3 knowledge of different roofing systems in the ... · 14500 13500 . knowledge of different roofing...

12
J. Civil Eng. Architect. Res. Vol. 2, No. 4, 2015, pp. 593-604 Received: February 28, 2015; Published: April 25, 2015 Journal of Civil Engineering and Architecture Research Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation Francesco Augelli 1 , Anna Anzani 2 , Lorenzo Cantini 3 , Paola Condoleo 4 , Antonia Gobbo 2 and Roberta Mastropirro 1 1. Department of Architecture and Urban Studies, Politecnico di Milano, Milan 20158, Italy 2. Department of Design, Politecnico di Milano, Milan 20158, Italy 3. Department of Architecture, Built Environment and Construction Engineering, Politecnico di Milano, Milan 20133, Italy 4. Department of Civil and Environmental Engineering, Politecnico di Milano, Milan 20133, Italy Corresponding author: Francesco Augelli ([email protected]) Abstract: The paper presents a study carried out on the roofing system of Politecnico di Milano main Campus, a public complex that was built at early XX century and is a relevant part of a valuable architectural heritage. The research was aimed at its structural and health assessment as a first step for designing its subsequent safeguard intervention. The adopted method of analysis consisted of different phases that started from a photographic and 3D direct geometric survey of the trusses that span over large rooms and sometimes bear brickwork vaults hanging below. Material, morphological and constructive aspects of the roofing systems were analyzed also comparing the different building solutions adopted in the different parts of the construction; a recognition of the types of connections between the hanging devices and timber trusses was carried out as well. According to the current Technical Codes, direct inspection of the structural elements was performed in order to collect the following information: macroscopic determination of the timber species, determination of their natural durability, individuation and technological evaluation of the principal defects and alterations, individuation of the pre-existing or still present macroscopic pathologies, sample detection of the ambient conditions and of the timber humidity through non-destructive measurement by electronic thermo-hygrometer. Where needed, drilling tests were applied in the deteriorated or suspected zones and an estimate of the residual sections was achieved, as well as the identification of two different use classes. Considerations on the reliability and recoverability of the inspected timber components were then made possible. Key words: Roofing systems, timber trusses, connection devices, material pathology, diagnostic techniques, rot fungi attack, drilling tests, preservation. 1. Introduction An articulated procedure has been applied for the structural and health assessment of the roofing systems of Politecnico di Milano main Campus (Fig. 1), a public complex built at the beginning of the XX century. The research was aimed to designing a detailed program of maintenance and safeguard of a relevant part of a valuable architectural heritage. The very complex constructive system characterized by trusses of different typologies spanning over large rooms was analyzed through a careful photographic and 3D direct geometric survey [1]. Most of the trusses were built in timber, in few cases in reinforced concrete. Above the main rooms, more complex trusses of bigger size are present. In many cases, the timber trusses are bearing extremely deformable brickwork vaulted structures hanging below. They are characterized by a nearly flat central sector, supported by iron beams, the presence of

Upload: vokien

Post on 09-Sep-2018

216 views

Category:

Documents


0 download

TRANSCRIPT

J. Civil Eng. Architect. Res. Vol. 2, No. 4, 2015, pp. 593-604 Received: February 28, 2015; Published: April 25, 2015

Journal of Civil Engineering

and Architecture Research

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

Francesco Augelli1, Anna Anzani2, Lorenzo Cantini3, Paola Condoleo4, Antonia Gobbo2 and Roberta Mastropirro1

1. Department of Architecture and Urban Studies, Politecnico di Milano, Milan 20158, Italy

2. Department of Design, Politecnico di Milano, Milan 20158, Italy

3. Department of Architecture, Built Environment and Construction Engineering, Politecnico di Milano, Milan 20133, Italy

4. Department of Civil and Environmental Engineering, Politecnico di Milano, Milan 20133, Italy

Corresponding author: Francesco Augelli ([email protected])

