See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/260984995

On site interventions on decayed beam ends

Chapter · January 2008

DOI: 10.13140/RG.2.1.2343.4083

CITATIONS READS

4 342

2 authors:

Benedetto Pizzo Kay-Uwe Schober

Italian National Research Council Mainz University of Applied Sciences
94 PUBLICATIONS 1,232 CITATIONS 96 PUBLICATIONS 551 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Modeling of historic timber joints View project

BUW - The reliability of timber columns based on stochastical prinziples View project

All content following this page was uploaded by Kay-Uwe Schober on 06 August 2015.

The user has requested enhancement of the downloaded file.

COST E34 - WG1: Bonding on Site

6 ON SITE INTERVENTIONS ON DECAYED BEAM ENDS

Pizzo B., Schober K.U.

6.1 Overview

In the past, wood has been the material used mainly for the construction of horizontal load bearing
structures, for practically all types of buildings. It is a relatively high durable material per se, and it is
known that the simple passing of time is not responsible for any decrease of its mechanical
characteristics [1]. But wood is a biological material and hence it is naturally subjected to decay if some
precautions are not taken for its preservation. Unfortunately, this situation is scarcely encountered in old
buildings and decay of wooden elements has been experienced by engineers and people familiar with
structural restoration. Nevertheless, unless related to original design problems or new structural
requirements, the majority of “diseases” are concentrated at the end of beams (Figure 6.1), often in direct
contact with the masonry, whereas the rest of the element is in generally sound condition.

Figure 6.1: Two examples of decayed beam-end: on the left the end is already cut for the subsequent
intervention.

In several cases, mainly in the recent past, wooden structures have been integrally or partially
substituted, even without an effective need, due to a superficial technical and /or economical evaluation.
Some exceptions have been made only for a minor amount of historically important buildings. Nowadays,
notably starting from the last three decades, an ever increasing number of designers try to protect the
original wooden structures, even for buildings less interesting from a historical point of view, on the basis
of aesthetical or economical considerations. In fact it is by now accepted that restoring ancient wooden
structures allows applying all the selected designing solutions without substantially increasing the cost of
the intervention due to the use of products and techniques adaptable to the several service conditions.
However, the specific characteristics of wood make the approach to the diagnostic and the design
different from that one requested for the other building materials. The experiences of past interventions
on ancient wooden structures demonstrate that some precautions are necessary in order to assure the
greatest durability of the joints [2]. Some general rules can be described as follows:
1. A visual and instrumented diagnostic survey must always precede the interventions, in order to
obtain objective data and limit the intervention to the barest essentials;

2. Adhesives specific for structural intervention on wood have to be used ([3, 4]) observing the
specific suggestions of the adhesive’s producer;
3. The wood has to be allowed to exchange moisture with the surrounding ambient, and a certain
thermal stability should be favoured as far as can be possible;
4. Products should be rationally used and limited to the essential (e.g. the filling of shrinkage cracks
should be preferably avoided except in specific cases).
The interventions on single elements are generally carried out when a beam-end is biologically decayed
due to insect and / or fungal attacks, or when it is physically missing. It is indispensable that the rest of
the beam is in good conditions, which needs to be approved by a diagnostic evaluation, and hence only

40 On Site Interventions On Decayed Beam Ends

 COST E34 - WG1: Bonding on Site

the replacement of the attacked part is required. This process can be carried out by using a prosthesis,
made for example of solid wood or glulam, that is connected to the sound part of the beam by means of
connecting elements: they could be either bars (made of steel or GFRP) or plates (often made of steel,
most recently of carbon-fibre). This type of technique avoids the dismantling of the entire wooden
structure, thus considerably limiting the scale of the intervention. Additionally, in this way the original load
distribution in the whole structure is not changed.
Several companies all over the world have developed similar techniques, since the early seventies. The
most diffused have surely been:
ƒ the W.E.R. (Wood Epoxy Reinforcement) System, developed in Canada [5];
ƒ the Beta-System developed by Renofors originally in The Netherlands and extensively used in
the United Kingdom and in Germany ([6], Figure 6.2 taken from [7]).
Recently, some other techniques, have been used mainly in Italy but also in the U.K., based on a
“minimum intervention” approach and making use of the design rules described in the design code
ENV 1995-2:1997 [8]. More details are available on the Licons Project web site [9].

Figure 6.2: Schematic steps of Beta-System

technique

6.2 Design methods

In general terms, the repair is different, often more complicated, than a new construction. The design of
the repair interventions on decayed ends of wooden elements has frequently followed an approach based
on the basic construction rules, probably simplistic but evidently successful.
For example, the one used for the Beta-System [6] considers a strain distribution in repaired (or
reinforced) beams as similar to that one of sound wood and moreover requires that the Moment of
Resistance of the sound timber is equal to that one of the rods used for the intervention (Figure 6.3). In
such a way the total area of rods can be calculated. The length of anchorage is calculated simply based
on the resin-to-rods allowable bond stress and on shear strength of sound timber (so, rigid bond between
resin and wood is assumed relative to the design safety value of shear strength for wood).

On Site Interventions On Decayed Beam Ends 41

COST E34 - WG1: Bonding on Site

Figure 6.3: Design method used for the Beta-System

A slightly more sophisticated approach is that of the W.E.R.-System based on the philosophy that “the
reinforcement must have a load bearing capacity which is equal to the allowable capacity of the timber
itself, when it is in sound condition”. No hypotheses are made on the stress or strain distribution, and the
design of the reinforcement is undertaken with an expected load obtained by considering the allowable
bending stress of timber of the same species of the beam and the dimensions of the real element. The
size of the reinforcement is determined by the existing maximum moment of the beam. The effective
dimensions for the reinforcement (says for example the thickness and the height for a plate, Figure 6.4)
are then stated by verifying it in relation to the horizontal shear and for the crimpling and buckling forces.
Also in this case, the design of the length of the connection between prosthesis and sound wood is based
on the assumption of rigid bond between resin and wood at the allowable value of shear strength for
wood, even if other factors (e.g. wood conditions, load conditions, service environment, etc.) are taken in
account. An interesting feature of this approach is that a fire safety condition enters into the design of the
reinforcement, even if in an empirical way: a minimal distance from the edges of the beam is suggested in
order to ensure a sufficient wood cover.

Figure 6.4: Decayed end reinforcement (W.E.R.-System)

A most recent approach, broadly used for example in Italy, is that one in which connection bars are
calculated with reference to the ENV 1995-2:1997, where the Eurocode 5 is applied in this case not to
new timber structures but to the old. The design of rods is therefore based on the evaluation of their pull-
out strength (Limit States approach).
According to Lauriola [8] it is convenient to subdivide the procedure in two parts:
1. quantifying of parameters in the beam after the cutting of the element (Fig. 6.5)

42 On Site Interventions On Decayed Beam Ends

 COST E34 - WG1: Bonding on Site

2. designing of the connection bars.

As for phase 1 the procedure considers that, after the estimation of the loads (distributed load on the
interested beam, constraining force at the support, and shear force at the section of calculation, bending
moment at the section of calculation) the tensile forces on each of the two lower connection bars can be
calculated as:
N1 N 2
Nd = + (1)
4 2
where

⎛ Tcalc ⎞
N1 is the load due to the shear force at the section of calculation ⎜ ⎟ (2)
⎝ tan α ⎠

⎛ M calc ⎞
N2 is the load due to the bending moment at the section of calculation ⎜⎜ ⎟⎟ (3)
h
⎝ u ⎠

