Forwarded message from Michele Ciavarella...



Dear Dr. Cebon

Following I attach a draft version of the introduction of a paper I am
preparing
with some collegues in Italy, about the luck of design standards for
Semitrailers and their connections to frames in trucks.   I would
appreciate any preliminary comments from you on that text, as I am
getting more familiar with the situation outside Italy.   At some point,
I am going to submit the paper to the periodical.  Please feel free also
of suggesting any possible direction to take into the research, from a
practical (grants, legislations, standards, etc.) as well as theoretical
standpoint.

Thanks in advance,

Regards, Michele

Dr. Ing. Michele Ciavarella. PhD
Lincoln College
Ox1 3DR
Oxford UK
Fax: 01865 279802

at Southampton:
Senior Research Fellow
Mech. Eng. Dept.
Univ. Southampton
Highfield
SO17 1BJ
UK
[log in to unmask]
http://www.mech.soton.ac.uk/
+44 1703 59 2899 (off.)
+44 1703 59 3230 (fax.)
+39 338 63 64137 (mob.)

general use:  [log in to unmask]

in Bari (Italy):
Dip Prog. Prod. Ind.le
Politecnico di Bari
V.le Japigia 182
70126 BARI (Italy)
Fax: +39 080 5460 777

[log in to unmask] (Home),  [log in to unmask] (off.)


The distribution of loads in the connections between bodies and frames in
semitrailers lorries: structural analysis and design recommendations

M.Ciavarella1, G. Bergamini2 , G. Demelio3

1Dept Mech Eng - Univ. of Southampton - Highfield Southampton SO17 1BJ - UK
2 Bergamini & C s.a.s. - Via C.Rosalba 46 F,  70124 - BARI
3 Dip Prog Prod Ind - Politecnico di BARI, Viale Japigia 182, 70126 - BARI

ABSTRACT

Recently, the needs for weight and size reduction have motivated the use
of more flexible frames in semitrailers lorries, together with stiffer
body structures. In some cases (for example in Italy), when the reduction
of stiffness is made, there is no need for a special design procedure or
homologation of the resulting vehicle, according to the actual
legislation. In some other cases (for example in Germany) the reduction
of stiffness of the frame may be so high that a test is required for the
actual configuration frame/van.

It is shown in this paper, however, that the structural behaviour of the
coupling often escapes the checks of common design procedures or
legislation. This may explain the source of various dramatic accidents
that are occurring with presently circulating vehicles.

In particular, the structural behaviour of various kinds of couplings
between bodies and frames in articulated refrigerated lorries has been
studied using the Finite Element Method (FEM). The distribution of loads
shared between bodies and frames and the amount of the structural
coupling between the bodies have been determined when the stiffness of
the frames is reduced. It is shown that the stresses acting on the
fasteners can reach a dangerous level in the fifth wheel zone. An
accurate evaluation of these stresses requires far more computational
effort than is actually recommend by legislation, and indeed examples of
this kind are not available in the open literature, to the best of the
authors' knowledge. The lack of standards available in order to assess
safety conditions for fastener design is highlighted, together with the
fact that actual classification of these vehicles appears inadequate, and
needs serious reconsideration.

1. INTRODUCTION

	During the 90s, several frames for semitrailer vehicles have
appeared in the North-European market in order to respond to the needs of
vehicle weight and size reduction. Their dimensions are such as to
require a contribution from the stiffness of the body for use on the road.
The dimensioning criteria used to date permit to neglect the structural
contribution of the body and consequently have lead to the realisation of
classical types of frames usually called "rigid and undeformable". They
are typically built with longitudinal bars and crossbars with a height of
the order of 140 mm, in the front part.
	The description "rigid and undeformable" suggests that these
frames are apt to any use. With the aim of reducing the weight, and so
giving up on rigidity and "undeformability", in Germany several frames
have been designed where the overall height in the region of the fifth
wheel goes down to a limiting 45 mm. Such frames are designed for use
with monocoque bodies as no longer capable of supporting, within the
maximum stresses within limits of regulations, a uniformly distributed
load of magnitude equal to the maximum load prescribed by homologation
procedures for an independent chassis.
In such cases, the height of the frame is insufficient not only in the
region of the fifth wheel, but also in the area of the suspensions, where
it can be as low as 160 mm.

In Italy, the manufacturing firms operating in this area (at least, the
ones that we know) have adopted less drastic solutions . The weight
reduction has been obtained by improving the quality of steels and by
realising longitudinal bars with an height of about 60 mm in the region
of the fifth wheel and greater than 300 mm in the region of the
suspensions. These dimensions have been chosen to keep the load capacity
of the independent frame equal to the nominal value. In other words,
notwithstanding the reduction of dimension and mass, the frame is capable
of sustaining the nominal load. This within the safety margin prescribed
by legislation for new designs (a ratio of three with respect to the
nominal value of the least between the yielding limit and 75% of the
ultimate strength) [1, 2, 3]. The need of homologation of the frame-van
ensemble may then be avoided. This is an important point for
manufacturers, as homologation of the frame-van ensemble would require
calculation, verification and certification of the body unit, all
operations which are considered not possible (at least in-house) by the
majority of manufacturers of isothermal bodies.

	However, in the absence of legislation which would constrain the
maximum admissible deformations, (as happens in railways and aeronautical
constructions), the deformability of frames under distributed nominal
load has continuously increased alongside the improvements in the
mechanical characteristics of the steels adopted.

	In contrast to the frames, the refrigerated bodies are still
rigid and actually the tendency of manufacturers has been to increment
this rigidity to compensate for the reduction of flexural and torsional
deformability of the frames. This has made less plausible the reference
to a non-continuum distributed load, such as the ensemble of concrete
blocks used for homologation verifications, as a base for the calculation
of stresses in frames and the relevant certifications.

	In the coupling of an independent but flexible frame with rigid
(even if theoretically not structural) body, in practise a transfer of
stresses arises from the more flexible element, the frame, to the more
rigid one, the body. The latter is loaded, as a first approximation, in
inverse proportion to the ratio of the rigidities.
The trade-off solution of a frame that is independentently homologable
has not prevented the bodies to be partially structural  we propose at
this purpose the term "collaborant"  which gives rise to problems of
stresses, particularly in the connections. Indeed, these problems are
ignored by referring improperly to the non continuum distributed load
mentioned above, which only poses constraints that are relatively modest
and predictable in an elementary fashion.
This loading hypothesis which is implicit in the actual regulation has
often been taken as a reference for the design of bodies. Indeed, even in
recent publications [5] the coupling of a body with an independent frame
seems to constitute enough justification to consider the stresses
exchanged between different segments of the body as negligible, when
considered as equal elementary chunks.

The first question that we raise is: is it possible to define a clear
distinction between monocoque bodies, conceived to be connected to
non-independent frames or even non connected, and bodies connected to low
rigidity independent frames?

	Many elements in effect give the impression that, even if this
distinction were possible, the stresses on body units are anyway not to
be considered negligible. Also, a greater attention should be paid to the
connections between frame and body unit. In fact, the amount of loads
exchanged between the two is not that trivial to estimate, considering
the different loading conditions and the dependence on relative rigidities.

As a matter of fact, the original motivation for the present
investigation has come from the knowledge of a series of accidents, some
of which happened in Italy and for which one of the authors has been
called as technical witness. These accidents have involved vehicles with
light frames, characterised by the presence of deformations, ruptures and
even tearing of connections mainly in the region of the fifth wheel.

The problem that we face is to verify if the accidents are deriving: (i)
from accidental causes connected to the dynamics of the vehicle; or (ii)
by macroscopic non conformity in the manufacturing; or else (iii) if
there is an insufficient analysis (and legislation), of the stresses on
bodies and connections that come into play when the body "collaborates"
with the only formally independent frame.

In the European legislation we have not found any particular
prescriptions for the connection of body of trailers to "independent"
frames apart from a generic reference to "state of the art" design
procedure.  This "state of the art" would probably consist in assuming
that the body unit would not slide off under the action of longitudinal
accelerations, the braking actions, or transversal accelerations, due to
centrifugal actions. However, clear indications are missing for any
quantitative reference values .

The background leading to the common practise of neglecting the
structural contribution of a body unit to an independent frame, consists
in the following thought: "a collapse of the body unit itself would cause
a complete loss of rigitidy such that the loading conditions would get
close to the distributed nominal load". The latter, as we know, is well
sustained by the frame-chassis within the appropriate safety margins,
according to the existing regulations. However, this is an oversimplified
assumption, and does not fully exclude the potentially dangerous global
collapse.
>From the point of view of connections, the hypothesis of distributed load
leads to neglecting all additional loading contributions coming from the
"collaboration" between frame and body unit which sum up to very high
peak loads and may end up by reaching dangerous levels.

On these grounds, the authors are conducting a large program of analysis
whose main results to date are summarised here, particularly with respect
to the computation of effective stresses in the chassis, in the body unit
and especially in connections with different rigidities of the frame.




%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%