Abstract: The paper presents a study carried out on the roofing system of Politecnico di Milano main Campus, a public complex that was built at early XX century and is a relevant part of a valuable architectural heritage. The research was aimed at its structural and health assessment as a first step for designing its subsequent safeguard intervention. The adopted method of analysis consisted of different phases that started from a photographic and 3D direct geometric survey of the trusses that span over large rooms and sometimes bear brickwork vaults hanging below. Material, morphological and constructive aspects of the roofing systems were analyzed also comparing the different building solutions adopted in the different parts of the construction; a recognition of the types of connections between the hanging devices and timber trusses was carried out as well. According to the current Technical Codes, direct inspection of the structural elements was performed in order to collect the following information: macroscopic determination of the timber species, determination of their natural durability, individuation and technological evaluation of the principal defects and alterations, individuation of the pre-existing or still present macroscopic pathologies, sample detection of the ambient conditions and of the timber humidity through non-destructive measurement by electronic thermo-hygrometer. Where needed, drilling tests were applied in the deteriorated or suspected zones and an estimate of the residual sections was achieved, as well as the identification of two different use classes. Considerations on the reliability and recoverability of the inspected timber components were then made possible. Key words: Roofing systems, timber trusses, connection devices, material pathology, diagnostic techniques, rot fungi attack, drilling tests, preservation.

1. Introduction

An articulated procedure has been applied for the

structural and health assessment of the roofing

systems of Politecnico di Milano main Campus (Fig.

1), a public complex built at the beginning of the XX

century. The research was aimed to designing a

detailed program of maintenance and safeguard of a

relevant part of a valuable architectural heritage.

The very complex constructive system characterized

by trusses of different typologies spanning over large

rooms was analyzed through a careful photographic

and 3D direct geometric survey [1].

Most of the trusses were built in timber, in few

cases in reinforced concrete. Above the main rooms,

more complex trusses of bigger size are present. In

many cases, the timber trusses are bearing extremely

deformable brickwork vaulted structures hanging

below. They are characterized by a nearly flat central

sector, supported by iron beams, the presence of

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

594

buttress walls on the extrados, and special devices that

connect the vaults with the roofing trusses [2].

A comparison between different building solutions

adopted in different parts of the construction and a

recognition of the types of connections between the

hanging devices and the timber trusses has been

carried out.

Direct inspection of the structural elements has been

performed in order to identify the timber species, their

principal defects, macroscopic pathologies and class

of biological use.

All the analyzed elements turned out to be in Larch

and present a surface treatment, presumably creosote,

which gives to the surface a quite dark color. Among

the objectives of the study there was the assessment of

the conservation state of truss elements, the estimate

their residual section and the quantification of possible

damage due to the observed pathologies. Based on the

adoption of a gimlet to test the timber consistency and

of a chisel to locally evaluate the aggression depth, an

initial evaluation of the decayed zones was carried out,

followed by a drilling characterization of the wooden

elements that showed evident damage. On these

elements, the ratio between the calculated resistant

profile (Pr) and the total measured profile (Pt) allowed

to highlight two use classes [3].

Fig. 1 Plan of the studied complex.

Critical situations have been pointed out through

both visual inspection and drilling test results, which

will be useful to suggest a maintenance and safeguard

program.

2. Geometrical Survey

The very complex constructive system

characterized by trusses of different typologies was

analyzed through a careful photographic and 3D direct

geometric survey. The geometry of every attic was

therefore traced to understand the behavior of the

structural elements and their connections.

Most of them are built in timber, in few cases in

reinforced concrete.

Above the main rooms, more complex trusses of

bigger size are present.

In building 3_a1, truss E3_a1-C93 (Fig. 2) consists

of a timber-iron system and is placed very close to

truss E3_a1-C104(Fig. 3) from which it is distant only

0.52 m.

It presents an iron tie sustained by four metal

elements, two of which are connected with the upper

horizontal tie and the other two are connected with the

diagonal struts. Details are shown in Figs. 4 and 5,

Similar trusses with a metal tie were used in building

7 (Fig. 6). In building 6, different bodies are

characterized by different trusses: in wing 6_a1 (Fig. 7)

the structures are characterized by the presence of

horizontal metal ties, whereas in wing 6_a2 (Fig. 8)

trusses of different dimensions with horizontal timber

ties were used.

Fig. 2 Truss E3_a1-C93 in building 3_a1.

Fig. 3 Truss E3_a1-C104 in building 3_a1.