Figure 6.5: Shear forces and bending moment at the beam section

Regarding part 2, the pull-out strength, Fax,Rd, is given by considering the minimum value among the joint
strength related to the:
ƒ rupture of the bond between wood and adhesive, F1
ƒ tensile cohesive rupture of wood, F2
ƒ rupture of the rod, F3
with:
k mod ⋅ π ⋅ d equ ⋅ l a ⋅ f v ,k kmod ⋅ f t ,0,k ⋅ Aeff f y ⋅ Ares
F1 = (4) F2 = (5) F3 = (6)
γ m,w γ m,w γm
where:
kmod coefficient of modification (depending on the duration of the accidental load)
dequ equivalent diameter (minimum of hole diameter and 1.25·d)
λa anchorage length of the rod
fv,k shear strength at the bond interface
1.2 ⋅ ρ k1.5
is given by: , ρ k being the density of wood. (7)
1000 ⋅ d equ
0.2

The new version of ENV 1995-2 (2004) presents a different expression for fv,k
ft,0,k characteristic tensile stress of wood parallel to grain
Aeff effective area for tensile rupture of wood
fy characteristic tensile stress of the rods
Ares resistant area of rods

γm partial safety coefficient of materials, γ m,w being the partial safety coefficient for wood

By this way the diameter of the rods can be derived from the expression:

On Site Interventions On Decayed Beam Ends 43

COST E34 - WG1: Bonding on Site

Fax , Rd = min( F1 , F2 , F3 ) > Nd (8)

The minimal length of anchorage of rods is given by:

⎧0.4d 2
l a ,min = max ⎨ (9)
⎩ 8d
The effect of adhesive rheology on the performance of the joint is not considered in all these approaches,
and probably a too precautionary value of the shear strength of the adhesive on wood is used. Moreover,
all the calculations are based on the hypothesis of conservation of the plane sections.

6.3 Methods of application

6.3.1 Repair of decayed ends using adhesives and steel reinforcement

Considering the variability of real cases and the variability of their requirements the following systems
illustrated should only be considered as examples, and intended as general models. According to the
more recent orientation and in order to make the interventions as less intrusive as possible, the
connecting elements are confined to the four corners of the beam, and their length is reduced to the
minimum necessary. As already provided in the new version of Eurocode 5, some experimental tests
should preferably be carried out in order to evaluate such optimised length (see chapter 5, Glued-in rods).
The operating procedures depend on the specific requirements of the site, but they are generally of a
common approach, that can be described as follows:
ƒ propping of the beams;
ƒ removal of the decayed portion of the timber, usually terminated in inclined cut (Figure 6.6);
ƒ substitution of the decayed wood with a new section of wood, shaped in such a way that external
dimensions are the same (Figure 6.7);
ƒ preparation of internal holes or external grooves between the original wood and the new
introduced wood for the positioning of the connecting elements. (Figure 6.8);
ƒ partial filling of the grooves or holes by an adhesive (generally but not necessarily thixotropic)
and introduction of the connecting elements inside the grooves or holes (Figure 6.9);
ƒ insertion of a final wood fillet, hiding the grooves (Figure 6.10), not necessary in the case of
holes;
ƒ removal of the beam supports after the complete curing of the adhesive.
The final appearance of an intervention is shown in Figure 6.11.

Figure 6.6: Inclined cut of decayed end Figure 6.7: Shaping of integration element

44 On Site Interventions On Decayed Beam Ends

 COST E34 - WG1: Bonding on Site

Figure 6.8: Linking grooves between integration Figure 6.9: Partially filling of the grooves with
and original timber adhesive

Figure 6.10: Insertion of a final wood wedge Figure 6.11: Repaired beam

Several real interventions realised in the recent years in historic monuments in Italy are described and
documented in [10].

6.3.2 Repair of decayed ends using epoxy polymer concrete and GFRP reinforcement (bars and
plates)
Epoxy resins have been used in specific cases to repair timber that has deteriorated from decay or insect
attack and certain structural deficiencies in existing construction. The compressive strength and filling
capabilities of epoxies can aid in repairing timber structures. The tensile, shear, and bond strength of
epoxies as structural wood adhesives are, however, limited and are further subject to variability due to
conditions of use. Mechanical reinforcement should be used in conjunction with epoxies for repairs
intended to develop shear capacity. For structural rehabilitation and restoration the decayed ends were
cut and replaced by timber or polymer concrete where the shear and tension forces resulting from loads
are transmitted by GFRP (glass fibre reinforced plastic) reinforcement with a synthetic resin filling
compound between the new and existing part of the structure (Figure 6.12).

Figure 6.12: Shear reinforcement with GFRP

On Site Interventions On Decayed Beam Ends 45

COST E34 - WG1: Bonding on Site

In some cases the decayed part of the structure will be replaced by a polymer concrete (PC) supplement
using rapid-setting organic polymers as binders and small grade aggregates such as sand or gravel. PC
is increasingly being used as an alternative to cement concrete in many applications. Today, PC is used
for finishing work in cast-in-place applications, precast products, highway pavements, bridge decks and
waste water pipes.
The application of PC in structural rehabilitation and restoration has several advantages
ƒ Historically significant structures can be protected by a minimum disturbance of the construction
and minimum replacement of decayed parts.
ƒ The appearance of the timber structure will not be changed.
ƒ The section design can be easily executed by a timber formwork on the level of the necessary
construction height.
ƒ All work can be undertaken from the top side; the suspended ceiling will remain unaffected.
ƒ The floor below the reconstruction work can be used with some restrictions
ƒ Suitable for inhabited floors and restoration of complicated timber joints
ƒ The full load-carrying capacity is achieved after one day.
Research work has been carried out at the Bauhaus-University of Weimar to improve the Beta-technique
for glued-in rods and replacement of the original Beta resin with bonding adhesive PC Compono® to
obtain a higher quality of adhesion and the load-carrying capacity. Compared to the Beta-system an
increase of the structural performance of 140% for timber-on-timber supplements and 30% for timber-PC-
supplements could be achieved with the new epoxy concrete.

Compression repair
Deterioration caused by decay or insect attack can be repaired with epoxy PC by removing the damaged
wood and filling the void with an epoxy or epoxy / wood fibre mixture, thus restoring strength sufficient for
the particular design. Because the curing of epoxies is typically an exothermic reaction, it is
recommended that larger voids be filled with wood prior to epoxy injection to minimize the amount of
epoxy used. Shrinkage and swelling properties and modulus of elasticity of wood and epoxy differ, and
these differences must be considered in large volume epoxy fillings. Consideration should be given to the
suitability of compression area epoxy repairs if there will be exposure to alternating wetting and drying
cycles. Some epoxy formulations can provide adequate bearing in areas of compression at bolts and
shear connectors and can be used to repair such areas.

Tension repair
The use of epoxy PC is not recommended for the repair of wood stressed in tension parallel to grain,
whether occurring on the tension side of bending members or for wood stressed in axial tension. Radial
reinforcement repairs for curved bending members can be accomplished by placing GFRP rods or rebars
in oversize holes filled with epoxy PC. In this type of repair the radial stresses are transmitted through the
epoxy PC in shear.

Shear repair
If longitudinal checks or splits in the wood occur that require re-establishment of the horizontal shear
strength of bending members, it is possible to use GFRP reinforcement in conjunction with epoxy PC for
this purpose in a similar fashion as described for radial tension repair. For horizontal shear reinforcement,
the FRP reinforcing elements in epoxy-filled holes act to transmit longitudinal shear forces between
adjacent beam sections. In all cases involving the use of mechanical reinforcement in combination with
epoxy adhesive, the net section remaining after the required holes for the steel are drilled should be used
when checking for resulting bending stresses.
Splits are openings which travel from one side of the timber to the opposite side or from one side and
through to the adjacent side. Seasoning checks can develop into splits under certain conditions. Splits
can also be caused by shrinkage of the wood in connection details with widely spaced bolts in tight-fitting
holes. Radial tension in curved members or notches or tapered cuts on the tension side of bending
members may also cause splits. The conditions causing the split should be determined and remedied.
This may require re-design of connection details, providing protection to keep the wood dry, or designing
appropriate radial reinforcement. Tight-fitting fasteners, such as fully-threaded lag screws or epoxy-

46 On Site Interventions On Decayed Beam Ends

 COST E34 - WG1: Bonding on Site

bonded GFRP bars, should be used in conjunction with epoxies in repairing splits, especially for long
splits. Epoxies without mechanical reinforcement are not recommended for repairing splits.

6.4 Applicable standards

As already evidenced in the above text, no specific standards exist at European level for the repair of old-
timber beam ends. For such reason, usually designers refer to the general approach described in ENV
1995-2:1997, valid only for the specific case of interventions carried out by using glued-in rods. Designing
with glued-in, not-nailed plates for example is still a sort of ‘challenge’ assigned to the skills or expertise of
the individual designer.
Nevertheless, some general rules exist for projects of structural interventions in the specific context of old
buildings that belong to the Cultural Heritage. The Italian organization for standardization has developed
a significant activity in this field and produced a standard [11]. A brief summary of the contents of this
standard, with a generic reference to the repair of beam ends, is therefore given in the following.
The standard gives some general indications on the criteria that should be followed for the preliminary
evaluation, the planning and the eventual interventions on wooden structures in old buildings. Some
preliminary verifications and evaluations have to be ever considered, including for example the state of
conservation and the service conditions for each wooden element of the structure by means of a
diagnostic survey. The accuracy of this diagnostic phase depends on the complexity of the structure and
on the requested level of detail. The preliminary procedures include:
ƒ historical analysis: if the original construction techniques or working abilities would be preserved,
this information could even suggest the choice of the eventual intervention
ƒ characterisation of materials: this may even include strength grading of wood members,
following specific standards
ƒ geometric characterization: for relieving the dimensions of structural members but also for
having a geometric survey of joints (carpentry or mechanical devices) and of deformations
ƒ characterisation of decay: the possible interactions between decay and service environment
have to be also analysed (for example, the heads of beams inserted in walls)
ƒ structural analysis: the evaluation of the static conditions and of the state of stress is considered
as mandatory. In the definition of the structural scheme an addition to the static scheme of single
timber members, including those of the structural units and joints etc have to be considered. In
order to identify the contribution of the timber structure in relation to the static performance of the
building. Only in simple cases, for example the substitution of the decayed beam end is it
possible to refer to models with single timber elements whilst neglecting the overall static
analysis. The preliminary structural analysis includes the verification of the repaired elements of
structure in order to demonstrate that timber replacement can guarantee an acceptable level of
safety.
In respect of the planning of interventions, several aspects of the conservation project need to be
considered, including some recommendations for a caution on the compatibility between the used
materials and for the correct periodic maintenance works (type and space time after intervention have to
be clearly fixed). The fact that repair is not even necessary is also evidenced, and in any case the
concept of “reversibility” is considered as a tendency which should be preferably pursued.
The project has to respect, whenever possible and if the original configuration is compatible with actual
levels of safety, the pre-existing overall static model. Therefore, for example, joints and constraints have
to be restored by maintaining their original stiffness (unless improper) and longitudinal shrinkage splits not
treated in any way that could prevent the transversal movements of wood. Emphasis is placed on the fact
that important modifications to the timber frame have to undergo an overall static analysis in order to
include the interactions between wooden and non-wooden elements. This induces an adequate
justification of interventions bringing to a significant alteration of the original overall static performance.
Caution in the choice of the relevant safety factors to apply on loads is also recommended, by
considering the importance of the structure and the consequent acceptable level of risk. In this context,
interventions on ends of beams are considered as “reintegration” of missing timber, and hence the use of
materials other than wood (for example epoxy polymer concrete or grouts) has to be evaluated very
carefully and is not generally recommended. Moreover, adhesives specifically formulated for structural
use with wood are only permitted. Conversely, “indirect” improving expedients such as facilitating micro-
climatic exchanges with the surroundings, mainly at the beam ends, are strongly favoured. In case of

