Décembre 1995: Boutique AFNOR Pour: Solios Carbone Client 8319200 Commande N-20100112-383554-TA Le 12/1/2010 16:25

NF E83-100-1
Décembre 1995
Ce document est à usage exclusif et non collectif des clients Normes en ligne.
Toute mise en réseau, reproduction et rediffusion, sous quelque forme que ce soit,
même partielle, sont strictement interdites.
This document is intended for the exclusive and non collective use of AFNOR Webshop
(Standards on line) customers. All network exploitation, reproduction and re-dissemination,
even partial, whatever the form (hardcopy or other media), is strictly prohibited.
Boutique AFNOR
Pour : SOLIOS CARBONE
Client 8319200
Commande N-20100112-383554-TA
le 12/1/2010 16:25
Boutique AFNOR pour : SOLIOS CARBONE le 12/1/2010 16:25
FE038722 ISSN 0335-3931
NF E 83-100-1
French standard December 1995
Classification index: E 83-100-1
ICS: 01.040.25; 25.160.10
Construction of fabricated assemblies

Welding processes
Part 1: General : Terminology — Weld quality classes —
Scope of weld inspections
F : Construction d’ensembles mécanosoudés — Techniques de soudage —
Partie 1 : Généralités : Terminologie — Classes de qualité de soudure —
Étendue des contrôles
D : Schweisskonstruktionen — Verfahren — Teil 1: Allgemeines : Terminologie —
Schweissnaht -Qualität — Prüfbereich
© AFNOR 1995 — All rights reserved
French standard approved

by decision of the Director General of AFNOR on November 20, 1995, taking effect
on December 20, 1995.
Replaces the approved standard with the same index dated December 1987.
Correspondence When this document was published, there were no international works dealing with
the same subject.
Analysis This document forms part of a series of normative documents concerning the
construction of fabricated assemblies.
This part defines especially:
— the terminology;
— welding classes and quality;
— the extent of inspections and the degree of aptitude of welders and operators.
This document gives in its annexes recommendations for fatigue analysis of welded
assemblies, as well as the examples of welding quality class determination, for
informational purposes.
Descriptors Technical International Thesaurus: welding, quality classes, inspection, welders

(personnel), qualification, fatigue tests, computation.
Modifications With respect to the previous edition, taking into account of the standards
NF EN 287-1, NF EN 288-1 and the following.
Corrections
Published and distributed by Association Française de Normalisation (AFNOR — French standard institute) — 11, rue Francis de Pressensé —
93571 La Plaine Saint-Denis Cedex — Tel.: + 33 (0)1 41 62 80 00 — Fax: + 33 (0)1 49 17 90 00 — www.afnor.org
© AFNOR 1995 AFNOR 1995 1st issue 1995-12-F

NF E 83-100-1 —2—
Contents
Page
Foreword ............................................................................................................................................................. 3
1 Scope ................................................................................................................................................. 3
2 Normative References ...................................................................................................................... 4
3 Definitions ......................................................................................................................................... 4
4 Welding quality class ..................................................................................................................... 11

4.1 Definition of welding quality class ..................................................................................................... 11
4.2 Risks encountered in case of the weld’s failure ................................................................................ 11
4.3 Types of operational stresses of welded joints ................................................................................. 12
4.4 Choice of welding quality classes ..................................................................................................... 13
5 Extent of the inspections ............................................................................................................... 14

5.1 General case ..................................................................................................................................... 14
5.2 Extent of the supplementary inspections .......................................................................................... 14
6 Bibliography .................................................................................................................................... 15
Annex A (informative) Recommendations for the fatigue analysis of welded assemblies .................. 16
A.1 General Remarks .............................................................................................................................. 16
A.2 Fatigue tests on welded assemblies ................................................................................................. 18
A.3 Calculation rules ................................................................................................................................ 22
A.4 Example of application ...................................................................................................................... 34
Annex B (informative) Examples of welding quality class determination ............................................... 37