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

595

In building 9 only three trusses are present whereas

in building 5 (Fig. 9) trusses with simplified geometry

can be observed.

Differently from other roofs, those in buildings

4_a1 and 8 are characterized by trusses with double

elements (Figs. 10 and 11).

Fig. 4 Truss E3_a1-C93 in building 3.

Fig. 5 Truss E3_a1-C104 in building 3.

Fig. 6 Trusses in building 7.

Fig. 7 Trusses in building 6 a1.

Fig. 8 Trusses in building 6 a2.

Fig. 9 Roofs in building 5 a1-5 a_2: View of the lower (a) and higher (b) simplified parts of the structures.

b

a

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

596

Fig. 10 Trusses in building 4.

Fig. 11 Trusses in building 8.

2.1 Connection between Roofs and Vaults

In many cases, the timber trusses of the studied

roofing systems span over large rooms and bear brick

work-vaulted structures hanging below. Figs. 12 and

13 show these extremely deformable large structures,

characterized by a nearly flat central sector, supported

by iron beams, the presence of buttress walls on the

extrados, and of hanging devices that connect the

vaults with the roofing trusses [2].

3. Inspection on Timber Elements

The state of conservation of the trusses of all

buildings, including secondary structural elements,

was assessed through inspections and non-destructive

testing. In each attic a careful inspection of all the

elements of any single truss was carried out and

possible defects, attacks of microorganisms and the

fibers characteristics were individuated [3].

According to the results of the inspection, some

points were chosen where drilling tests had to be

carried out, especially near the connections between

timber elements and walls. In building 3, since the

roof suffered for the failure of a support due to rot

caused by water absorption, many tests were carried

out close to the supports of every truss. In other

buildings, tests were only performed where the

inspections highlighted anomalies in the timber

elements.

Fig. 12 Intrados of a vault.

Fig. 13 Extrados of a vault.

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

597

3.1 Purpose and Procedure of the Inspection Phase

The purpose of the inspections was the assessment

of the state of conservation and the estimate of the

performances of the timber elements through

nondestructive methods and techniques [4].

According to the current Italian and European

Technical Codes [5-26], the following information has

been collected: macroscopic determination of the

timber species, determination of their natural

durability; individuation and technological evaluation

of the principal defects and alterations; individuation

of the pre-existing or still present macroscopic

pathologies; sample detection of the ambient

conditions and of the timber humidity through

non-destructive measurement by electronic

thermo-hygrometer; estimate of the residual strength

in the deteriorated or suspected zones (mainly in the

supports and connection points) through drilling

technique, estimate of the residual sections and

considerations on the reliability and recoverability of

the inspected timber components. All the accessible

surfaces of the timber components were therefore

examined.

3.2 Identification of the Timber Species

The individuation of the timber species has been

performed from a macroscopic point of view,

according to UNI 11118:2004 [17] and UNI

11119:2004 [18], through the recognition of the

presence or the absence of anatomic elements that are

visible to the naked eye or with a 10× magnifying lens:

porosity or chromatic differentiation of the growth

rings; differentiation or non-differentiation between

sapwood and heartwood; visibility, shape and

dimensions of the medullar rays; presence of resin

and/or resin pockets; characteristic smell, color and

texture of wood, etc. These elements are sometimes so

clear that the species, or at least the order, gender and

family recognition is very reliable. It has to be noted

that, when carrying out the macroscopic timber

individuation and, subsequently, the visual inspection

of the defects of anatomic constitution, the presence of

deposits and of variously deteriorated surface

treatments are very disturbing [27].

In the analyzed case, all the elements have a surface

treatment, presumably creosote, which gives to the

surface a quite dark color. All the inspected elements

turned out to be in Larch.

3.3 Individuation and Technological Evaluation of the

Main Defects

During the survey, according to UNI 11119: 2004

[18] critical zones have been individuated, i.e. those

longitudinal portion of an element not shorter than 150

cm where relevant defects of the element reliability

turn out to be visible. Through the individuation of the

defects, a classification according to the strength has

been formulated [28].

3.4 Classification according to the Strength

From the Italian Code of Practice UNI 11119:2004

[18] the following table has been extracted

thatindicates the value of allowable tensions, for in

work categories of the principle timber species,

applicable for a timber humidity of 12%.

