Computational assessment of seismic resistance of RC framed buildings with masonry infill

Loots, Jurie (2005-12)

Thesis (MScIng)--Stellenbosch University, 2005.

Some digitised pages may appear illegible due to the condition of the original hard copy.

Thesis

ENGLISH ABSTRACT: Reinforced concrete (RC) frames with unreinforced masonry infill form the structural system of many buildings and this is also true for South Africa. It is common practice to consider the masonry infill as a non-structural component and therefore it does not contribute to the performance of the Re frame buildings under lateral loading such as earthquake loading. This is done by leaving a sufficient gap between the Re frame and the infill. This ensures that there is no contact between the frame and the infill during an earthquake event. However, it has been suggested that masonry infill can play a significant role in the performance of a Re frame building under lateral loading. The first part of the study focuses on the South African situation. The relevance of shear walls in these Re frame buildings as well as the size of the gap (between frame and infill) left in practice, are investigated. This is done by finite element analysis. The second part of the study focuses on the effects that the infill can have on the global performance of the structure when there is full contact between the Re frames and infill. The effect of openings in the infill to the response of the frame is also investigated. Finite element models of single span Re frames with infill is built and analyzed in order to investigate possible damage to the infill, frame infill interaction and to obtain the non linear stiffness of the frame with infill as a whole. This obtained non linear stiffness can be modelled in Diana as a non linear spring that will be used in the development of a simplified analysis method. The simplified method developed consists of a frame and two such non linear springs, placed diagonally, and which have the same force versus displacement behaviour as the original frame with infill. These single span frames can be added together to model a whole frame. In a first step to generalise the simplified method, various geometries of infills are considered, varying span and height, as well as opening percentage, representing windows and doors of varying total area and positioning. However, in this study a single masonry type, namely solid baked clay bricks set in a general mortar, is considered. To generalise the approach further, other masonry types can be considered in the same way. The use of these springs in a simplified model saves computational time and this means that larger structures can be modelled in Diana to investigate response of'Rf' frame buildings with infill. The work reported in this thesis considers only in-plane action. Out-of-plane-action of the masonry infill has been reported in the literature to be considerable, under the condition that it is sufficiently tied to the frame to prevent mere toppling over, causing life risking hazards in earthquake events. This matter should be studied in continuation of the current research to generalise the simple approach to three dimensions.

AFRIKAANSE OPSOMMING: Gewapende betonrame (GBR-e) met ongewapende messelwerk invulpanele (invul) vorm die strukturele ruggraat van vele geboue en dit geld ook vir geboue in Suid-Afrika. Dit is algemene praktyk om die invulpaneel in sulke geboue as 'n nie-strukturele komponent te beskou. Daarvolgens dra dit nie by tot die gedrag van 'n GBR gebou onderhewig aan 'n aarbewing nie. Dit word bereik deur 'n groot genoeg gaping tussen die betonraam en die invul te los. Die gevolg is dat daar geen kontak tussen die betonraam en die invul plaasvind indien daar 'n aardbewing sou voorkom nie. Dit is egter voorgestel dat invul 'n noemenswaardige rol kan speel in die gedrag van 'n GBR gebou onderwerp aan 'n horisontale las. Die eerste deel van die studie fokus op die Suid-Afrikaanse situasie. Die relavansie van skuifmure in GBR geboue asook die grootte van die gaping (tussen die raam en invul) wat in die praktyk gebruik word, word ondersoek. Dit word gedoen met behulp van eindige element analises. Die tweede deel van die studie fokus op die effek wat invul kan hê op die globale gedrag van 'n struktuur wanneer daar volle kontak tussen die GBR en die invul is. Die effek wat die teenwoordigheid van openinge in die invul kan hê op die gedrag van 'n GBR is ook ondersoek. Eindige element modelle van enkelspan GBR met invul is gemodelleer en geanaliseer om die moontlike skade aan die invul, die interaksie tussen die GBR en die invul asook die nie-lineêre styfheid van die raam en invul as 'n geheel, te ondersoek. Hierdie nielineêre styfheid kan in Diana as 'n nie-lineêre veer gemodelleer word en word gebruik in die ontwikkeling van 'n vereenvoudigde metode. Hierdie vereenvoudigde metode wat ontwikkel is, bestaan uit 'n raam en twee sulke nielineêre vere (diagonaal geplaas). Die raam met vere het dieselfde krag teenoor verplasingsgedrag as die van die oorspronklike raam met invul wat dit voorstel. Hierdie rame kan saamgevoeg word om 'n raam uit 'n gebou as 'n geheel te modelleer. Verskeie invul geometrieë word gebruik in die analises in 'n eerste stap om die vereenvoudigde metode te veralgemeen. Die span en hoogte asook opening persentasie van die invul word gevariëer om vensters en deure van veskeie grootte en posisie voor te stel. In die studie, 'n enkel messelwerk tipe, naamlik solied klei bakstene geset in algemene mortar, word gebruik. Ander messelwerk tipes kan gebruik word om die metode verder te veralgemeen. Die gebruik van die vere in die vereenvoudigde metode spaar berekenings tyd en dit beteken dat groter strukture in Diana gemodelleer kan word om die gedrag van GBR geboue met invul te ondersoek. Die werk gedoen in die tesis neem slegs in-vlak aksie in ag. Literatuurstudie dui daarop dat goeie uit-vlak-aksie van messelwerk invul bestaan, mits dit goed geanker is aan die raam om te verseker dat dit nie kan omval en 'n gevaar vir lewens in 'n aardbewing inhou nie. Dit behoort verder bestudeer te vord in die vervolging van die huidige ondersoek om die vereenvoudige metode na drie dimensies te veralgemeen.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/50299
This item appears in the following collections: