This paper focuses on the study of a technique for recovering damaged wooden beams with the aid of structural adhesives. The technique being studied is based on the realization of a laminated wood prosthesis that will replace the part of the degraded beam, by using only a suitable structural adhesive as connection element. The paper takes up the research previously carried out by prof. Marco Andrea Pisani and the colleagues Eng. Antelmi and Eng. Pisacane who deal with the study of wood as an essence and as a structural material, its mechanical properties and the degradation mechanisms; on the research for an effective surface treatment of the glued wood and the type of glue to be used for this purpose. These studies have led to choosing oak as the essence and LOCTITE UK 1351 B25 / UK 5452 as a glue to be used in subsequent experimental tests. The research continues with the conception of a suitable test setup to be able to test the chosen material which consists of two metal confinement plates which, combined with the specimens, avoid out of plane movements and aims to calibrate a contact law between the two bonded surfaces. In the previous papers, by testing gluing lengths of 30, 60, 90, 120, 150 and 180 mm, a local bond-slip model was calibrated in the line parallel to fiber direction. The purpose of this paper is to focus on: calibration of the local bond-slip model of contact between two wooden bonded surfaces in the line orthogonal to fiber direction, because the orthotropy of the material does not allow the study of the local bond-slip model in only one direction; on the verification of the local bond-slip model in the line orthogonal to fiber direction through finite element models and on the application of the local bond-slip model obtained on the head of a timber beam consolidated with laminated wood. The calibration of the local bond-slip model in the line orthogonal to fiber direction was carried out through experimental tests with the help of the test setup, studied in the paper of Eng. Pisacane, and the new non-contact measurement technique DIC (Digital Image Correlation) which allows to determine, through digital images, the displacement of the specimens with time and load variations. The bonding was tested by tensile tests performed on two 76x50x8 mm first class oak strips glued together. The type of test chosen falls on the single lap shear of traction and the setup previously studied has been readapted for the new dimension of the specimens. Six glued specimens were tested for each adhesion length and three different bonding lengths were chosen, namely 10, 20 and 30mm, for a total of eighteen samples analyzed. Furthermore, three tensile tests were carried out on individual strips, to determine the strength and the modulus of elasticity. The study of the results allowed to obtain a local bond-slip model in terms of interfacial shear stress-slip (τ-s) between the two bonded surfaces. The calibrated model, in both directions, has been validated by simulation on finite element software, Abaqus, in which the strips tested in the laboratory have been shaped. After the local bond-slip model, using the publication "Stress Transfer and Fracture Propagation in Different Kinds of Adhesive Joints" (Zhishen Wu; Hong Yuan; and Hedong Niu) the minimum bonding length to be used was calculated. The local bond-slip model and minimum bonding length obtained were then used to model the beams tested on the software by the Eng. Antelmi in her paper. From the modeling results that the length used, obtained from the previous calculations, was not enough. This will lead the analysis on the effectiveness of the technique used today, which involves the consolidation of the head of the wooden beam by gluing vertically the arranged strips, and an alternative, more performing will be proposed: the consolidation of the beam wooden head by gluing horizontally arranged strips. With the aim of confirming the best effectiveness of the proposed technique, the two consolidation methods will be compared by modeling on the software Abaqus, comparing the tangential stresses present on the bonding surfaces. Finally, the new technique will be examined in depth on the issues concerning the number and length of bonding of the strips to be used for consolidation.