3.5 Determination of the Natural Durability

In the studied complex, the natural durability of

the elements in work was determined according to the

Table 1 From Italian Code UNI 11119:2004 [18] applicable for timber humidity of 12%.

Maximum stress (N/mm2)

Species In work category

Compression Static bending

Tension parallel to the fibres

Shear perpendicular to the fibres

Flexural modulus of elasticity Parallel to the

fibres Perpendicular to the fibres

Larch (Larix spp.)

I II II

12 10 7.5

2.5 2.2 2.0

13 11 8.5

12 9.5 7

1.1 1.0 0.9

15500 14500 13500

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

598

Italian Code UNI EN 335-1:2006 [7] and UNI EN

335-2: 2006 [8]. In the case of Larch it is as follows:

Moderately durable to the fungi rot attacks; not

resistant to the attacks by wood borers beetles

(Anobium punctatum and Hylotrupes bajulus).

3.6 Individuation of Present and Previous

Macroscopic Pathologies

Besides the assessment of the conservation state of

single truss elements (primary and secondary struts and

ties) and the estimate of their residual section, main

objective of the study was also to inspect, where

possible, the structural nodes between main struts and

main ties close to the connection to the walls and to

quantify possible damage due to the observed

pathologies. Also in this case the codes UNI

11119:2004 [18] and UNI 11130:2004 [26] were

followed. On the structural nodes, where the

macroscopic examination let one suspect the presence

of current or previous infections but was not sufficient

to achieve a complete diagnosis, as well as on each

accessible wall support, preliminary beating tests were

performed, aimed at detect fungi attacks [27]. For an

initial evaluation and to quantify the decayed zones, a

gimlet to test the timber consistency and a chisel to

locally evaluate the depth of the aggression were

adopted. The results of this phase were determinant for

the subsequent choice of the points where drilling test

were required [28].

In the cases where significant biological decay was

detected, the assumed load-bearing section was

conveniently reduced with respect to the real geometry.

In particular, it has to be noted that in building 3 on

struts of truss E3_a1-C115 corresponding to the East

masonry support and on strut E3_a1-FP10 a severe

fungi attack was detected.

4. Non-destructive Drilling Tests

The tests carried out on the timber elements of the

roof were aimed to assess the material density related to

the humidity and to the timber species. Initially, the

ambient and local thermo-hygrometric conditions in the

testing points were recorded (Fig. 14a). Then, only in

some representative positions, ultrasonic

characterization was carried out in cross-section mode

(Fig. 14b). Finally, drilling tests were performed using a

penetrometer (IML RESIF400®) (Figs. 14c and 14d),

that measured the resistance opposed by timber to the

penetration of a pin which rotates and penetrates at

constant speed [29]. This parameter is related to the

material density, which of course is influenced by the

presence of knots, irregularities, decay, etc. For each

structure presenting symptoms of decay, different

profiles located at the ends and at midpoint of each

principal element were usually collected [30]. If

necessary, also secondary elements were tested as

shown in Fig. 15a. As an example, a profile obtained on

the end of a horizontal element is shown in Fig. 15b.

ba

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

599

Fig. 14 Tests methods carried out on the timber roof: thermo-hygrometric survey (a), ultrasonic test (b), drilling test on a truss loaded in tension (c) and in compression (d).

Fig. 15 Drilling tests: localization of the tests (a); amplitude vs. depth profile (b).

In general very few profiles, localized at the ends of

the diagonal elements, showed internal discontinuities.

In these cases, the profile obtained with the drilling

tests was used to estimate the real working section of

the timber structure.

The graph provided by the device shows a curve

where the horizontal axis reports the penetration depth

reached by the pin during the tests, while the vertical

axis reports the percentage amplitude used for the

penetration. Low values of amplitude can be related to

a smaller density of the material: this condition, in a

wooden structure, is typical of decayed portions of the

material that lost their integrity due to the presence of

water, of parasites insects and micro-organisms. Once

evaluated the extension of the decayed portions, an

estimate of the working section can be obtained as the

difference between the whole dimension of the tested

section and that of the decayed portions [31].

