i
l
op
I
SV
PV
:
.
.
TOFI
ORNL P
2615
.
.
-
14
-
S
og
?
r.
arine
.
.
het
.
nl
.
1
-
ha
:
.
7
{
:
...
.
i
..
.
.
1.25 1.1.4 1.6
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS – 1963
-
...o
nt
finizle 15.......
9-28-66 WIMSICA
NOV 2 9 1966

H..
CODES, SPECIFICATIONS AND ACCEPTANCE STANDARDS FOR WELDMENTS
10.8 1.00 en 50
Edward C. Miller
Oak Ridge National Laboratory*
CONF-660936
Let me preface this talk with a disclaimer--that while I am a member of several
committees involved in writing or maintaining certain codes, standards and
specifications, the statements and opinions I offer are strictly my own and do
not represent official opinions or interpretations of these code committees or
77
of my employers.
The terms codes and specifications can be defined with some exactness, but
they are all too frequently used without much regard for the definitions. A
specification represents a compilation of technical details used as a basis for
purchase; this may be a company specification, or, if generally recognized,
accepted and used by the greater part of an industry, it is better described as
a standard specifications. Specifications may contain technical provisions re-
lated to functioning, economy, and appearance; Codes may include some of
these technical details too, but they place principal emphasis on items related
to safety; they are frequently adopted as the legal requirements of some
-"*"
national or local governmental regulatory agency, or, if they are generated and
-
- ...
-
accepted as standards for an industry by a consensus of manufacturers, users,
and public interest groups, they may be acceptable to courts as prima facie
* Operated by Union Carbide Corporation for the U.S. Atomic Energy Commission

LEGAL NOTICE
RELEASED FOR ANNOUNCEMENT
IN NUCLEAR SCIENCE ABSTRACTS
This report was prepared as an account of Government sponsored work. Neither the United
States, nor the Commission, nor any person acting on behalf of the Commission:
A. Makes any warranty or representation, expressed or implied, with respect to the accu-
racy, completeness, or usefulness of the information contained in this report, or that the use
of any information, apparatus, method, or process disclosed in this report may not infringe
privately owned righta; or
B. Assumed any liabilities with respect to the use of, or for damages resulting from the
use of any information, apparatus, methrd, or process disclosed in this report.
As used in the above, "person acting on behalf of the Commission" Includes any em-
ployee or contractor of the Commission, or employee of such contractor, to the extent that
such employee or contractor of the Commission, or employee of such contractor prepares,
disseminates, or provides access to, Any information pursuant to his employment or contract
with the Commission, or his employment with such contractor.
"
Ti
FIT
..............
..
. in...
.
powertheca.-.
.......
thithuaid
that time..'
9-28-66
- 2
-
evidence of a fixed standard of professional performance.
For the purposes of this talk, I'll not be too concerned with exact definitions
but will simply discuss codes and specifications which have a relatively wide
acceptance, including some mandatory legal implications, and which contain
provisions contributing to safety, reliability, and functional operability of a
component or engineering device. My experience has been largely in the
fields of nuclear components, relatively conventional pressure-containing
equipment, and corrosion-resisting chemical and radiochemical process equip-
ment, most of which use relatively ductile materials as compared with some of
the high strength-to-weight ratio, sometimes low ductility materials of the
cerospace industry; however, there are still many similarities between the
materials characterstics and applications problems in these areas of tech-
nology, particularly in the matter of developing precise, practical and in-
telligible codes (specifications) and acceptance criteria for welds and weldments.
The ASME Boiler and Pressure Vessel Code is one of the oldest, most successful,
widely recognized and accepted documents of this kind; much of this discussion
will be based on the requirements of this ASME Code; and extension of these
requirements to other codes and specifications.
The first ASME Boiler Code appeared just over 50 years ago. It resulted from
....
ber.wo..
.