Questo elaborato di tesi è incentrato sullo studio di una tecnica di recupero delle travi lignee ammalorate tramite l’ausilio di adesivi strutturali. La tecnica in corso di studio si basa sulla realizzazione di una protesi di legno lamellare che sostituirà la parte della trave degradata, mediante il solo utilizzo di un idoneo adesivo strutturale come elemento di connessione. L’elaborato riprende le ricerche precedentemente svolte dal prof. Marco Andrea Pisani e dai colleghi Ing. Antelmi ed Ing. Pisacane le quali vertono sullo studio del legno come essenza e come materiale strutturale, le sue proprietà meccaniche ed i meccanismi di degrado; sulla ricerca di un trattamento superficiale efficace del legno sottoposto ad incollaggio e la tipologia di colla da utilizzare per questo fine. Questi studi hanno portato a scegliere il rovere come essenza e la LOCTITE UK 1351 B25/UK 5452 come colla da utilizzare nelle prove sperimentali successive. La ricerca prosegue con l’ideazione di un idoneo setup di prova per poter testare il materiale scelto che consiste in due piastre metalliche di confinamento che, accorpate ai provini, evitano movimenti fuori piano e si pone l’obbiettivo di calibrare di conseguenza una legge di contatto tra le due superfici incollate. Negli elaborati precedenti, testando incollaggi di lunghezza 30, 60, 90, 120, 150 e 180 mm, è stata calibrata una legge di contatto in direzione parallela alle fibre. Tutte le suddette ricerche precedentemente svolte ed i risultati ottenuti sono stati riassunti all’interno della prima parte di questo elaborato in modo da poter fornire un quadro il più chiaro possibile sulle informazioni necessarie e gli studi effettuati in merito al consolidamento della trave lignea ammalorata. Come prosecuzione del lavoro svolto in precedenza questa tesi verterà: sulla calibrazione della sopracitata legge di contatto tra due superfici incollate lignee in direzione perpendicolare alle fibre, in quanto l’ortotropia del materiale non permette lo studio della legge di contatto in una sola direzione; sulla verifica della legge di contatto in direzione perpendicolare alle fibre tramite modelli ad elementi finiti e sull’applicazione della legge di contatto ricavata sulla testa di una trave lignea consolidata tramite legno lamellare. La calibrazione della legge di contatto in direzione perpendicolare alle fibre è stata effettuata tramite prove sperimentali con l’ausilio del setup di prova, studiato nell’elaborato dell’Ing. Pisacane, e la nuova tecnica di misura senza contatto DIC (Digital Image Correlation) la quale consente di determinare, tramite immagini digitali, lo spostamento dei provini al variare del tempo e del carico. L’incollaggio è stato testato tramite prove di trazione eseguite su due listelli in rovere di prima categoria di dimensioni 76x50x8 mm incollati tra loro. La tipologia di prova scelta ricade sul single lap shear di trazione ed il setup studiato in precedenza è stato riadattato per la nuova dimensione dei provini. Sono stati testati sei provini incollati per ogni lunghezza di adesione e sono state scelte tre lunghezze di incollaggio diverse, precisamente di 10, 20 e 30mm, per un totale di diciotto campioni analizzati. Inoltre, sono state effettuate tre prove di trazione su singoli listelli, cosi da determinare la resistenza ultima di rottura e il modulo di elasticità E. Lo studio e la rielaborazione dei risultati ottenuti hanno permesso di ottenere una legge di contatto in termini tensione-spostamento (τ-s) tra le due superfici incollate. La legge calibrata, in entrambe le direzioni, è stata validata tramite simulazione su software agli elementi finiti, Abaqus, in cui sono stati rimodellati i listelli testati in laboratorio. Dopo la legge di contatto, utilizzando come riferimento la pubblicazione “Stress Transfer and Fracture Propagation in Different Kinds of Adhesive Joints” (Zhishen Wu; Hong Yuan; and Hedong Niu) è stata calcolata la lunghezza di incollaggio minima da utilizzare. Legge di contatto e lunghezza minima di incollaggio ottenute sono poi state riutilizzate per modellare nuovamente su software le travi testate dall’Ing. Antelmi all’interno del suo elaborato. Dalla modellazione effettuata risulta che la lunghezza utilizzata, ottenuta dai calcoli precedenti, non è sufficiente. Questo porterà all’analisi dell’efficacia della tecnica tutt’oggi utilizzata per il consolidamento di una trave lignea attraverso l’utilizzo del legno lamellare, la quale prevede il consolidamento della testa della trave tramite l’incollaggio di listelli disposti verticalmente, e ne verrà proposta una alternativa, più performante: il consolidamento della testa della trave lignea tramite l’incollaggio di listelli disposti orizzontalmente. Con la finalità di confermare la miglior efficacia della tecnica proposta, le due metodologie di consolidamento verranno confrontate tramite modellazione su software Abaqus mettendo a confronto gli sforzi tangenziali presenti sulle superfici di incollaggio. Infine, la nuova tecnica verrà approfondita sulle tematiche che riguardano il numero e la lunghezza di incollaggio dei listelli da utilizzare per il consolidamento.