In Figs. 16-18 the plan of the attic of Building 3 and

the localization of the drilling tests carried out on the

roof structure covering the big vaulted room are

shown.

dc

b

a

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

600

Referring again to Building 3, in Fig. 19 the results

obtained on a sound element not showing decay

phenomena are reported as a comparison.

In Figs. 20-22 diagrams obtained on severely

damaged elements are shown.

In Table 2 the results of drilling tests carried out on

the roofs of Building 3 are reported as an example. The

ratio between the calculated resistant profile (Pr) and

the total measured profile (Pt) has been considered:

when Pr/Pt > 0.7 use class A has been indicated, when

Pr/Pt < 0.7 use class B has been indicated. Numerous

drilling tests were carried out on the elements where

the inspection indicated a severe fungi attack. It has to

be noted that in the cases of elements E3_a1-C115-P13,

E3_a1-C115-P14, E3_a1-C115-P16, E3_a1-C115-P18,

E3_a1-FP10-P24, E3_a1-FP10-P25 (Table 2) the ratio

Pr/Pt was equal or lower than 0.3, indicating a

particularly severe safety condition.

Fig. 16 Plan of the attic of Building 3: A) vaulted ceiling underneath the roof; B) plane ceiling underneath the roof.

2 5 m0

E 3_a1-C71-P11

E 3_a1-C71-P12

E 3_a1-C82-P10E 3_a1-C82-P8

E 3_a1-C82-P9

E 3_a1-C93-P28E 3_a1-C93-P29

E 3_a1-C93-P4

E 3_a1-C93-P5

E 3_a1-C104-P2E 3_a1-C104-P3

E 3_a1-C104-P1

E 3_a1-C71-P13

E 3_a1-C71-P14E 3_a1-C71-P24

E 3_a1-C71-P23E 3_a1-C82-P15E 3_a1-C82-P16

E 3_a1-C82-P22E 3_a1-C93-P17

E 3_a1-C93-P18E 3_a1-C93-P19

E 3_a1-C104-P20E 3_a1-C104-P21

E 3_a1-FP12-P27E 3_a1-FP10-P26

E 3_a1-FP10-P24E 3_a1-FP10-P25E 3_a1-C115-P19

E 3_a1-C115-P20

E 3_a1-C115-P21

E 3_a1-C115-P18

E 3_a1-C115-P22

E 3_a1-C115-P15

E 3_a1-C115-P23

E 3_a1-C115-P16

E 3_a1-C115-P17

E 3_a1-C115-P12E 3_a1-C115-P13E 3_a1-C115-P14

E 3_a1-C71-P25

E 3_a1-C71-P26E 3_a1-C71-P27

E 3_a1-C93-P6

E 3_a1-C93-P7

Fig. 17 Roof of the big room of Building 3: localization of the drilling tests.

CAPRIATA E3_a1-C115 - lato EST

0 1 m 2 m

E3_a1-C11-P17

E3_a1-C11-P16

E3_a1-C115-P15

E3_a1-C115-P14E3_a1-C115-P13

E3_a1-C115-P22

E3_a1-C115-P23

E3_a1-C115-P19

E3_a1-C115-P18

E3_a1-C115-P20

E3_a1-C115-P21

Fig. 18 Building 3, truss E3_a1- C11: localization of the drilling tests.

Fig. 19 Test E3_a1-FP12-P27 carried out on a strut of dimensions 18 × 23 cm. Distance from the support 2 cm, T 13 °C, R.H. 66%, α = 45°, penetration depth 30.5 cm, estimated resistant length 30.2 cm.

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

601

Fig. 20 Test E3_a1-C115-P16 carried out on a strut of dimensions 21 × 25 cm. Distance from the support 48 cm, T 13 °C, ambient R.H. 66%, timber R.H. 5.4%, α = 0°, penetration depth 21.4 cm, estimated resistant length 4.6 cm.

Fig. 21 Test E3_a1-C115-P18 carried out on a tie of dimensions 21 × 25.5 cm. Distance from the support 6 cm, T 13 °C, ambient R.H. 66%, timber R.H. 4.9 %, α = 45°, penetration depth 9.3 cm, estimated resistant length 2.2 cm.