9-28-66
- 3
:
two sets of related circumstances:-first, in the period around the turn of the
century there was a very high and rapidly increasing f.equency and severity
of boiler explosions and resultant fatalities, which ran into several hundred
each year in the U.S., as boilers were built and operated at higher and
higher iemperatures and pressures; second, as a result of these many serious
explosions, many states and municipalities developed their own rules for
boiler construction, and soon there were so many different sets of rules that
.
boiler manufacturers, users, and insurors were faced with a completely
**
chaotic jumble and confusion of rules. The ASME was asked to form a
committee to bring some semblance of order into this confusion; this re-
sulted in the appearance of the first ASME. Boiler Code in 1914. In the en-
suing years, this Code was widely adopted as a legal as well as voluntary
standard for the construction of new boilers; and these boilers were operated
within the temperature and pressure limitations imposed by the design rules;
as a consequence, the incidence of failures and fatalities was drastically
reduced. Failures have since continued at a low level even though the Code
has been obliged to keup pace with revolutionary engineering developments
in steam generation, pressure containment, size of components, and environ-
mental conditions ranging from cryogenic temperatures to some of the very high
temperature, high pressure applications of the chemical and petrochemical
Lun.
CSVU
industries, and to the nuclear field.
Before I get into the details of Inspection reouirements for welds in the
Codes, I should point out that they have not been free from criticism-some
perhaps merited, and some based on the fact thut while the Codes present
technical requirements for adequats construction, they do not attempt to
make detalles mandatory provisions for administrative enforcement of their
technical requirements-the obligation for compliance with these Codes rests
with the legal jurisdictions who adopt the Code, designer, and the fabricator
and their professional responsibility for producing a system meeting Code
requirements and specifications, and the purchaser who has the obligation of
obtaining technically competent design and fabrication personnel and organi-
zations to provide a system meeting functional and safety requirements.
Those of you who are involved in the development and application of more
elaborate techniques of quality control and nondestructive testing may find
it difficult to accept the inspection and nondestructive examination require-
ments of the conventional codes and even some of the more elaborate and
restrictive codes and specifications. Technical and scientific people in the
nondestructive testing field have generated many highly and effective means
of locating and identifying defects in welds and welded construction; people
in the fields o: fracture mechanics and weld flaw evaluation can give us
.
.
.
9-28-66
- 5
-
approximations of the types, dimensions and orientations of discontinuieus
which can be tolerated in a structure subjected to stated sets of service
conditions and environment; people in Code-writing bodies are then obliged
to translate this information into practical rules which can be uniformly
interpreted and applied to a broad spectrum of service conditions by people
with different talents and capabilities in a competitive, cost-conscious fab-
ricating industry; this perhaps explains some of the things for which Codes
are sometimes criticized.
But to get back to the specific subject of code requirements and acceptance
standards for welds and weldments, let's start with the ASME Boiler Code.
The Sirst edition in 1914 covered riveted boilers. A few years later the Code
was extended to unfired pressure vessels of forged or riveted construction,
but rules for fusion welding of strength joints were not incorporated into the
Code until around 1928-31, some 15 years later. At that time and even today,
visual examination was the principal method for the inspection of components-
more particularly, pressure-containing "Code" wolds; it was used along with
a pressure (usually hydrostatic) cest or leak-test, destructive examination of .
weld test plates, and a hammer test-the hammer test requirement has been
largely discontinued today. Radiography was also introduced, not much later,
for welds in critical or "lethal" service, but, even in such services, was
9-28-66
- 6-
required only for longitudinal and circumferential main seam. butt welds;
the nuclear vesse). requirements, beginning with the Code Cases of the late
50's and continuing in Section III, Nuclear Vessels, requires radiography of
all pressure-containing full penetration welcs. Ultrasonic, magnetic particle,
and liquid penetrant methods of examination of welds have also been lacor-
porated into codes to varying extents, the setting of suitable acceptance
standards for these techniques has been difficult and even controversial.
The requirements for visual examination are quite similar in the ASME and
other codes.-most of them call for evaluation on the basis of qualitative de-
scriptions of acceptable weld conditions; quantitative restrictions are applied
in some areas. Among the visual requirements are the examination of weld
joint preparations to ensure proper fitting, alignment, positioning, joint
almensions, soundness and cleanliness of surfaces to be joined, the use of
qualified procedures and welders or welding operators, adequate protection
of the work from unfavorable weather conditions, and proper storage and
moisture control of welding materials. Strictly speaking, some of these are
a type of surveillance or process control carried out before welding. Codes
require or imply that these things should be done, but don't necessarily say
how, when or by whom
these detalls are frequently included in the welding
procedure rather than the code.
