testing
The development of computer interfacing for data collection, analysis, and display has led to
multiple-frequency testing, eddy current imaging, multiple-element eddy current probes, and
automated data interpretation systems that significantly enhance both the capabilities and relia-
bility of the eddy current inspection [71].
13.1
5
Tube-Sheet Diagram for Windows
The tube-sheet diagram for Windows (TSD) is computer software intended for use by inspec-
tion groups in the oil, gas, and power industries
to
inspect tubes periodically in heat ex-
changers, boilers, steam generators, chillers, air conditioners, and similar multitube equipment
and to award a classification based on user-defined criteria such as corrosion severity, material
type, defect position, etc. [77]. TSD is shown in Fig. 42.
Automated Tube Inspection System
(ATIS)
ATIS is a software package that carries out automated analysis of the data from eddy current
and electromagnetic in-service tube inspection systems in real time
[78].
The system discrimi-
nates between baffle support plate signals and defect signals and can be utilized on a wide
variety of tube materials. The results are classified in terms of defect severity and stored in a
QA,
QC,
Inspection,
and
NDT
949
hard disk, from which they can be retrieved and summarized at any time. TSD can accept
manual input from the keyboard or the mouse and can also import results from the silver-wing
ATIS (automatic tube inspection system) data files. Presently the system is compatible with
both standard eddy current instruments and the CENTEST 91 ferrous tube inspection flaw
detector.
14
LEAK TESTING (LT)
Leak testing (LT) is a nondestructive testing technique used for the detection and location of
leaks and for the measurement of fluid leakage in either pressurized or evacuated vessels and
components [79]. A leak may be a crack, fissure, hole, or passage that admits any fluid or lets
fluid escape. The operational reliability and serviceability
of
many devices under pressure or
vacuum is greatly reduced if sufficiently large leaks exist. Leak testing is performed for three
basic purposes:
1.
To prevent material leakage loss.
2.
To prevent environmental contamination or hazards, particularly ones that involve han-
dling of radioactive or lethal fluids.
3.
To
detect unreliable components.
For pressure vessels and heat exchangers, leak testing is used to ensure the leak-tightness of
various welded joints, flanged joints, tube-to-tube-sheet joints, mechanical joints, enclosures,
etc.
14.1
Written Procedure
Leak testing procedures should encompass one or a combination
of
the following [go]:
Locating leaks
Determining the leakage rate
Monitoring for leakage
ASME Code Section
V
specifies the minimum information to be provided for the LT proce-
dure, which include the following:
1.
Scope and extent of the examination.
2.
Type of equipment to be used for detecting leaks or measuring rate
of
leakage.
3.
Surface cleanliness and surface preparation.
4.
Leak test method.
5.
Temperature, pressure, gas and percentage concentration to be used.
14.2
Methods
of
Leak
Testing
The basic principle of leak testing involves identification and location of the leaks or the
measure of the flow
of
fluid or the testing medium quantitatively through the leak. This could
be the actual flow rate, drop in pressure or vacuum level, etc. Various leak testing methods
are detailed in Refs.
79
and 80, ASME Code Section
V,
and Yokel1
[
181. The following are
the different leak testing methods available, arranged in a rough order of increasing sensitivity.
1.
Ultrasonic detection
2.
Bubble test-direct pressure technique or immersion technique, or vacuum box method
3.
Gas leak lake testing
4.
Pressure change test
950
Chapter
I4
5.
Halogen diode detector probe test
6.
Helium mass spectrometer detector (sniffer) probe technique
7.
Helium mass spectrometer tracer probe technique
8.
Helium mass spectrometer hood method.
Tests for gross leaks involve direct visual inspection for signs of tracers or leakage prod-
ucts, or scanning of leaking systems with an ultrasonic leak detector to locate the sources of
hissing sounds. Small leaks can be detected by sophisticated leak testing methods utilizing
tracer gases such as halogen vapors and helium with detectors such as the heated anode halogen
detector and the helium mass spectrometer, respectively. Leak testing methods listed in items
1
to 8 are discussed next. Yet another method of leak testing is acoustic emission testing, and
this method is explained in the section on acoustic emission.
Ultrasonic Leak Detection
The leakage of a gas through small openings, even for small pressure differences, generates
noise in the ultrasonic range. Therefore, it
is
possible to detect the leaks by listening for them
using an ultrasonic listening device, which can record low noise levels produced
in
the high
frequencies (40-50 kHz) that occur due to turbulent flow [34].
Bubble Testing
Bubble testing can be performed by the direct pressure technique or vacuum-box technique.
The direct pressure technique using bubble solution is the more common one.
Bubble Testing, Direct Pressure Technique.
The objective of the direct pressure technique is
to locate leaks in a pressurized component by application of a solution or by immersion
in
liquid that will form bubbles as the leakage gas passes through it. To perform this test, pressur-
ize the shell with inert gas or air; the required pressure differential can be from 1 kgkm’ to
1.25 times the design pressure of the vessel but less than
20
kg/cm’. After sufficient pressure
soak
time, apply a soap solution over the joints. Generally the surfaces must be in the tempera-
ture range of 40-125°F (4.5-52°C). This method can detect leaks of the order of
10-s
pascal
m’/sec when the test objects are immersed in detection liquids.
Bubble Testing, Vacuum Box Technique.
The objective of the vacuum box technique is to
locate leaks in a pressure boundary that cannot be directly pressurized. In this method, on one
side of the test joint after application of soap solution, vacuum is created and
if
there are any
through leaks, atmospheric air from the other side will come out to the vacuum side and will
be observed as the formation of a bubble.