—3— NF E 83-100-1
Foreword
1) This document forms the first part of a series of normative documents grouped under the same index E 83-100
concerning the welding techniques relating to the construction of fabricated assemblies, of which the outline is the
following:
— Part 1: General remarks: terminology, weld quality classes, extent of weld inspections
— Part 2: Materials — Design
— Part 3: Guide for the choice of materials and for constructive provisions
— Part 4: Manufacturing — Inspection
— Part 5: Qualification of a welding operational procedure
This collection of normative documents corresponds to the needs expressed by the different concerned
professions in order to facilitate the manufacturer-client relationships. In particular, it allows one to:
— unify the language in this domain;
— unify the welding specifications required by the different order-givers;
— define for each welded joint, a level of reliability corresponding to the use requirements;
— serve as reference at the implementation of provisions relating to welding within the framework of quality
assurance.
2) While waiting for elasticity criteria to be brought into line with national and European standards and taking into
account the fact that the symbol Re has no meaning in European standards, the following measures are adopted
in this document on the subject of the elastic limit:
— the elastic limit to take into account is the upper flow limit ReH (generally specified in the EN standards), or, in
its absence, the conventional elastic limit at 0.2% (Rp0.2), or the extension limit at 0.5% (Rt0.5). In case of
dispute, the conventional elastic limit at 0.2% (Rp0.2) must be determined;
— the limit specified in this document is the upper flow limit (ReH); if one does not have guaranteed ReH values,
one can use the guaranteed values of Rp0.2 or Rt0.5, all the while maintaining the specified limits.
1 Scope
This document defines, within the framework of construction and repair of fabricated assemblies relating to its
scope of application:
— the terminology of the terms most commonly used;
— welding quality classes;
— the extent of inspections and the degree of aptitude of welders and operators according to welding quality
classes.
It gives in its annexes for informational purposes the recommendations for fatigue analysis of welded assemblies,
as well as examples of welding quality class determination.
This document applies to the construction of fabricated assemblies in steel relating to mechanical industries, such
as, for example, those concerning mechanical parts, machine frames, mining materials, agricultural machinery,
handling gear, etc., with the exception of those that are subject to special standards or regulations, especially
pressure devices, hoist devices and certain handling devices.
It does not apply to metallic construction (fixed or mobile framing and metallic structures, included in a building
operation, civil engineering, public works, facilities and equipment).
NF E 83-100-1 —4—
2 Normative References
This document incorporates by dated or undated reference, provisions from other publications. The normative
references are cited at the appropriate places in the text, and the publications are listed hereafter. For dated
references, subsequent amendments to or revisions of any of these publications apply to this document only when
they have been incorporated into it by amendment or revision. For undated references, the last edition of the
publication referred to applies.
NF EN 287-1, Qualification test of welders — Fusion welding — Part 1: Steels (classification index: A 88-110-1).
NF EN 288-1, Specification and approval of welding procedures for metallic materials — Part 1: General rules for
fusion welding (classification index: A 89-010-1).
NF A 03-400, Iron and steel — General principles of the fatigue test.
NF E 83-100-4, Construction of fabricated assemblies — Welding processes — Part 4: Manufacture Inspection.
3 Definitions
For the purposes of this document, the following terms and definitions apply.
3.1
buyer
The buyer is the physical or legal person who buys from the manufacturer a fabricated assembly, either on his own
behalf or on the behalf of a third party.
It is up to the buyer to indicate, under his own responsibility, the data necessary for the requested construction
and, if the need arises, special requirements in addition to those in this document.
3.2
welded assembly
A welded assembly is a group of elements assembled by using one or several welds.
3.3
welding quality class
A welding quality class establishes the conditions necessary and sufficient to be required at design and to be
respected at manufacturing in order to obtain an assembly capable to support the stresses from use and those
that can result from manufacturing without damage.
3.4
designer
The designer defines, on behalf of the manufacturer or the purchaser, the product, its conditions of use, its
characteristics, and drafts the technical specifications accompanying the order.
The designer may work for the manufacturer’s or the purchaser’s own operation.
3.5
manufacturer
The manufacturer is the physical or legal person who assumes the responsibility for the fabricated construction,
in accordance to the specifications provided by the purchaser.
The manufacturer may entrust the operations or work to subcontractors but maintains complete liability of the
operations or work thus subcontracted.
3.6
fabricated construction
Group that is part of a fabricated construction (machines, installations, etc.) constituted of one or several welded
assemblies.
—5— NF E 83-100-1
3.7
supplementary inspection
A supplementary inspection is an inspection intended to assess the quality of the weld on both sides of an anomaly
outside of the detected tolerance during an inspection.
3.8
inspector
Physical person responsible for performing the inspection operations.
This person may answer to the manufacturer, the purchaser, or to a specialized outside organization, following the
agreement in the order.
3.9
test plate
Representative specimen welded at the same time and with the same parameters as the welded joint to be
assessed. This plate may be adjoining or not to that of the welded joint, or be sampled directly on the part, in the
joint itself.
3.10
thinning by heat
Before welding, the action of bringing the parts to a temperature less than 100°C in the interest either of drying
them, or of bringing them to a temperature sufficient for welding in the case of parts that are too cold.
3.11
welder’s degree of aptitude (see standard NF E 83-100-4)
Qualification level obtained by a welder at a given qualification test.
3.12
stress relieving (relaxation treatment)
Thermal or mechanical treatment intended to diminish the internal stress of the weld.
3.13
extent of non-destructive testing
The extent of non-destructive testing is, for a welded joint, the weld length proportion to be inspected.
3.14
weld throat thickness
3.14.1
nominal throat thickness
The nominal throat thickness "a" is the thickness of the weld (or reference throat thickness) used for
the calculation.
The figures of table 1 (see the end of this clause) define the nominal throat thicknesses of the most common
welded assemblies for welds with or without chamfer.
3.14.2
actual throat thickness
The actual (or real) throat thickness "au" is the effective throat thickness obtained after welding.
3.15
gouging
Operation intended to eliminate irregularities, defects, possible slag deposits to leave only the sound metal before
restart, on a weld bead.
3.16
inspector
Physical person entrusted by the purchaser with supervising manufacturing and/or inspection operations.
NF E 83-100-1 —6—
3.17
joint
Space between two or several elements to be assembled by welding whose edges have been prepared for this
purpose (see figure 1).
In the case where the edges are straight, the joint volume may be null.
Primary metal
α Angle of chamfer
g Space or play
s Lock or flat spot
Figure 1
3.18
welded joint or weld
The term welded joint, or weld, indicates the joint after making the weld (see figure 2).
Consumable metal
Molten metal Welded joint
Figure 2
3.19
welding operational procedure (see standard NF EN 288-1).
3.20
welding operator (see standard NF EN 287-1).
3.21
preheating
Before welding, the action of bringing parts to be assembled to a determined temperature and maintaining the
parts at this temperature for the entire duration of the welding.
3.22
post-heating
After welding, the action of maintaining the assembled parts at a determined temperature and for a given time,
and cooling them.
3.23
welding procedure (see standard NF EN 288-1).
3.24
welding program or descriptive of a welding operational procedure (DWOP) (see standard NF EN 288-1).
—7— NF E 83-100-1
3.25
welder qualification (see standard NF E 83-100-4).
3.26
root
Region of the first run furthest from the welder.
3.27
repair
Important operation applied to a fabricated construction before or after implementation, in the interest of making
it conform to the previously defined quality criteria.
3.28
resumption (not to be confused with touch-up or sealing run)
Continuing a weld after a stop.
3.29
sealing run
Operation performed to the back side of a weld for full penetration in the interest of improving its finish and/or
resistance.
3.30
touch-up
Local operation performed in the interest of eliminating minor defects detected in the welded assemblies, before
implementation.
3.31
inspection severity
Inspection requirements determined according to the welding quality classes.
3.32
welder (see standard NF EN 287-1).
3.33
weld (see welded joint).
3.34
partial penetration weld
Welding concerning only one part of the thickness of at least one of the assembled elements; the penetration may
be normal or strong, according to the depth of the molten zone in the primary metal.
3.35
guaranteed penetration weld
Welding performed by one (or several) procedure(s), in specified conditions, assuring the sure connection of the
assembled elements over a definite depth.
3.36
full penetration weld
Welding performed by one (or several) procedure(s), in specified conditions, assuring the sure connection of the
assembled elements over the thickness of at least one of the assembled elements.
3.37
welding with natural chamfer
Welding performed in a joint whose chamfer is constituted by the rounding or the angle of the parts to be
assembled.
3.38
support
Device (metallic strip, fibreglass, ceramic, flux, etc.) supporting the metal in fusion during welding.
NF E 83-100-1 —8—
Table 1: Nominal throat thicknesses of the most common welded assemblies

for welds with or without chamfer
Designation Schematic representation
1 — partial penetration
(normal penetration) angle weld
without chamfer
with chamfer
(strong penetration) angle weld
without chamfer
(strong penetration) angle weld
with chamfer
(to be continued)
—9— NF E 83-100-1

for welds with or without chamfer (continuation)
5 — full penetration angle weld

with or without chamfer
6 — full penetration angle weld

with chamfer
7 — strong inclination angle

weld
— open angle:
strong penetration weld
— closed angle:
normal penetration weld
(to be continued)
NF E 83-100-1 — 10 —

8 — welding with natural

chamfer
with chamfer
(strong penetration) butt weld
with chamfer
(to be continued)
— 11 — NF E 83-100-1