.
9-28-66
-
7.
The next requirement le surveillance monitoring of the actual welding process-
control of volts, amperes, arc length, travel speed, deposition rate, and ob-
servation of the arc and weld puddle. These details are generally included
in the welding procedure; again the codes don't say how or by whom the
weiding conditions shall be monitored. While this may sometimes be done
by an unsupervised welder or limited supervision, one of the most important
requirements for quality welds is that these welding conditions be frequently
and competently monitored and controlled by highly qualified Inspectors or
welding engineers, or by elaborately automated tape-or computer-controlled .
systems. Such control requirements may be administrative rather than
technical-but a significant shortcoming of some Codes is the lack of anything
like MIL-Q-9858, "Quality Control System Requirements"—this does not spell
out how surveillance control shall be accomplisi.ed, but does require that
detailed control and inspection procedures be established and maintained.
The most effective inspection system is the one which prevents the introduction
of a defect into a weld-rather than the one which finds the defect after the
fact, although the final check is ordinarily a necessary one. A weld originally
free from defects is far better than one which is repaired free of defects.
Returning to specific visual examination requirements for completed welds,
the ASME Code and others require that finished welds have complete joint
penetration, freedom from undercut (some codes are more realistic in this
requirement), no overlap, and no abrupt ridges or valleys; they also require
hne.......
-
.
F"
je
9-28-66
bil
-8.
that weld suríaces be reasonably smooth and either flush with the adjoining
surfaces or have a limited amount of weld reinforcement merging smoothly
into the base metal surface. The interpretation of these requirements is
generally a matter of interpretation and agreement (sometimes argument);
it is most difficult to state these requirements by description in writing, a
far better way is to use actual "Workmanship Standards" as indicated in
the sketch, Fig. 1, taken from the American Welding Society's Arc Welding
Inspection Manual and the AWS Welding Handvook. Some visual inspection
requirements are stated in measurable dimersional tolerances-alignment,
extent of reinforcement, dimensions of joint preparation, and, sometimes,
length and depth of undercut and incomplete penetration.
The ASME Code requirements generally call for hydrostatic testing at 1-1/2
times (this varies) the design or anticipated working pressure, or for
pneumatic testing at a somewhat lower overpressure. The generally stated
purposes of the overpressure test are for strength verification, leak testing,
and, sometimes, stress analysis purposes. To avoid brittle fracture in the
overpressure test, it is usually required that ferritic moterials in nuclear
and other vessels be pressure-tested above the brittle fracture transition
temperature, variously stated as NDT + some stated number (30 to 120,
commonly 60) of degrees F. In addition, there is a growing conviction among
some pressure vessel people that an overpressure test in the ductile
-
-
- -
- ----
-
-
---
.
.
.
-
-
9-28-66
- 9-
temperature range accomplishes "mechanical stress relief", which, together
with thermal stress relief, is credited with having permitted many vessels,
some with sizeable flaws, to querate without failure at temperatures well
within the brittle fracture susceptible temperature range. There is further
bellef, particularly in the United Kingdom nuclear vessel program, that this
overpressurization should be done somewhat above the highest anticipated
operating temperaturen-rather than just EDT + 60°F-to establish that there
are no defects in the vessel larger than a size-which can be calculated from
the test conditions and materials properties-safely less than the "critical
crack size" required for catastropic rupture by brittle fractur or ductile tear.
Hammer testing of a vessel during the hydrostatic test was once a require-
ment of codes and is still performed to some extent. However, it has been
largely discontinued because of the inherent danger to personnel performing
the test and some question regarding its actual validity as a test-(including
the difficulty of getting a "calibrated" hammer!) However, some modification
of the hammer test might well be considered for welds in materials like
titaniun which can be embrittled by inadequate control of the welding pra ess;
without being readily detectable by the usual nondestructive tests (although
portable hardness testing can be used to do this).
