This article was downloaded by: [University of Cincinnati Libraries]

On: 05 January 2015, At: 05:44

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41
Mortimer Street, London W1T 3JH, UK

Welding International
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/twld20

Dimensional tolerances of welded products

E Richter
Published online: 05 Jan 2010.

To cite this article: E Richter (1987) Dimensional tolerances of welded products, Welding International, 1:1, 58-61, DOI:
10.1080/09507118709449026

To link to this article: http://dx.doi.org/10.1080/09507118709449026

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the
publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or
warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed
by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with
primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings,
demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly
in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is
expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 ~~

. $ C h l ~ C i S 3 f C C / J t l i k1986 36 (6) 246-248

Dimensional tolerances of welded products

E Richter
Selected from Srhwpircterhnik(~er~;erlin)
1986 36 (6):Reference ST/86/6/246; TransIation 012

The principal aims of production engineering are to
obtain high productivity with low costs while ensuring
product quality. The individual stages of development
from manual production to automated mass production
can be considerably affected by each respective stage of
development.' Tolerance specifications and tests based
on scientific experiment are important to ensure and
improve the quality of products and production
Downloaded by [University of Cincinnati Libraries] at 05:44 05 January 2015
Dimensional tolerances are determined by
experiment if no previous model exists and if calculation
is not possible.
T h e four methods of calculation described in TGL
19115 (maximum-minimum method, theoretical
probability method, compensation method and method
of group inter~hangeability)~form the basis for
calculating dimensional and tolerance chains. Because of

processes. the importance of these methods of calculation for

With increasing automation in welding processes,
together with the need to use robot technology in
assembly, and the improved quality and reliability of
component parts and function groups with growing co-
operation, it is often necessary to keep geometrical
tolerances within narrow limits. According to ref.2
dimensional tolerances can be established in three ways:
determining tolerances of dimensions on individual
component parts and products for correct manufacture,
the possibilities of using these for welding production
are examined, looking in particular at product groups
and accuracy classes, joint split and assembly as well as
seam cross section and distortion, and are explained
using practical examples.

- by formally adopting tolerances from similar DIMENSIONAL TOLERANCES

constructions or specifying tolerances based on
empirical values;
- by determining tolerances
experiments;
- by calculating tolerances.
If the first method is used, the dimensional
tolerances are on average too small with regard to
functions to be fulfilled. This method makes production
considerably less economic because manufacturing costs
increase exponentially with decreasing dimensional
tolerances.
T h e dimensional tolerance (T)is the difference between
the maximum permissible value (G) and minimum
on the basis of permissible value (K):
T=G-K ... [l]
The permissible variations for length measurements
of welded constructions are given as a function of four
accuracy classes in Table 1? Differences in function and
manufacture can be allowed for with the four accuracy
classes. Since observance of narrow limits within which

Table 1 Permissible variations for length measurements of welded products?

Accuracy class Nominal dimension range, mm
30-120 120-315 315- 1000 1000-2000 2000-4000 4000-SO00
A (fine) +I +1 +2 f 3 +4 f 5
B (mean) +2 f 2 +3 f 4 +6 +8
C (approximate) f 3 f4 +6 +8 f l + 14
D (very approximate) f 4 +7 . f 9 f 12 + 16 + 21
Product groups Accuracy classes '

Rail vehicles B
Motor vehicles B; C
Crane construction B; C
DAR construction B; C
Steel construction B; C
Machine construction B; C; D
Shipbuilding C; D

58 WELDING INTERNATIONAL 1987 No.1

 Table 2 Dimensional variations from welding FIT
production which determine the dimensional variations Correct fit of component parts when assembled is
of welded products. ensured if the condition given in [3] is fulfilled'
.m
Dimensional variations Abbreviation
for Ts = 1 Ti S To = TFu ...[3]
i = l
dimensional
tolerance Ts - sum tolerance; Ti - individual tolerance; To -
final term tolerance; TFu - function or operating
Individual parts TE-r tolerance.
Assembly Tz The operating tolerance here can, for instance, be
Distortion Tv the tolerance for the width of the weld joint in
Measuring uncertainty M accordance with ref.8 o r for a welded assembly as in
Table 1, depending on how one looks at this.
The fit of the component parts largely determines
the economy of the assembly process and is a pre-
requisite for automation of the joining and handling
process in the assembly.
The simplest and most economic method of
manufacturing is to increase the operating tolerance, i.e.
Downloaded by [University of Cincinnati Libraries] at 05:44 05 January 2015

to change from a precise to a more approximate

accuracy class (see Table 1). This can only be done
providing that the functions are only impaired within a
permissible range and the customers o r employer are in
agreement with larger accuracy classes.
T h e sum tolerance can only be minimised if the