(end)
11 — full penetration butt weld

with chamfer
4 Welding quality class

4.1 Definition of welding quality class
4.1.1 In use, the welded joints of a fabricated assembly is subjected to various stresses according to their
destination, their placement, etc.
Otherwise, the failure of a welded joint can have different consequences according to the risks encountered and
the conditions for replacing the defective element.
In order to optimise the manufacturing according to these conditions, this document establishes for welded joints,
by order of decreasing severity, three welding quality classes A, B and C (see 4.4), which are determined
accordingly:
— risks encountered in case of failure of the weld (see 4.2);
— operational stresses of the welded joint (see 4.3).
A same fabricated structure can therefore include different welding quality classes.
4.1.2 The quality class of each of the welds is defined at the time of design
It determines:
— the extent of the inspections;
— the severity of the inspections;
— the degree of aptitude of welders and operators;
— the welding program;
— the particular conditions to be respected during manufacturing and inspection.
4.2 Risks encountered in case of the weld’s failure

Table 2 establishes three grades of risks according to the consequences of a failure during operation:
• R1 = significant risks;
• R2 = risks of medium significance;
• R3 = low risks.
The grades of risks must be determined from one or several criteria from table 2 and must be adapted to the
environmental conditions in which the fabricated structure is located.
NF E 83-100-1 — 12 —
Table 2: Grades of risks according to the consequences of a failure in operation

and replacement conditions of the defective element
Condition
Necessity of the replacement of replacement
Defect in use Repair on site
or repair of the defective element or repair of the
Grades defective element
of risks Not Deferred
Requiring
requiring Impossible
immediate Immediate Short Long Difficult Easy Easy
immediate or difficult
stop duration duration
stop
R1 X X X X
R2 X X X X
R3 X X X X
4.3 Types of operational stresses of welded joints

Fabricated assemblies are classified according to stresses to which they are subjected. One differentiates:
— non-calculated assemblies;
— the calculated assemblies not subjected to a fatigue analysis;
— the calculated assemblies subjected to a fatigue analysis.
NOTE For the definitions and symbols used in this clause, refer to the standard NF A 03-400.
4.3.1 Calculated assemblies not subjected to a fatigue analysis (determinant static loads)
These assemblies are designed and calculated to resist static loads, which are:
— structures subjected to essentially static loads;
— structures and constructions for which, in addition to static loads, variable loads produce a stress spectrum for
which one can presuppose that fatigue shall not be a decisive factor.
One considers that a weld is not to be calculated and/or verified in fatigue if the number of cycles is equal or inferior
to the following value:
10
7 × 10
N ≤ --------------------
m
-
∆σ
where :
N is the number of cycles stipulated for the lifespan of the structure;
∆σ is the extent of stress applied in megapascals ( ∆σ = σmax – σmin calculated at the limit state of use);
m is equal to 3 (the most usual coefficient).
4.3.2 Calculated assemblies not subjected to a fatigue analysis (determinant cyclic loads)
4.3.2.1 These assemblies are designed and calculated to resist stress of periodically variable intensities
(estimated number of cycles), of which the maximum level remains inferior to the metal’s elastic limit.
4.3.2.2 The cyclic load is evaluated according to the ratio:
∆σ
----------
∆σ a
where
∆σ is the extent of the applied stresses:
∆σ a is the extent of the admissible stresses.
— 13 — NF E 83-100-1
This extent depends on the type of assembly (see table A.1 of the annex A which gives recommendations for
fatigue analysis of the welded assemblies).
4.3.2.3 In the case where the extent of stresses varies over the use of the structure, a verification must be
performed using the Miner rule (see annex A).
If the load is of a random character, a treatment for obtaining a load cumulative must be performed prior to the
application of this rule.
4.3.3 Non-calculated assemblies

These assemblies are designed and dimensioned by comparison according to experience.
These are assemblies for which it is not possible, or not necessary, to calculate the stresses and the variation of
the applied forces. They are classified in "stressed" or "low stressed" assemblies.
4.4 Choice of welding quality classes

Table 3 gives, according to the grades of risks defined in paragraph 4.2 and operational stresses of joints defined
in paragraph 4.3, the corresponding welding quality classes A, B an C.
When an assembly is subjected to determinant cyclic loads and elevated static loads, the most severe welding
quality class must be applied but when an assembly is subjected to severe cyclic loads, the angle weld profile
determines its performance in use. This superior quality accessible by a qualified welder is defined in paragraph
6.4 of the standard NF E 83-100-4.
Annex B gives examples of welding quality class determination.
Table 3: Choice of welding quality
Welding quality classes

Operational stresses of welded joints according to grades of risks
R1 R2 R3
σc
0,8 ≤ ------ ≤ 1 A A B
σe
Calculated assemblies
not subjected σc
Static stresses 0,4 ≤ ------ ≤ 0,8 A B C
to a fatigue analysis 1) σe
(determinant static loads)
σ
-----c- < 0,4 B C C
σe
∆σ
0,7 ≤ ---------- ≤ 1 A A B
∆σ a
Calculated assemblies
subjected ∆σ
Cyclic stresses 0,3 ≤ ---------- ≤ 0,7 A B C
to a fatigue analysis 1) ∆σ a
(determinant cyclic loads)
∆σ
---------- < 0,3 B C C
∆σ a
Stressed A A B
Non-calculated assemblies
Low stressed A B C
σc = calculated factored stress.

σe = stress at the conventional elasticity limit.
∆σ = Extent of applied stresses.
∆σ a = Extent of admissible stresses at 2 × 106 cycles (see annex A).
1) The fatigue analysis does not exclude static testing; apply the most severe.
NF E 83-100-1 — 14 —
5 Extent of the inspections
5.1 General case

Table 4 establishes for each welding quality class:
— the extent of inspections of density and surface.
The data appearing in table 4 applies to a fabricated assembly.
In the case of mass-manufactured fabricated assemblies, the percentage of fabricated assemblies to be inspected
is defined by agreement between the designer and the manufacturer.
For lack of the ability to apply the inspections stipulated in table 4 in the correct conditions of execution and
interpretation, an inspection of welds must be performed during manufacturing according to the methods to be
defined in accordance with the designer.
Table 4
Extent of the inspections

Welding quality class
Visual Radiography or ultrasound Magnetoscopy or penetrant testing
A 100 > 50% 1) 2) 3) > 50% 1)
B 100 — > 10% 4)
C 100 — —
1) By agreement between the manufacturer and the purchaser, inspection by radiography or ultrasound may be
replaced with an inspection by magnetoscopy or by penetrant testing and vice versa.
2) For angle assemblies, the inspection by ultrasound is replaced by inspection by magnetoscopy or by
penetrant testing if the thickness is less than or equal to 10 mm.
3) The stress concentration points (nodes, crossings, weld extremities,...) are inspected at 100% over at least
50 mm on both sides of the concentration axis or of the edge of the welded joint.
4) Inspection by magnetoscopy or by penetrant testing is cancelled on steels whose elastic limit is less than
355 MPa, if the thickness "t" or the throat thickness "a" is less than or equal to 20 mm.
REMARKS: The inspections performed must be distributed as regularly as possible over the whole of each
relevant welded joint from the same welding quality class.
The inspected length must not be less than 400 mm.
Any welded joint of a length less than 400 mm is inspected over the total length.
For welded joints of a length greater than 400 mm, all the extremities must be inspected over a length at least
equal to 50 mm.
5.2 Extent of the supplementary inspections

In case of surpassing the limit of acceptance, the initial inspection is extended on both sides of the zone in question
over a minimum length of 400 mm.
During this complementary inspection, if the acceptance limit is once again surpassed, the inspection is then
extended to the total length of the assembly.
— 15 — NF E 83-100-1
6 Bibliography
NF E 52-109-1, Lifting and handling — Welded joints — Part 1: Fabrication.
NF E 52-109-2, Lifting and handling — Welded joints — Part 2: Weld quality classes — Extent of the non-
destructive testing.
NF P 22-470, Steel construction — Welded connections — Details and design of welds.
Document IIS/IIW-693/81, Recommendations for the calculation of steel welded constructions, subject to cyclic
stresses.
NOTE This document of the International Welding Institute should soon be revised.
A further revision of this document (annex A) is stipulated so that this remains coherent with the IWI document.
Terms and definitions used in welding and associated techniques (publications of Autogenous Welding and of the
Welding Institute).
NF E 83-100-1 — 16 —
Annex A
(informative)
Recommendations for the fatigue analysis of welded assemblies 1)
Init numérotation des tableaux d’annexe [A]!!!

Init numérotation des figures d’annexe [A]!!!
Init numérotation des équations d’annexe [A]!!!
This annex gives the rules relating to performance in fatigue of welded assemblies.
These rules based essentially on the tests can be used for the design and calculation of welded assemblies.
A.1 General Remarks

Any welded assembly is the seat of stress concentrations which occur at very different steps.
The first effect concerns the general welding melting rate (figure A.1) that one finds evoked by the inspection
specifications (elimination of excessive rounding for butt welding, recommendation of a flat angle or concave
weld). It is a question of an effect of macrogeometry (at the scale of the weld bead).
Strongly curved weld (to be avoided) Concave weld Convex weld (to be avoided)
Figure A.1
One then finds a microgeometry effect. It is a question in this case of very localized stress concentrations only
concerning the particular zones of a weld (figure A.2):
— connecting zone of a butt weld;
— toe and root of a angle weld.
Weld toe angle

Connection
Root
Figure A.2
It’s in these regions that what one usually qualifies as natural weld notches are located.
1) Text drafted from the document IIS/IIW-693/81.

— 17 — NF E 83-100-1
Finally, the third effect results in the geometry of the whole, i.e. the related arrangement of the layout of the parts
to be assembled (figure A.3).
Figure A.3
The most important effect regarding fatigue seems to be the effect of microgeometry because the fatigue cracks
are preferentially located in the regions where it occurs. (Figure A.4 gives examples of cracks for informational
purposes.)
Crack
Crack
Direction of the force
Crack
Crack
Crack
Figure A.4
In reality, the combination of three geometric effects condition the localization of the cracks, the microgeometric
effect playing a predominant role.
Even if, as indicated below, this observation cannot be used quantitatively, it is, however, a question of extremely
precious qualitative information.
NF E 83-100-1 — 18 —
A.2 Fatigue tests on welded assemblies

A.2.1 Test principle
The fatigue tests on welded assemblies resume the general principles of fatigue tests (see standard NF A 03-400).
As a general rule, the tests are conducted with non-random simple cycles. Only the side effects of geometric
imperfection bring a complication to the practical realisation of pure loading systems. On the other hand, the fact
of working on specimens taken from large-sized welded plates must cause an alteration of the field of residual
stresses and an almost complete disappearance of the effect of global geometry, although it is always useful to
confirm the results on small specimens through tests on parts or on real-size models.
A fatigue test on a welded specimen is to be performed in the following way (see figure A.5):
Figure A.5
It concerns a repeated tensile test (Rσ = σmin / σmax = 0), for which one imposes a force variation
∆F (∆F = Fmax – Fmin). By definition, the butt assembly is subjected to an extent of variation of nominal stress ∆σ
equal to ∆F / S.
This magnitude ∆σ is equal to σmax – σmin ; it is designated hereafter in the text by "Extent of stress".
At the end of a certain number of cycles, a crack appears in the connection zone. Initiation occurs in the zone
where the geometry is the most unfavourable and corresponds thus to the maximum notch effect; but, if the weld
is regular, one sees very often multiple initiation which produces a continuous shallow surface crack.
Continuing the test, one causes the propagation of the crack which generally occurs at depth, i.e. in the thickness
of the product or in the thickness of the weld (see figure A.4) continuing until the specimen breaks.
A.2.2 Choice of a representation of the results

Although on first thought, one can choose to represent the fatigue test results on the welded assemblies in any
representation system, it’s the Wöhler curve traced in a system of bilogarithmic coordinates (figure A.6) which is
the most frequently used.
— 19 — NF E 83-100-1
Figure A.6
It remains possible to choose, for the Y-coordinates, either ∆σ = σmax – σmin or σa, amplitude of the stress cycle.
It suffices to remember that σa = ∆σ / 2 and the choice of one or the other parameter becomes indifferent. In the
system of representation thus chosen, one makes a representative point of the test result at the coordinates
(∆σnominal, Ncrack). The graph is completed by showing all the results classically in the line of the Wöhler curve
(figure A.7).
Extent of stress
Figure A.7
NF E 83-100-1 — 20 —
A.2.3 Analysis of results

The fatigue test results on welded assemblies are affected by a dispersion. The mean plot of the general equation
N = C / ∆σm which is traced between the points, corresponds to a cracking probability — or survival — of 50 %.
This point is particularly important in the perspective of using the test results for dimensioning. By a statistical
analysis of the results, it is possible to determine a dispersion band of which the lower and upper borders
correspond respectively to cracking probabilities of 2.3% (or 97.7% of survival) and of 97.7% cracking
(2.3% of survival).
This dispersion band is also extended to the horizontal part of the Wöhler curve even if the idea of dispersion is,
in this domain, more difficult to appreciate assuming that it occurs vertically rather than horizontally and it must be
appreciated from deterministic results (cracking) or not deterministic (tests stopped before cracking).
A.2.4 Influence of residual stresses

Any welded assembly, as-welded, is the seat of residual stresses. These result in hindered contraction of the
molten metal during its cooling and of the restraint effect resulting from the mass of assembled parts and of the
heterogenous character of heating (and of cooling) created by the welding operation.
These residual stresses combine with stresses due to applied forces and it’s the resultant which determines the
performance in fatigue. It is easy to show that residual welding stresses attains the primary metal’s elastic limit
and this maximum level explains in reality why only the quantity ∆σ has an influence on the lifespan. Whereas, as
is the case for materials alone, one could expect an influence on the average stress or, put another way, an
influence on the Rσ relationship. Experimentation confirms this reasoning.
Figure A.8 constitutes an example of the result obtained for different values of Rσ; it shows that whatever the value
of Rσ, the endurance level is noticeably identical
Extent of stress
Undulating tension
0 Repeated tension
1 Symmetrical compression tension
α Repeated compression
Figure A.8
It is the presence of these residual tensile stresses that explains why fatigue cracks can be noticed on welded
assemblies subjected to compression cycles which often have the reputation of not being dangerous in fatigue.
— 21 — NF E 83-100-1
A.2.5 Influence of the nature of steel

In the majority of cases, cracking by fatigue initiates on an acute notch that constitutes the weld’s toe. In reality,
the existence of small very acute intrusions causes the duration of the initiation phase to be reduced to its most
simple expression since it transforms a crack-type defect into a "true" fatigue crack. The total lifespan is therefore
entirely and chiefly represented by the duration of the propagation phase. The law of propagation in fatigue being
not very dependent on the resistance of steels implemented, it appears normal that a fatigue crack needs
approximately the same time to propagate itself no matter what the constituent steel. An illustration of this
phenomenon is given in Figure A.9. By this fact, the fatigue resistance of an assembly depends, above anything
else, on its configuration.
Smooth specimen
∆σ for 106
Notched specimen
Welded specimen
Tensile force
Figure A.9
One notices in this figure, between the performance domain of the constituent material and the performance
domain of a welded assembly, the intermediary location of a specimen bearing a mechanical notch. This must
naturally be approximated to that which is known in other respects, especially in regards to the increase of
sensitivity to the notch effect when static resistance increases. This effect, amplified in the case of welded
assemblies by the maximum acuteness of notches, provides another route of explication as to the non-
dependency of performance in fatigue of assemblies with respect to the elastic limit.
This observation is also partially explicable by the increased unfavourable effect of residual stresses which
increases with the elevation of the elastic limit.
An immediate consequence of the preceding is the immobilisation of a reflex attitude when an assembly (or a
welded part) breaks from fatigue. This reflex attitude consists of not changing anything in the assembly's plan or
execution other than using a more resistant steel. In reality, the steel will be effectively more resistant statically
since it has greater static mechanical characteristics but its intrinsic aptitude in fatigue, although greater, will be
completely eliminated by the indicated effect.
The expected improvement will not be obtained; the result may even be contrary to the desired effect.
The improvement in fatigue of a welded joint can only occur through an increase in thickness (leading thus to a
global diminution of the stress level) or an improvement of the plan (through recourse to a better assembly).
In general, the use of steels with a high elastic limit only presents an important interest in the case when a finishing
treatment (grinding, remelting, blasting, hammer,…) is applied, after welding. The effectiveness of these
treatments is much greater if the steels used have a higher elastic limit.
NF E 83-100-1 — 22 —
A.3 Calculation rules

A.3.1 Principle of calculation rules for a range of constant stress
Let’s consider the welded assembly in figure A.5. The problem to be resolved may be posed either in terms of
dimensioning, or in terms of verification.
A.3.1.1 Dimensioning
The elements that the designer may use are:
— stresses (forces, resultant moments) to be transmitted by the assembly;
— the operational conditions of the construction.
The calculation must allow dimensions to be established (thickness, inertia) so that the assembly gives satisfaction
during the totality of the contractual lifespan.
The first work consists in calculating the number of cycles N to which the assembly must resist.
The second step is to determine the value of ∆σ for the assembly.
In consideration of which, we find:
∆F known
∆σ admissible = --------------------
-
S min
or:
∆F known
S min = -----------------------------
-
∆σ admissible
From this simple example, we therefore see that the calculation does not directly concern the welded assembly
(the welds themselves) but the constituent elements which must be sufficiently substantial so that, under the stress
transmitted the nominal stress level remains at a sufficiently low value in order for there to be no fatigue cracking
from the notches made by the welds.
Later will be covered:
— the determination of ∆σ admissible;
— the dimensioning of the welds themselves.
A.3.1.2 Verification
This time we have at our disposal the following elements:
— stresses to be transmitted by the assembly;
— operational conditions;
— dimensions of the assembly.
Despite being able to be conducted on the level of stress (verification that the real nominal stress level remains
inferior to the level of admissible nominal stress), the verification is generally made on the number of cycles.
According to the operational conditions and lifespan, one calculates a number of cycles to be performed and,
according to the dimensions and stresses, one calculates the ∆σnominal to which the assembly is subjected.
∆F known
∆σ nominal = --------------------
-
S min
According to ∆σnominal, one determines a number of admissible cycles and one verifies that:
Nreal < Nadmissible
— 23 — NF E 83-100-1
A.3.2 Calculated curve

The determination of ∆σadmissible from a number of determined cycles (dimensioning) or a number of admissible
cycles from a known ∆σ creates a calculated curve.
This calculated curve is determined from the average performance curve (50% of survival) by making use of the
safety margin by the designer.
The rules of calculation in general do not establish this safety margin, at the risk of taking on a regulatory aspect
or of replacing the designer who shall remains the one who chooses this security margin, welded constructions
generally being under the responsibility of the manufacturer.
It is appropriate however, to emphasize that this safety margin will intervene in the choice of an accepted cracking
probability (or, put another way, of a guaranteed survival probability).
This general principle (see: guaranteed elastic limit) results from the statistical nature of the test results.
An analysis of dispersion allows, as indicated above, to determine a dispersion band. This may be wider or
narrower, but it is naturally the lower border which will be applied. This border is obtained simply by shifting the
mean plot in the direction of an increase of the survival probability by a certain number of standard deviations
which globally characterize the dispersion of the results.
As indicated above, this shift results from the designer establishing the required survival probability, it being
understood that the number of standard deviations result from this.
A currently allowed value for the survival probability is 97.7% corresponding to a shift of two standard deviations
(figure A.10).
Mean distribution (50% of survival)
Shift of 2 standard deviations
Calculated distribution (97,7% of survival)
Figure A.10
The plot thus determined from the mean plot representing the mean distribution (figure A.10) serves indifferently
for dimensioning or verification because it is what allows one to go from the Nknown to ∆σadmissible (figure A.11a)
or from ∆σknown to Nadmissible (figure A.11b).
NF E 83-100-1 — 24 —
Mean distribution
Mean distribution
known
Nknown
Figure A.11
This clearly indicates the principle retained for the inclined part of the Wöhler curve.
In regards to the fatigue limit, the procedure is different.
One can make a "vertical" statistical analysis and establish a guaranteed operational probability calculated fatigue
limit Furthermore, as it concerns a clearly more delicate procedure, we prefer to take recourse to another
technique that consists in establishing the calculated fatigue limit at the level ∆σ reached by the calculated plot for
a number of determined cycles (5 × 106, 107, 2 × 108 for example) according to the fatigue case and especially in
case of an aggressive environment (especially corrosion) (figure A.12). This point is essential not only for
calculating welded assemblies for long life spans but also for the calculation of damage in variable or random
cycles.
Mean distribution
Mean fatigue limit