Fig. 22 Test E3_a1- FP10-P25 carried out on a tie of dimensions 18 × 23 cm. Distance from the support 1 cm, T 13 °C, ambient R.H. 66%, timber R.H. 4.3 %, α = 0°, penetration depth 23 cm, estimated resistant length 7.3 cm.

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

602

Table 2 Example of use class estimate carried out on the roof of Building 3.

Test denominated Angle α Pt (cm) Pr (cm) Pr/Pt Use class

E3_a1- C71-P26 45° 27.0 27.0 1 A

E3_a1- C71-P14 45° 28.4 24.7 0.86 A

E3_a1- C93-P29 0° 24.3 16.8 0.69 B

E3_a1- C93-P19 45° 28.23 18.3 0.64 B

E3_a1-FP10-P26 0° 23 15.3 0.66 B

E3_a1- C93-P28 0° 24.67 22.67 0.91 A

E3_a1-FP12-P27 45° 30.5 30.2 0.99 A

α = angle between the penetration trajectory and the horizontal direction; Pt = penetration depth; Pr = estimated resistant profile obtainedby subtraction from Pt of the decayed portions identified by the results of drilling test; Use class A: Pr/Pt> 0.7; Use classB: Pr/Pt< 0.7.

5. Conclusions

An articulated procedure has been applied for the

structural and health assessment of the roofing

systems of Politecnico di Milano main Campus, a

public complex built at the beginning of XX century.

The research was aimed to designing a safeguard

intervention of a relevant part of a valuable

architectural heritage [30]. Direct inspection of the

structural elements allowed to identify the timber

species, their principal defects, macroscopic

pathologies and class of biological use. Where needed,

drilling characterization of the wooden elements with

evidence of damage was carried out and the ratio

between the calculated resistant profile (Pr) and the

total measured profile (Pt) allowed to highlight two

use classes.

Critical situations have been evidentiated through

both visual inspection and drilling profiles. In

particular, a partial failure of the secondary strut S17

in truss E7_C174 of building 7 have been noticed, that

provoked also a drop of the vertical tie (Figs. 23-26).

In building 3, on struts of truss E3_a1-C115

corresponding to the East masonry support and on

strut E3_a1-FP10 a severe fungi attack was detected.

The drilling tests carried out on the same elements

highlighted that in six cases the ratio Pr/Pt was equal

or lower than 0.3, which indicates a particularly severe

safety condition.

E 7C101

E 7C111

E 7C121

E 7C131

E 7C163

E 7C174

2 5 m0

Fig. 23 Building 7, Truss E7-C174, plan.

Fig. 24 Building 7, Truss E7-C174, plan.

Fig. 23 Partial failure of the secondary strut S17,truss E7-C174.

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

603

Fig. 23 Drop of the vertical tie, truss E7-C174.

Acknowledgment

The technicians of the Material testing Laboratory

of Politecnico di Milano M. Antico, M. Cucchi, C.

Tiraboschi and the students of the Faculty of

Architecture J. Migliavacca and L. Ronchi are

gratefully acknowledged for their fundamental support

in the in-situ survey and laboratory testing, C. Arcadi

for her helpful editing contribution.

References

[1] F. Augelli, Wooden cultural heritage. Guidelines, standards and methods of representation, National University of Architecture & Construction of Armenia 2 (53) (2014) 21-33

[2] A. Anzani, P. Condoleo, A. Gobbo, A. Taliercio, Modeling the static behaviour of a double curvature brickwork vault, Advanced Materials Research 133-134 (2010) 367-372.

[3] F. Augelli, The diagnosis of woodworks and timber structures, The inspections, Xilema Series1, II Prato, Saonara (PD), 2006. (in Italian)

[4] M. Riggio, R.W. Anthony, F. Augelli, B. Kasal, T. Lechner, W. Muller, T. Tannert, In situ assessment of structural timber using non-destructive techniques, Rilem, Materials and Structures, Springer 47 (5) (2014) 749-766.

[5] UNI 8829: 1986 Sawing timber, Determination of moisture gradient.

[6] UNI 9091-1: 1989 Wood, Determination of moisture content. Electric method.

[7] UNI EN 335-1: 2006 Durability of wood and wood-based products, Definition of uses classes, Part 1: General.