Indagine numerica e sperimentale su una tecnica di recupero di travi lignee
VILLA, MARTINA
2018/2019
Abstract
This paper focuses on the study of a technique for recovering damaged wooden beams with the aid of structural adhesives. The technique being studied is based on the realization of a laminated wood prosthesis that will replace the part of the degraded beam, by using only a suitable structural adhesive as connection element. The paper takes up the research previously carried out by prof. Marco Andrea Pisani and the colleagues Eng. Antelmi and Eng. Pisacane who deal with the study of wood as an essence and as a structural material, its mechanical properties and the degradation mechanisms; on the research for an effective surface treatment of the glued wood and the type of glue to be used for this purpose. These studies have led to choosing oak as the essence and LOCTITE UK 1351 B25 / UK 5452 as a glue to be used in subsequent experimental tests. The research continues with the conception of a suitable test setup to be able to test the chosen material which consists of two metal confinement plates which, combined with the specimens, avoid out of plane movements and aims to calibrate a contact law between the two bonded surfaces. In the previous papers, by testing gluing lengths of 30, 60, 90, 120, 150 and 180 mm, a local bond-slip model was calibrated in the line parallel to fiber direction. The purpose of this paper is to focus on: calibration of the local bond-slip model of contact between two wooden bonded surfaces in the line orthogonal to fiber direction, because the orthotropy of the material does not allow the study of the local bond-slip model in only one direction; on the verification of the local bond-slip model in the line orthogonal to fiber direction through finite element models and on the application of the local bond-slip model obtained on the head of a timber beam consolidated with laminated wood. The calibration of the local bond-slip model in the line orthogonal to fiber direction was carried out through experimental tests with the help of the test setup, studied in the paper of Eng. Pisacane, and the new non-contact measurement technique DIC (Digital Image Correlation) which allows to determine, through digital images, the displacement of the specimens with time and load variations. The bonding was tested by tensile tests performed on two 76x50x8 mm first class oak strips glued together. The type of test chosen falls on the single lap shear of traction and the setup previously studied has been readapted for the new dimension of the specimens. Six glued specimens were tested for each adhesion length and three different bonding lengths were chosen, namely 10, 20 and 30mm, for a total of eighteen samples analyzed. Furthermore, three tensile tests were carried out on individual strips, to determine the strength and the modulus of elasticity. The study of the results allowed to obtain a local bond-slip model in terms of interfacial shear stress-slip (τ-s) between the two bonded surfaces. The calibrated model, in both directions, has been validated by simulation on finite element software, Abaqus, in which the strips tested in the laboratory have been shaped. After the local bond-slip model, using the publication "Stress Transfer and Fracture Propagation in Different Kinds of Adhesive Joints" (Zhishen Wu; Hong Yuan; and Hedong Niu) the minimum bonding length to be used was calculated. The local bond-slip model and minimum bonding length obtained were then used to model the beams tested on the software by the Eng. Antelmi in her paper. From the modeling results that the length used, obtained from the previous calculations, was not enough. This will lead the analysis on the effectiveness of the technique used today, which involves the consolidation of the head of the wooden beam by gluing vertically the arranged strips, and an alternative, more performing will be proposed: the consolidation of the beam wooden head by gluing horizontally arranged strips. With the aim of confirming the best effectiveness of the proposed technique, the two consolidation methods will be compared by modeling on the software Abaqus, comparing the tangential stresses present on the bonding surfaces. Finally, the new technique will be examined in depth on the issues concerning the number and length of bonding of the strips to be used for consolidation.File | Dimensione | Formato | |
---|---|---|---|
Tesi Villa Martina - Indagine numerica e sperimentale su una tecnica di recupero di travi lignee.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
19.34 MB
Formato
Adobe PDF
|
19.34 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/150790