On Site Interventions On Decayed Beam Ends 47

COST E34 - WG1: Bonding on Site

using permanently pre-stressed elements, it is necessary for an evaluation of the performance of the
timber structure in terms of long-term duration and, additionally, a plan of the procedures for future
inspections have to be described. A check on the efficacy of a proposed intervention is required before its
implementation, either through experimentation, numerical simulation or by some factors that are of
unique knowledge to the individual case. Such an approach can be also used in order to qualify new
products and/or methods of intervention. Finally, requirements are provided regarding the periodic
inspection of the interventions which have to be clearly specified (by indicating typology, time and space)
and reported in a maintenance plan. The inspections serve the purpose of confirming the compliance with
the specifications of the plan. This includes factors concerning ambient conditions, permanent and/or
moving loads, and a visual examination with the aim of relieving new or accentuated damage on single
elements, with a special attention to the possibility of new biotic attacks. Additionally, the results of these
inspections have to be registered in a specific paragraph of the maintenance plan.

6.5 Research needs

Research in the field of repair of ancient timber structures involves different aspects, most of them
overlapping with the relevant topics of other issues presented in this document (e.g. glued-in rods, timber
reinforcement etc.). Particularly important are:
ƒ Elaboration of a specific design concept for on site applications of old timber structures, including
consideration of long-term behaviour of connections
ƒ Further improvement of structural wood adhesives designed for on site repair with the following
characteristics:
ƒ easy-to-use and apply on site
ƒ optimised compatibility for applications in wood
ƒ curing at room temperature but characterised by values of Tg higher than 60-70°C
ƒ high values of interfacial fracture toughness, such that a high long-term durability could be
guaranteed
ƒ values of shear strength and modulus of elasticity comparable to those of the actual
commercially available products, normally characterised by good creep behaviour.
ƒ Evaluation of the quality of bonding for existing and repaired structures by NDT techniques

48 On Site Interventions On Decayed Beam Ends

 COST E34 - WG1: Bonding on Site

6.6 References

[1] Kuipers, J. 'Effect of age and/or load on timber strength'. CIB W18, Meeting 18, Florence, Italy,
1986,
[2] Berti, S., Pizzo, B., Lauriola, M., Lavisci, P. and Rizzo, G. 'Consolidamento con adesivi epossidici
delle strutture lignee. Parte 1: Indagine sulla durabilità degli interventi', Arkos 3 (1) (2002) 32-37
[3] Pizzo, B., Lavisci, P., Macchioni, N. and Misani, C. (2002) 'A new approach to the design of epoxy
adhesives for the restoration of timber structures'. In: (eds) European Timber Buildings as an
Expression of the Technological and Technical Cultures. Elsevier, Heritage series, Paris, 245-254.
[4] Schober, K.U. and Rautenstrauch, K. 'Upgrading and repair of timber structures with polymer
concrete facing and strengthening'. WCTE, Portland, USA, 2006,
[5] Stumes, P. 'The W.E.R.-System Manual: structural rehabilitation of deteriorated timber',
Association for Preservation Technology Pubs, 1979
[6] Waterhouse, M. 'The restoration of structural timbers'. Annual Convention of British Wood
Preserving Association, 1981, 26-31
[7] Mettem, C.J. and Davis, G. 'Resin bonded repair systems for structural timber', Construction
Repair March/April (1996)
[8] Lauriola, M. (2002) 'Esempi di calcolo secondo i criteri dell’Eurocodice 5'. In: Mapei (eds) Il
restauro delle strutture lignee. 19-25.
[9] Anonymus 'Low Intrusion Conservation Systems for Timber Structures.' 2006, Website:
http://www.licons.org/
[10] Pizzo, B., Lavisci, P., Macchioni, N. and De Ciechi, M. (2004) 'On site consolidation systems of old
timber structures”'. In: (eds) Interaction between Science, Technology and Architecture in Timber
Construction. Elsevier Heritage Series, Paris, 323-352.
[11] UNI 11138 'Cultural Heritage. Wooden Artefacts. Load bearing structures of buildings: Criteria for
the preliminary evaluation, the design and the execution of interventions' (2004)

On Site Interventions On Decayed Beam Ends 49

View publication stats