1
A" -
Transverse distortion Ab as a function of the weld
quantities in [2] are minimised.
T h e tolerances of the individual parts and the
assembly can be considerably influenced through
seam cross section A,; free distortion
suitable choice of the method of manufacture. Using
AN
AN= 1.7.. . 7.5AZw,Ab up 0.03.- 4- 0.22;
S
holding jigs is another way of reducing tolerances. The
theoretical pre-conditions for this were discussed in
AN ref.8 taking into account the joint width and the
AbMAcc 0.03- 4- 0.03.
S
distortion. Functional relationships between individual
and equipment tolerances are explained using graphs.
Item S, bspr
The amount of distortion, i.e. shrinkage after welding, is
mm mm
determined by the welding method, weld cross section
1 10 0 and restraint of the welded parts (Fig.1). T h e size of the
2 10 4
3 12 4 weld cross section A, can on the one hand be seen from
the joint shape and the parts of the joint which
determine the measurements e.g. the joint width bsp,
the dimensions may vary involves extra costs in the
angle of opening a, height of weld c.
welding operation the accuracy classes must be selected
Alternatively, the weld cross section is determined
with care. In many cases the mean accuracy classes B
from influencing quantities specific to the particular
and C suffice?.5
process such as: welding current I,, welding voltage Us,
T h e advantage of the tolerances specified in Table 1
welding rate v,, electrode diameter d,. The effect these
over those specified in TGL 28973 is that these have
have on the cross section A, can be represented in a
been tried out in practice over many years with
general form by equation [4]:
particular consideration being given to the special
requirements of fusion welding. The dimensional . ..[4]
variations occurring on welded constructions are in fact
not only determined by the dimensional variations of the In addition there are the effects of the shielding
individual component parts but are also dependent on a gases (pure shielding gas o r mixed gas), welding flux,
number of influencing factors from the manufacturing weld pool formation erc which must be considered
process: the most important of which are listed in Table separately.
2. The measuring uncertainty is determined by the
According to this, the sum tolerance T, is: method of measurement and apparatus chosen and
should not exceed 20% of the dimensional tolerance.
T, = JTET2 + Tz2 + Tv2 + M Z . ..[2] CALCULATING WITH DIMENSIONAL
TOLERANCES
which agrees well with results obtained in practice. When calculating with dimensional tolerances, one

WELDING INTERNATIONAL 1987 No.1 59

 Table 3 Nominal dimensions and permissible variations
of the component parts for the container in Fig.2.
Tolerances for component parts to TGL O-S570 and
ref.9,kiven in mm
Abbreviation Nominal TE Tabs G K
for dimension
component
part
4
L 7000 k S 16 7008 6992
A 198 f 2 4 200 196
B 1646 k4 8 1650 1642
b,, 4 f l 2 5 3
b 1 k0.6 1.2 1.6 0.4
Ts = Ti = 56;Ts = = 17.75 > TFu= 16
1 s Tolerances for component parts to TGL 0-8570 and
- + <f-- TGL 0-2310, given in mm
c c
Abbreviation Nominal TE Tab, G K
Downloaded by [University of Cincinnati Libraries] at 05:44 05 January 2015

for dimension
2 Guide dimension and dimensional chain for container component
example: Sheet thickness, s = 20mm; joint gap, 6, = 4
f lmm; shrinkage Ab = 1 k 0.6. part
L 7000 k 8 16 7008 6992
starts with the graph of the dimensional chain. Following A 198 k0.5 1 19S.5 197.5
this the starting equation is drawn up and solved B k 1.5 3 1647.5 1644.5
according to the final term. In accordance with the 2nd
principal theorem of the tolerance condition T, = G, -
2, 1646
4
1
k1 2
k0.6 1.2
5
1.6
3
0.4
K, the final term tolerance To and other sought
quantities are determined. The control calculation is T, = Ti = 30;T, = JCTf = 8 < TFu = 16
then done with the 1st principal theorem of the
tolerance condition:
The resulting starting equation for the dimensional
m chain is:
Ts = 1 Ti
i = l
2A + 4B + 5b,, - 5Ab - L = 0 .. .[6]
To explain this procedure a container (Fig.2) is
chosen as an example; it comprises two bases (A and F)
ifA = F a n d B = C = D =i= E.
and .four shell rings (B, C, D and E). The dimensional
chain can then be identified as a closed continuous line The solution after the final measurement M, = L
consisting of the construction dimensions A to F, the gives:
joint gaps b,, and the transverse deformation (shrinkage)
Ab. M, = 2A + 4B + 5b,, - 5Ab .. .[7]
To simplify things it is assumed that the nominal
dimensions and dimensional tolerances are the same for
the two container bases. The same should also apply for The final term tolerance To = G, - K, is found
the four shell rings, the joint gaps and the transverse from
deformations.
T h e construction dimensions KM, are calculated G, = 2GA + 4G" + 5G,,, - 5K,, .. . [Sa]
from the difference between the guide measurement
RM, and the share of the joint gap b,,, therefore
K, = 2KA + 4K" + 5K,,, - 5G,, ...[Sb]
KMA = K M = R M -!h
F " 3
:. . [5a]
and for the values in Table 3a gives the final dimensional
and tolerance To = 56mm, i.e. it was assumed that all
positive maximum dimensions and all negative minimum
KM,, = . . . = KM, = RM,, - b,, .. .[5b] dimensions occur in the welding process, in order to
determine G, and conversely K,,; a situation which is
The nominal dimensions (construction dimensions), very improbable. If a quadratic error propagation T, =
tolerances, maximum and minimum dimensions for the which better reflects the real operating conditions is
example chosen are listed in Table 3. assumed, the value drops to T, = 17.75. This value is
60 WELDING INTERNATIONAL 1987 No.1

TE, - tolerance of individual part
3 Relationship between equipment and individual part
tolerance according to Taober.
Downloaded by [University of Cincinnati Libraries] at 05:44 05 January 2015
however still too high compared with the function
tolerance TFu = 16. A check should then be made, in
accordance with the working sequence for the tolerance
synthesis, to see whether the function tolerance can be
increased, i.e. change from accuracy class B to C (Table
1). This possibility gets a negative answer for the chosen
example so the next step is to reduce T,, i.e. lower the
tolerance of the individual parts and thus pre-fabrication
in assemblies. For the first step more accurate cutting
methods and production means including cutting
equipment can be used so that finer tolerances to TGL
0-2310 are ensured.
T h e finer tolerances lead to a sum tolerance T, =
= 8 6 16 = TFu. This would theoretically
guarantee that the system dimension for the container L
= 7000 k 8mm is within the pre-determined limits in
Table 3.
WELDING PROCESS
When calculating the dimensional variations to be
anticipated, those relating to assembly and distortion
were taken into account in addition to those for cutting
the individual parts to size. For the welding operation,
this means that the joint gaps are to be kept within the
assumed limits through the use of. devices such as
production means (Fig.3): Moreover, the distortion,
also the weld seam measurements to TGL 11776 must
be kept within the assumed o r given limits through the
welding system, such as parameter combination,
electrode diameter, shielding gas, polarity and their
permissible variations.’” T h e effects of component shape
and size and welding sequence on the dimensional
WELDING INTERNATIONAL 1987 No.1
variations cannot as yet be calculated. The effects of
these influencing factors must still be determined
experimentally for the time being.
SUMMARY
The effects of many influencing factors on dimensional
variations in welded construction can largely be
determined by calculation using the described method
and can therefore be used for manufacturing
preparation. The interrelationships between parts
preparation, assembly, welding and distortion were
considered in a composite manner and the multifarious
special test results relating to accuracy in thermal
separation and to welding deformation were included as
an organic whole in the overall presentation of welding
production.
The systematically calculated tolerance
specifications ensure an assembly with little need for
adaptations and continuous welding production. It was
also shown that partial solutions in the form of more
exact cutting to size o r more precise assembly only give
partial effects in the direction of quality improvement,
lowering adaptation costs and material and power
consumption. The comprehensive rationalisation effect
will only be achieved with a general system of the given
production tolerances for a production where the
dimensional variations for each respective stage of the
process are within the given limits.
References
1 Trumphold H, Beck Ch, Riedel T ‘Finding dimensional
tolerances and dimensional chains in interchangeable manufacture.’
Berlin: VEB VerlagTechnik 1984.
2 Trumphold H, Beck C h ‘Fixing optimum tolerances taking into
account the dimensional chain theory and the statistical characteristics
of the manufacturing process.’ Fertigungstechnik und Betrieb, Berlin
21 (1971)4, pp. 242-246.
3 Uelze A ‘Inaccuracy of manufacture and finding tolerances of
welded components.’Scbweirstechnik 1981 Berlin 31 (6) 260-263.
4 Diesnt H: ‘Permissible dimensional variations in welded
constructions in accordance with the new DIN 8570.’ Prukfiker 1976
Diisseldorf 28 (7) 128-130.
5 Cramer K ‘Permissible deviations for measurments on welded
constructions. Examples of use.’ Pruktiker 1977 Diisseldorf 29 (5) 85-
86.
6 Cramer K ‘Dimensional variations on welded constructions.’
Il‘erksturtrtndBefrieb 1981 Munich 114 (11)831-832.
7 Richter E, Schilling W, Weise M: ‘Book of tables for assembly.’
Berlin: VEB Verlag Technik 1984, p. 32.
8 Heinz 0. ‘Calculating dimensions and permissible tolerances for
individual parts of the reqcired component tblerance.’ Schweirstechnik
1973 Berlin 23 (4) 176-178.
9 Neumann-A, Richter E: ‘Book of tables on welding and brazing
processes.’ Berlin: VEB Verlag Technik 1979, 3rd edition, p. 32, 191,
365.
10 Richter E, Matthes K-J, Renatus W. ‘Production tolerances in
1983 Berlin 33 (3) 114-116.
fusion welding.’ Sch~r~eisstec/znik
61