Calculated plot
Calculated fatigue limit
Nlimit
(break)
Figure A.12
— 25 — NF E 83-100-1
A.3.3 Classification of the assemblies

One notices that everything rests on experimental results because it is using them that one deduces the mean
distribution and then the calculated distribution.
Therefore, in order to establish a fatigue calculation rule, we must have at our disposal a collection of results as
complete as possible in its diversity (taking all the current fundamental assemblies in consideration) as well as by
its wealth (a maximum of results for a given assembly allowing a serious statistical treatment). Said work requires
time and collaboration between laboratories. From this collection of results, we proceed to statistical processing
and will, on this occasion, notice that assemblies of sometimes different natures, or, on the contrary of
neighbouring plans have a comparable performance, expressed globally by the resulting figures. One groups them
into the same class to which one single and same calculated curve is allocated.
The complexity of a calculated specification may be assessed according to the number of proposed classes.
The number of these classes is inclusive between 5 and 10, without however the possibility of making a quality
judgment on the calculation rules according to the number of classes.
A.3.4 Detail of the calculation

The following must be based on the calculation rules of the International Institute of Welding (document IIS/IIW-
693/81).
A.3.4.1 Field of Application

Rules have been established according to the available test results (see A.3.3); for this fact they cover current
welded structural steel non-alloyed or low alloyed constructions, created by an arc welding procedure.
Consequently, one must specify that all that concerns:
— strongly allied steels;
— welding procedures such as electron beam, friction, etc. ;
— non-ferrous materials;
is not covered by these rules, which means furthermore that the test results are not available. In particular,
assemblies in aluminium and aluminium alloys have been subject to a great number of tests, and the calculation
rules in this domain are currently in the process of being extended at the I.I.W, whereas the national rules have
already been published.
A.3.4.2 Scope of validity

Based on a conservative principle (as-welded assemblies), these rules are safely applicable for all the steels of
the type defined above of which the elastic limit is less than or equal to 700 MPa.
It is appropriate however to highlight that, even if high yield stress steel (HSLA) brings no particular advantages
in the most common cases, their use may be beneficial in two circumstances; the first when the static load proves
preponderant (elevate stress mean), in this case a double verification is necessary (see figure A.13), the second
when finishing techniques are implemented (see A.2.5).
NF E 83-100-1 — 26 —
Double verification
with:
(k = safety coefficient)
Time
Figure A.13
A.3.4.3 Field of use — Static load

The Wöhler plot, defined on figure A.14, establishes the limit from which an assembly must be dimensioned or
verified in fatigue (see paragraph 5.3 of this document).
If the operational regime of the assembly is found below this limit plot, there is no reason to dimension (or verify)
this assembly in fatigue; it may be calculated according to the rules of static calculation.
A limitation should also be established on the number of cycles.
Actually, the fatigue calculation rules (dimensioning and verification) are not applicable to the generalized plasticity
oligocyclic regime, i.e. for the numbers of cycles lower than 100,000.
Fatigue analysis zone
Figure A.14
A.3.4.4 Damage accumulation

In the case where the cycles have a random or variable character, direct dimensioning, figure A.11a, is not
possible and only the verification, figure A.11b, is usable for subsequent tests and approaches.
From the described or measured operational conditions, one determines a load cumulative with the help of an
appropriate counting method. One then applies the Miner cumulative law of damages leading thus to verifying that,
for the considered assembly:
ni
∑ ----N-i ≤ 1
— 27 — NF E 83-100-1
with:
ni = number of applied cycles under the extent ∆σi;
Ni = number of admissible cycles under the extent ∆σi only.
A.3.4.5 Applied Classification

By definition, the class of an assembly is the value of ∆σadmissible for 2 × 106 cycles at a cracking probability 2.3 %
(figure A.15).
Mean distribution
Class
Calculated distribution
at 97.7% of survival
Figure A.15
A.3.4.6 Table of notch cases

Table A.1 gives an assembly classification type (notch cases). It includes:
— an order number of the assembly type;
— a sketch of the assembly with the location of potential cracking;
— a description of the assembly with some additional specifications or restrictions that may concern a particular
welding procedure or the implementation of non-destructive testing in order to assure the quality of execution;
— the class of each assembly.
The use of this table is two-fold:
— first, it constitutes an aid in design, because, while the relative disposition of the elements is established, we
have the choice between several assemblies, one can select the best among them, which is that with the
highest class;
— second, it allows, after having selected an assembly and, consequently, having determined the class to which
it belongs, to select, on figure A.16, the calculated plot to which it corresponds.
It should be noted that this table includes two particular cases which concern respectively the edges of thermal
cut steel plates cleaned up by grinding (n. 27), and the molten metal of angle welds (n. 28).
For these, (n. 28), a verification of their dimensioning must be systematically performed when they transmit a
transversal force (types nos. 17 and 18).
NF E 83-100-1 — 28 —
Extent of stress — MPa
Lifespan, cycles
Figure A.16
— 29 — NF E 83-100-1
Table A.1: Classification of standard assemblies

In this table, a weld is called "transversal weld" when the force is applied perpendicular to the weld, and "longitudinal weld" when
the force is applied parallel.
Assembly diagram Description of the assembly Class

Type n.
indicating the fracturing mode and non-destructive testing (NDT) to be performed N/mm2
Machine flushed transversal full penetration butt

1 welding 125
NDT 100%
Transversal full penetration butt welding done in the

shop on a flat surface by any procedure excluding
2 100
welding done under flux
NDT
Transversal full penetration butt welding performed

3 in different conditions than weld n. 2 80
NDT
Transversal butt welding with permanent support

4 (extent of stress in the steel plate with exclusion 71
made for the support effect)
Full penetration longitudinal angle welding.

5 Automatic continuous welding 125
(extent of stress in the flange adjacent to the weld)
Partial penetration longitudinal angle welding.