[8] UNI EN 335-2: 2006 Durability of wood and wood-based products, Definition of uses classes, Part 2: Application to solid wood.

[9] UNI EN 1995-1-1: 2014 Eurocode 5, Design of timber structures, Part 1-1: General-Common rules and rules for buildings.

[10] UNI EN 14081-1: 2011 Timber structures, Strength graded structural timber with rectangular cross section, Part 1: General requirements.

[11] UNI EN 844-4: 1999 Round and sawn timber, Terminology, Terms relating to moisture content.

[12] UNI EN 844-8: 1999 Round and sawn timber, Terminology, Terms relating to features of round timber.

[13] UNI 10969: 2002 Cultural heritage, General principles for the choice and the control of the microclimate to preserve cultural heritage in indoor environments.

[14] UNI 11035-1: 2010 Structural timber, Visual strength grading for structural timbers, Part 1: Terminology and measurements of features.

[15] UNI 11035-2: 2010 Structural timber, Visual strength grading for structural timbers, Part 2: Visual strength grading rules and characteristics values for structural timber population.

[16] UNI EN 13183-2: 2003 Moisture content of a piece of sawn timber, Estimate by electrical resistance method.

[17] UNI 11118: 2004 Cultural heritage, Wooden artefacts, Criteria for the identification of the wood species.

[18] UNI 11119: 2004 Cultural heritage, Wooden artefacts, Load-bearing structures-On site inspections for the diagnosis of timber members.

[19] UNI 11120: 2004 Cultural heritage, Field measurement of the air temperature and the surface temperature of objects.

[20] UNI 11138: 2004 Cultural heritage, Wooden artefacts, Building load bearing structures-Criteria for the preliminary evaluation, the designand the execution of works.

[21] UNI 11131: 2005 Cultural heritage, Field measurement of the air humidity.

[22] UNI 11161: 2005 Cultural heritage. Wooden artefacts, Guideline for conservation, restoration and maintenance.

[23] UNI 11202: 2007 Cultural heritage. Wooden artefacts, Determination and classification of environmental conditions.

[24] UNI 11203: 2007 Cultural heritage, Wooden artefacts, Load-bearing structures-Terms and definitions of the structural configurations and of the constituent elements.

[25] UNI 11204: 2007 Cultural heritage, Wooden artefacts, Determination of moisture content.

[26] UNI 11130: 2004 Cultural heritage, Wooden artefacts, Terms and definitions concerning wood deterioration.

Knowledge of Different Roofing Systems in the Politecnico di Milano: Geometric Survey and Material Investigation

604

[27] F. Augelli, C. Colla, R. Mastropirro, Inspection & NDT to verify structural reliability of historic wooden roofs in the ex-Meroni spinning-mill, in: SAHC, Structural analysis of historical constructions, Padova, November 10-13, Vol. 1, 2006, pp. 377-386.

[28] F. Augelli, A. Grimoldi, L. Jurina, D. Meroni, S. Zanzani, Silva-Persichelli Palace in Cremona, Diagnostics and cataloguing of the timber elements for the preservation and strengthening design, in: Proceedings of International Conference on the Conservation of the Wooden Structures, Collegiodegli Ingegneri della Toscana, UNESCO, Vol. I, Florence, February 22-27, 2005, pp. 307-321. (in Italian)

[29] A. Anzani, L. Binda, L. Cantini, G. Cardani, P. Condoleo, G.E. Massetti, The basilica of S. Lorenzo in cremona:

Structural investigation and monitoring, in: Proceedings of the10th Tenth North American Masonry Conference, 2007, pp. 1088-1099.

[30] L. Binda, L. Cantini, P. Condoleo, A. Saisi, Non destructive testing techniques applied to the masonry and timber structures of the Crocifisso Church in Noto, in: Proceedings of the 10th International Conference on Structural Studies, Repairs and Maintenance of Heritage Architecture, 2007, pp. 237-248.

[31] L. Binda, L. Cantini, C. Tiraboschi, C. Amigoni, Not destructive investigation for the conservation design of a Monastery near Bergamo (Italy), in: Proceedings of the 1st International RILEM Symposium on Site Assessment of Concrete, Masonry and Timber Structures, 2008, pp. 797-806.