Lith
.
..
.
...
9-28-66
10 -
The ability of sample or test welds to pass certain visual, destructive and
nondestructive tests can be considered a weld acceptance test of sorts.
Several types of test plates are required in codes. The ASME Code requires
that each welding procedure be qualified by the preparation of a sanple weld --
in plate or pipe--for each stated class of base material, weld metal and
thickness range. Samples taken from the test plates are subjected to
specified iests-largely destructive to establish that the procedure is
capable of producing sound welded joints meeting certain visual and mech-
anical requirements--principally tensile strength and tensile and bend ductility-
stated in the Code. Also, each welder welding to a given procedure is similarly
required to establish his ability to make sound welds to the procedure-sample
coupons taken from his test welds are also destructively tested
primarily for
bend ductility. Once a procedure or a welder is qualified, if he continues
welding to that procedu.'e, further requalification is not normally required
unless certain "essential variables" are changed. ".
Qualification requirements are not always the same as the acceptar.ce standards
applied to actual production welds, and it is possible to qualify on the basis of
weld test samples which may not themselves meet the nondestructive test
acceptance standards required of production welds made by the same procedure
or welder. Similarly, qualification of a procedure may not give positive
9-28:66
- 11 -
in
...
assurance that all heats of materials la all specifications falling within the
classification grouping can actually be welded by that procedure to the accept-
...
ance standards of the code.
...
...
Where the intended service is sufficiently severe or critical to warrant,
fabricators or purchasers often place more restrictions on materials, thick-
nesses and joint designs than those required for standard code qualifications.
Some codes require added "vessel test plates" which require that samples of
the same materials to be used in the actual vessel be welded concurrently with
the vessel welds, subjected to the same heat treatment history as the vessel
welds, and required to meet the same nondestructive tests as the vessel
welds and the same destructive tests as the procedure qualification test welds...
In some cases, certain impact test requirements on the vessel test plates serve
as acceptance tests for the actual vessel welds.
Visual examination, pressure and leak testing, and the qualification of weld
procedures, welders and vessel test plates, are important Code acceptance
requirements; however, when we talk about acceptance standards, we generally
think of nondestructive testing, particularly radiography. Radiography entered
the Codes about the same time is fusion welding, and-excepting visual
examination-it is the best known and most widely accepted nondestructive
test method. The capabilities and proposed acceptance standards of other
L
.
NDT methods-eg., ultrasonic-are too frequently evaluated on the basis of
Reli..
Arte
*
*
:
...........
,
.
9-28-186
- 12 -
com
o
their ability to reproduce the results of radiographic examination, rather than
...
on direct comparison with the size, geometry, orientation and nature of the
weld defects themselves. (Perhaps if ultrasonic testind had come first,
******
people would now be trying to make radiographic images correlate with the
amplitude rejection levels on a cathode ray tube!) Actually, a principal
merit of ultrasonic testing, as well as other NDT methods, is the ability to
show things that radiography doesn't.
The following quotation from an unpublished manuscript appears pertinent:
"Certain features and limitations apply to practically all nondestructive test
methods. A material flaw is generally presented as a signal or indication
which may approximate the magnitude of the discontinuity, but seldom gives
an exact representation of its nature, geometry and dimensions. No one test
can find all defects; two or more complementary tests are frequently used to
detect a broader range of material imperfections; and, on occasion, indications
observed by one method are further explored and evaluated by using the second
technique. While reference acceptance standards of varying suitability may
be established, most tests require experience and judgment on the part of the
operator : to interpret the results in terms of suitability of a product." It is well
to recognize-although it's frequently overlooked-that radiography does not
produce dimensionally exact photographic reproduction of a weld flaw-the
image on the film is a signal that requires competent interpretation as does the
9-128=1676
- 13 a
signal from the other NDT techniques.