Gas Leak Lake Testing
For this method, pressurize the shell side of the heat exchanger with air or nitrogen to the
maximum allowable working pressure
(MAWP)
at room temperature, set the unit vertically.
With a ring of about 2
in
height (50.8 mm), make a dam on the tubesheet face. Fill the tube
side with water to the level of the dam, and observe for bubbles.
Pressure Change Test
This test method describes the technique for detecting the leakage rate of the boundaries of a
closed compartment or the vessel at a specific pressure or vacuum by indicating pressure or
vacuum change over the given time period.
Halogen Diode Detector Testing
Principle.
A
detector or a sniffer sucks a mixture of air and leaking tracer gas through a
tubular probe into an instrument that is sensitive to small amounts of tracer gas. The tracer
gas
can be a halide gas (Refrigerant 12) to a maximum of 25% of the design pressure.
A
heated
- -
951
QA,
QC,
Inspection,
and
NDT
Suitable for use in..
.
*
PowaGmerationPlant
OilandGasRefining
PetdhemicalPlant
Air
Conditioning Plant
Engincuing andContracting
*
PlantMaintmance
onshore
and
mshore
utilities
InspectionDepartmmts
Savice
Inspection Companies
Use
the
Report Generator to create
a
complete report..
.
add
or ddete
frames
to
meet
individual
preference,
drag
and
drop
frames
anywhere on
the
page,
*
resize
diagrams
chart, or text
for
case
of viewing,
use
text
only
to create title pages
and
written reports
report
on
any
Windows
compatible
@a
(colour
or Ww).
*
print
in
A3,
A4,
US
Lma,
US
Legal,
etc..
use
portrait
or
landscape orientation.
Some
of
the
TSD
feature
include..
.
User defined templates
with
up to
200
x
200
tubes,
Up
to
48
user
defined
tube
classifications
per
template,
Up
to
12
tube
shapes
to
aid
recognition,
Unlimited number
of
data
Layers
per
template
Unlimited
colours
for
tube
fills,
and
tube
borders
Automatic generation
of
Charts
and
Tables
TSD
CM
report
and
analyse
the results
of
almost
any
tube
inspection
method
including eddy currents. ultrasonics, CCTV.
- -
material
types.
mechanical properties,
etc.
7
95%
10
35%
I
TSD
U
fu&
compatible
with
Siiva
Wirrg
'ATIS'
&taJila
4
-
00%
-
I
i
11
24%
1s
3.3%
CLa
3
(I I)
2
Q4%
I
14
3oy
I
Figure
42
Tubesheet
diagram
for
Windows.
952
Chapter
14
platinum element (anode) in the detector ionizes the halogen vapor, and the ions are collected
on a cathode plate.
A
current proportional to the rate of ion formation is measured.
Leak
Testing
Method.
Traverse the tube ends with
a
probe at a distance of about
!4
in
(3.2
mm) and at
a
predetermined scanning rate using a standard leak. Also, insert the probe into
the tube ends for about
!4
in
(6.35
mm). Check for cracked tube ends by inserting the probe
into each tube end to a minimum depth of
%
in
(19
mm)
and holding it there for about
3
sec.
If
the tubes are welded to the tube sheet, check the welds by encapsulating each tube end with
a funnel connected to the probe
[
1
11.
When the tubes are too small to allow the probe to enta
and too close
to
each other for effective scanning, use the halogen sniffer in a bag test system.
The halogen diode detector leak test method is a semiquantitative method to detect and locate
leakages and should not be considered quantitative. The sensitivity of this method is
10”
~a*m’/sec.
Helium Leak Detection Methods
Principle.
For the manufacture of nuclear components, leak testing using helium as a tracer
gas is one of the mandatory code and customer requirements. Helium leak detection can
be
done either by pressurizing the vessel with helium and sniffing (detector) the welds or joints
IHm
I&
test.-
of
evacuoted
vard
t+cliwn
kok
t&kq
d
pesur)zed system
system with
tracer
pok
with
sanp(ing
probe
a
b
Le&
trsiinq
of
evacuoted components inserted
inlo
1
I
1
I
2-
hood
or
em-elope
contairiqhrlium
miatwe
System
mder
test
(evacuated)
Hood
cmconng
helium-air
mixture
I
C
d
Figure
43
Helium mass spectrometer testing. (a) Tracer probe technique; (b) detector probe tech-
nique; (c) vacuum technique
or
hood method; and (d) typical test set-up. (a-c from Ref.
79.)
953
QA,
QC,
Inspection,
and
NDT
ATHOSPHERIC
vAcuun
n
PRESSURE
4b
I
PLUG
CONVENT
1
ONAL
BUBBLER
RESERVOIR TUBE PLATE
a
BOX
Figure
44
Inservice examination of leaks using. (a) Bubbler;
and
(b) soap foam.
(From
Ref.
43.)
from outside or by evacuating the vessel and spraying helium on the welds from outside by a
tracer probe or enclosed in a vacuum mask. In each case, the helium atoms passing through a
leak are collected to a mass spectrometer, where they
are
ionized. The current proportional to
the quantum
of
ions is amplified and measured as the leak rate.
Helium
Mass
Spectrometer Test.
The basic methods of helium leak detection include
[79]:
1.
Detector probe technique
2.
Tracer probe method
3.
Vacuum technique or hood method