6 Automatic continuous welding 112
Partial or full penetration longitudinal angle welding.

7 Manual continuous welding 100
(to be continued)
NF E 83-100-1 — 30 —
Table A.1: Classification of standard assemblies (continuation)


Type n.
Discontinuous longitudinal angle weld (extent of

8 80
stress in the flange at the ends of the welds)
Full penetration longitudinal weld, continuous angle

weld, or discontinuous angle weld with indentations
9 71
(extent of stress in the flange at the ends of the
welds)
Longitudinal angle weld of a gusset:

71
10 — length ≤ 150 mm 63
50
— length > 150 mm near the edge
11 Transversal angle weld of a gusset 80
12 Non-stressed pin connectors 50
(to be continued)
— 31 — NF E 83-100-1


Type n.
13 Welded gusset on the lip of a plate 80
Stiffener welded to the web plate of a girder

14 (principal extent of stress in the web plate at the end 80
of the stiffener)
Stiffener welded to the flange of the girder (extent of

15 80
stress in the flange, at the toe of the weld)
Cross assembly with full penetration transversal

16 angle weld on plate Z and alignment defect less than 71
15% of the plate thickness
Cross assembly with full penetration transversal

17 angle weld and alignment defect less than 15% of 63
the plate thickness (root cracking, see type n. 28)
Lap assembly by transversal angle weld

transmitting force. Cracking in the weld toe
18 (calculated stress on the base of a plate of a width 71
equal to the width of the butt straps) (cracking at the
root see type n. 28)
(to be continued)
NF E 83-100-1 — 32 —


Type n.
Lap assembly by transversal angle weld

19 50
transmitting force
Inserted flush-welded flange portion.

20 Plot or curve transition 112
NDT
Full penetration transversal butt weld between

elements of different widths or thicknesses.
Progressive transition according to indication (1/5):
21
— weld according to type n. 2 100
— weld according to type n. 3 80
NDT
Full penetration transversal butt welding.
22 Progressive transition according to indication (1/5) 112

flush weld:
NDT
Additional flange or butt strap on girder, welded

23 extremities (extent of stress in the flange, at the toe 50
of the weld)
(to be continued)
— 33 — NF E 83-100-1


Type n.
Additional flange or butt strap on girder, non-welded

24 extremities (extent of stress in the flange, at the 50
extremity of the weld)
Several flanges on girder, welded extremities

25 50
(extent of stress in the flange, at the toe of the weld)
Additional flange or butt strap wider than the flange,

26 non-welded extremities (extent of stress in the 50
flange, at the extremity of the weld)
Raw material with flame cut edges. Ground edges,

27 125
no cracks at inspection
Molten metal in the angle welds transmitting force

28 (extent of stress in the weld throat plan) (Cracking at 45
the root see types n. 17 and 18)
NF E 83-100-1 — 34 —
A.4 Example of application

This clause gives an example for determining the dimensions of a girder, of a welded flat on the flange of the girder,
and joint welds between the flat and the girder (figure A.17).
Figure A.17
The flat receives a variable force corresponding to ∆F = 60 kN.

This force determines in the central part of the girder in flexion, where the plat is found, a variation extent from the
flexion moment ∆M = 35,000 Nm. It is based on a lifespan of 106 cycles and a survival probability of 97.7%.
A.4.1 Analysis of the problem

Three crack types are possible (figure A.18).
Cracking in the girder (caused by the

notch that constitutes the flat toe of the
joint welds-tension flange).
Tension flange
Cracking in the flat (caused by the
notch that constitutes the flat toe of the
joint welds-flange).
Cracking in the welds from the root

(we suppose that angle welds are
Weld necessary — in order to be able to
calculate them — even if full
penetration welds are incontestably
better).
Figure A.18
— 35 — NF E 83-100-1
A.4.2 Girder dimensions

We do not find the corresponding type in the elementary assembly catalogue in table A.1. Therefore a similar
constructive provision must be found. For this, we must simply consider that the flange of a girder is the seat of a
longitudinal, tensile, or compression force. In the present case, the detail should be compared to type n. 11.
It is appropriate to note that the provision in type n. 10 which corresponds to orientating the flat longitudinally,
favourable for the transmission of force F to the web plate of the girder, which recovers it as shearing force, is less
favourable from a point of view of fatigue performance of the girder, as it corresponds to a lower class than that of
type n. 11.
Referring to fatigue A.16, we find for class 80 and 106 cycles, a ∆σ admissible of 100 MPa. We therefore get the
minimal inertia module:
I ∆M
---- = ------------------------------
V ∆σ admissible
or:
I 3
---- = 350 cm
V
Then either a rolled section, or a reconstituted section fulfilling this condition is chosen.
A.4.3 Dimensioning of the flat

The type of assembly does not appear in table A.1 but, it being a question of the assembly of a flat receiving a
force on another perpendicular flat, one refers to types n. 16 and n. 17. Only type n. 17 is appropriate because
type n. 16 corresponds to an assembly of full penetration beads. Therefore, partial penetration angle welds have
been chosen.
The type n. 17 assembly corresponds to class 63, which, for 106 cycles, gives a ∆σ admissible of 80 MPa on
figure A.16.
One gets:
∆F
S min = -----------------------------
-
∆σ admissible
or:
2
S min = 750 mm
a flat of 75 mm wide and 10 mm thick is chosen.

We find in paragraphs A.4.2 and A.4.3 that the dimensioning calculations of the constituent elements have been
performed taking into account the performance in fatigue of these elements, determined by the presence of welds.
A.4.4 Dimensioning of welds

One assumes two welds of a length l = 75 mm.
The detail of type 28 of class 45 specifies that it is appropriate to calculate the extent of stress in the weld throat
thickness plan. This is because, cracking is noticeably propagated in the throat thickness plan, its speed of
advancement, and consequently the lifespan of the assembly, is conditioned by the perpendicular tensile stress
in the throat thickness plan (otherwise equal due to symmetry to the shear stress in the throat thickness plan).
Referring to fatigue A.16, one finds for class 45 and 106 cycles, a ∆σ admissible of 57 MPa.
NF E 83-100-1 — 36 —
Moreover, calculating with respect to the throat thickness plan, we get:

∆F
∆σ nominal = ------------------------ ≤ ∆σ admissible
2 2⋅a⋅l
or:
∆F
a min = ---------------------------------------------------
2 2 ⋅ l ⋅ ∆σ admissible
We find, considering ∆σ admissible = 57 MPa, l = 75 mm et ∆F = 60 kN

a min ≠ 5 mm
or:
a = 6 mm
It must be noted, in passing, that the choice of a throat thickness value a = eflat/2 is limit.
This rule must be abandoned although it is in common use by design consultants. Overabundant in static stress,
it can be dangerous in the case of fatigue!
A.4.5 Comments
As has already been indicated, the arrangement of the flat transversal to the tension flange is most favourable in
regards to the fatigue performance of the girder; however, this arrangement is not recommended because it is
susceptible to provoke a secondary deformation effect of the flange (figure A.19).
Figure A.19 Figure A.20
In order to avoid this side effect, it is appropriate to add stiffeners which directly assure the resumption of the force
caused by the flat (figure A.20).
We notice that the addition of these stiffeners does not affect the fatigue performance of the girder. Effectively, this
type n. 15 assembly corresponds to class 80 which is identical to that of the flat associated on the tension flange.
— 37 — NF E 83-100-1
Annex B
(informative)
Examples of welding quality class determination
Init numérotation des tableaux d’annexe [B]!!!

Init numérotation des figures d’annexe [B]!!!
Init numérotation des équations d’annexe [B]!!!
B.1 Welding quality class determination of two welded joints of a fork lift truck apron
(figure B.1)
B.1.1 Determination of risk levels (paragraph 4.2)
— Weld 1: an operational failure causes the device to immediately stop.
— Conclusion: risk level R1.
— Weld 2: an operational failure does not cause the device to immediately stop: Repair on site is easy.
B.1.2 Determination of the type of operational stresses (paragraph 4.3)

In the two cases, the assemblies are not calculated, the stresses are estimated.
— Weld 1: partial penetration welded joint, stressed.
— Weld 2: partial penetration welded joint, low stressed.
B.1.3 Determination of welding quality class (paragraph 4.4)

— Weld 1: class A
— Weld 2: class C
Dimensions in millimetres
NF E 83-100-1
— 38 —
1 = weld 1
2 = weld 2
Figure B.1 : Fork lift truck apron

— 39 — NF E 83-100-1
B.2 Welding quality class determination of five welded joints of an 18 cylinder

internal combustion engine frame and 570 mm bore (figure B.2)
— Weld 1: principle weld between the crank (cylinder tie rod anchoring) and bearing partitions; work in tension.
Surfacing weld — Full penetration joint.
— Weld 2: bearing partition assembly — attachment flanges on pillar or engine bed. Full penetration joint.
— Weld 3: partition-flange connections (section F) and lateral bearing partitions. Full penetration joint.
— Weld 4: median plate assembly with crank and ridge. Full penetration joint.
— Weld 5: lower plate assembly of camshaft crankcases. Partial penetration joint (normal penetration).
B.2.1 Determination of risk levels (paragraph 4.2)

— Welds 1 to 3: an operational failure causes the immediate stop of the engine.
— Welds 4 and 5: an operational failure does not cause the immediate stop of the engine; repair on site is difficult.
B.2.2 Determination of the type of operational stresses (paragraph 4.3)

— Weld 1: the calculated assemblies subjected to a fatigue analysis.
Calculated extent of stresses: ∆σ = 49 MPa.
The assembly is compared to type n. 16, defined in annex A, of which the class of extent of stress at 2 × 106 is
∆σa = 71 MPa.
∆σ 49
---------- = ------ ≈ 0,7
∆σ a 71
— Welds 2 to 4: Non-calculated assemblies, stressed.

— Weld 5: Non-calculated assemblies, low stressed.
B.2.3 Determination of welding quality class (paragraph 4.4)

— Welds 1 to 4: class A.
— Weld 5: class B.
NF E 83-100-1 — 40 —
Dimensions in millimetres
Weld 1 details
Surfacing
Weld 4 details
Weld 5 details
Figure B.2 : 18 cylinder internal combustion engine frame and 570 mm bore

Décembre 1995: Boutique AFNOR Pour: Solios Carbone Client 8319200 Commande N-20100112-383554-TA Le 12/1/2010 16:25

Uploaded by

Copyright:

Available Formats

Décembre 1995: Boutique AFNOR Pour: Solios Carbone Client 8319200 Commande N-20100112-383554-TA Le 12/1/2010 16:25

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Décembre 1995: Boutique AFNOR Pour: Solios Carbone Client 8319200 Commande N-20100112-383554-TA Le 12/1/2010 16:25

Uploaded by

Copyright:

Available Formats

What is the purpose of this document?

What is the purpose of this document?

What terminology is defined in this document?

What terminology is defined in this document?

NF E83-100-1

Pour : SOLIOS CARBONE

FE038722 ISSN 0335-3931

Classification index: E 83-100-1

ICS: 01.040.25; 25.160.10

Construction of fabricated assemblies

French standard approved

Descriptors Technical International Thesaurus: welding, quality classes, inspection, welders

© AFNOR 1995 AFNOR 1995 1st issue 1995-12-F

2 Normative References ...................................................................................................................... 4

4 Welding quality class ..................................................................................................................... 11

5 Extent of the inspections ............................................................................................................... 14

Annex B (informative) Examples of welding quality class determination ............................................... 37

NF A 03-400, Iron and steel — General principles of the fatigue test.

NF E 83-100-4, Construction of fabricated assemblies — Welding processes — Part 4: Manufacture Inspection.

Molten metal Welded joint

Table 1: Nominal throat thicknesses of the most common welded assemblies

Designation Schematic representation

Table 1: Nominal throat thicknesses of the most common welded assemblies

Designation Schematic representation

5 — full penetration angle weld

6 — full penetration angle weld

7 — strong inclination angle

Table 1: Nominal throat thicknesses of the most common welded assemblies

Designation Schematic representation

8 — welding with natural

Table 1: Nominal throat thicknesses of the most common welded assemblies

Designation Schematic representation

11 — full penetration butt weld

4 Welding quality class

4.2 Risks encountered in case of the weld’s failure

Table 2: Grades of risks according to the consequences of a failure in operation

4.3 Types of operational stresses of welded joints

4.3.2.2 The cyclic load is evaluated according to the ratio:

4.3.3 Non-calculated assemblies

4.4 Choice of welding quality classes

Table 3: Choice of welding quality

Welding quality classes

σc = calculated factored stress.

5 Extent of the inspections

5.1 General case

Extent of the inspections

A 100 > 50% 1) 2) 3) > 50% 1)

B 100 — > 10% 4)

5.2 Extent of the supplementary inspections

NF E 52-109-1, Lifting and handling — Welded joints — Part 1: Fabrication.

NF P 22-470, Steel construction — Welded connections — Details and design of welds.

Init numérotation des tableaux d’annexe [A]!!!

A.1 General Remarks

Weld toe angle

1) Text drafted from the document IIS/IIW-693/81.

Direction of the force

A.2 Fatigue tests on welded assemblies

A.2.2 Choice of a representation of the results

A.2.3 Analysis of results