Considering Code acceptance standards for radiography, there are certain
restrictions on weld surface irregularities and film density, radiographs are
required to have a "sensitivity" capable of showing a "penetrameter"-
actually an
image quality indicator, not a defect simulator. The ASME
Code-required sensitivity is usually considered to be 2%; but the equivalent
sensitivity approaches 5 or 10% in thicknesses less than 1/4 inch, it is
about 2% in thicknesses from 3/4 to about 2 inches, and is about 1% for
thicknesses over 4 inches. With selected equipment, fine grain film, im-
proved techniques and competent radiographers, it is not difficult to achieve
a sensitivity as low as 1/2%, particularly in thicker sections. The achieved
S
sensitivity is in effect an acceptance standará, since margiral sensitivity
places a practical limit on the size and quantity of tight cracks and fine
porosity and inclusions which can be detected and rejected. ASME Code
rejectability standards also include any type of crack or zone of incomplete
fusion or penetration; elongated or intermittently aligned slag or inclusions
within certain measured limits of length and spacing; and the size, quantity
and distribution of porosity, frequently as represented by dots on a printed
chart. Evaluation of these standards should consider the ability of the
radiographic techniques to present an image which does represent the flaws,
and the significance of the flaws or combination of flaws detected or not
detected, on the ability of a welded joint to perform its designed function in
.
9.-28-66
Bl:
- 14 -
-
HIM
.
.
the service environment. These should be considered in establishing the
original acceptance standards, although it is difficult to establish standards
which are so precise that they do not warrant competent responsible technical
review in marginal or questionable situations.
Considering the rejectable defects separately, cracks, incomplete fusion
and incomplete penetration are generally the most serious--yet they are normally
detectable by radiography only wher: in line (plus or minus a few degrees) with
the radiation beam. This serious limitation to the radiographic presentation of
cracks and planar: defects is partly (but not very convincingly) compensated
by a radiographic setup which will show the cracks which are oriented in the
directions most likely to result in weld failures; however, if location of cracks
is a critical requirement, one is obliged to use some supplementary method -
pernaps ultra sonic testing or perhaps a mechanized and electronically instru-
mented radiographic setup which will provide multidirectional scanning.
Elongated slag inclusions are limited to 1/4 to 3/4 of an inch, and aligned in-
clusions have certain spacing limitations, which vary with section thickness.
These requirements are readily obtainable and acceptable in most Code appli-
cations using ductile materials-but I doubt if they'd be acceptable for high
strength, low ductility weldments in aerospace applications. Interpretation
of elongated and aligned inclusions requires a determination that some images
are 3-dimensional, blunt-ended slag stringers rather than representations of
continuous but intermittently oriented two-dimensional cracks or perhaps
Iwona
9-%8-66
- 15 -
interdendritic segregates. The adequacy of the Code requirements depends
heavily on the ability of the people interpreting the film.
The statement of rejectable dimensions of radiographic images, such as of
porosity, in thousandths of an inch, is not very practical, since the geometry
associated with radiography does not permit the statement of actual flaw
sizes to such accuracy. Several Codes have porosity charts showing the
permissible size, quantity and distribution of porosity in a given length, area
or volume of weld metal. The ASME Code limits total area of porosity to 0.060
x thickness of weld. in sq.in, in any 6 in. length of weld, with limits of 20%
of thickness on single spots of porosity, and added limitations on distances
between indications. Much of the "fine" porosity indicated on the charts is not
visible on film of marginal sensitivity-fortunately in ASME Code applications
this is seldom harmful unless indicative of something else such as the presence
of aluminum oxide inclusions in the weld, resulting from the aluminum-water
reaction. Another point-without establishing actual "acceptability standards",
the very fact or expectation that welds will be radiographed can have a psy-
chological influence in improving the quality of welding. For that matter, the
amount of inspection of any type which is to be applied influences the extent
of process control and thus contributes to weld quality.
Ultrasonic examination is an important NDT technique, particularly for
determining the presence and dimensions of cracks, even though it is not suited
1-28-66
- 16 -
to all welds and materials-particularly in the case of coarse-grained weld
deposits which cannot be refined by heat treatment or mechanical working. '
At present, the ASME Code only requires ultra sonic examination of electro-
slag welds and welds in certain nozzle considerations; in both cases it is
used in addition to rather than in place of radiography, but it is only a
matter of time (and effort) until ultrasonic examination will be used much
more widely in boilers, pressure vessels and piping in conventional and nucleark
applications. The introduction of ultrasonic procedures and acceptance
standards is complicated by the necessity for detailing procedures where
operator manipulative and interpretive skills are such important factors.
Other nondestructive tests, such as magnetic particle and penetrant exami-
nation are used particularly in the examination of surfaces of plates and
forgings, the cut edges of weld joint preparations, and the surfaces of
finished weldments. While it would be preferable to state rejectability
standards for NDT methods in terms of weld flaw dimensions, it is more con-
venient and practical to state them in terms of the dimensions of the NDT
signals whose correlation with actual defect sizes is often difficult and inexact.
Penetrant and magnetic particle techniques are sometimes considered simply
magnifying lenses, microscopes, and intra scopes. Some Codes require the
use of "metal identification" equipment to verify the materials used.
9-23-66.
- 17 -
The amount of inspection and the acceptance standards are sometimes estab-
lished on the basis of "classes" or "levels of quality", which in turn depend
on the requirements of the service and environment. Several years ago a
Qualification Procedure Committee of the American Welding Society suggested
some fifteen
classes of welds ranging from those used in a garbage can to
what today. would be used in space vehicles or nuclear work; this classifi-
cation was not formally adopted, but it indicates the number of possible
classes. An article in the Welding Journal of April 1965 by Ronald Clough
of the Canadian Welding Bureau suggests some 8 classes of welds. Foreign
codes frequently provide for 3 or 4 classes, and similar recommendations
have been made in this country; even the different sections of the ASME Code
provide a classification to some extent.
.
.
In preparing this talk I concerned myself largely with the ASME Boiler and Pressure
Vessel Code, since its provisions are well known and its problems and methods
are often common or similar to those of other codes. It was suggested that I
might compare some of the other Codes and specifications-but I should point out
that the March 1966 issue of Materials Evaluation contained a 6 page tabulation
of commonly used specifications and standards for nondestructive testing. One
specification suggested, with which you are familiar, is Radiographic Standard
*
NAS 1524. I would not have the presumption to criticize it in detail without
T
"
2
actual experience in its use; however, it appears to have many problems which
.
.
..
face other Codes, I note the calculation of sums of areas of porosity carried
.
7
9-20-66
. 18 -
out to 109 sq.in. I recognize that this is not an acceptance limit, but I
doubt if any regularly used NDT methods can provide this degree of accuracy.
Another document, NAVSHIPS 250-1500-1, has been mentioned several times it
i
this Symposium. Frankly, I think it an excellent document; many-but not
necessarily all-of its provisions should be considered by other codes when
their nondestructive requirements are critically reviewed. One of the best
descriptive statements of acceptability standards for radiography or other
method of examination is contained in the American Welding Society's
D2.0-66, "Specifications for Welded Highway and Railway Bridges". This
contains charts indicating the acceptability of size and spacing of porosity
and other indications in relation to weld thickness..
Other useful devices are reference radiographs such as those of the International
Institute of Welding and the American Society for Testing and Materials. They
represent actual radiographs of weld discontinuities and have been used by
other organizations as actual acceptance standards, even though the IIW and
ASTM specifically disclaim any such intent.
In ciosing, the best conclusions I can offer are restatements of a few recommendations
and suggestions. One is to use "Workmanship Standards" for comparison pur-
poses in establishing visual inspection standards for welds. Another is to use
actual films and photos where practicable and feasible, rather than too many
words, hypothetical sketches and porosity charts. Another is to recognize that
9-28-66
- 19 -
most nondestructive techniques complement each other, consequently we would
do well to not attempt to substitute one for the other. Finally, it is essential
in writing codes and specifications, that every effort be made to write them
so they are practical, economical, understandable, and actually relate NDT
indications to the size, shape and orientation of the discontinuities.
i
*
TIM
*
is
Arnis
..
.
'
6.
*
**
P
1.
END



DATE FILMED
12/ 28 / 66






LT
iTV
*
1: