### Abstract

The nonlinear seismic responses of steel frames with fully restrained, partially restrained and composite connections are evaluated and compared in terms of the maximum interstory and maximum top lateral displacements. Steel frames are usually analyzed assuming all the connections are fully restrained. However, considering the practical design aspects of connections, this is rarely true. This practice introduces unintended flexibility in the frame. Using a nonlinear time domain seismic analysis algorithm developed by the authors, three steel frames are excited by 13 earthquake time histories. Twelve of them were recorded during the Northridge earthquake of 1994. Any one of these 12 earthquake time histories can be used to represent the Northridge earthquake in future designs. To define the rigidity of a connection, a parameter called the T ratio is introduced. It is the ratio of the moment the connection would have to carry according to the beam line theory and the fixed end moment of the girder. Initially, the T ratio of all the connections is assumed to be 0.9, making them fully restrained. The results indicate that this assumption is inappropriate and gives unconservative responses depending upon which time history of the same earthquake is being used. Several frames with a T ratio of 0.95 developed very large lateral displacements causing instability, although they behaved properly when the ratio was assumed to be 1.0. For composite connections, slab steel has a significant beneficial effect on the overall structural response. It increases the T ratio, making it closer to the FR connection. Even for composite connections with a T ratio of 0.95, the frames developed large lateral displacement. Further parametric study indicates that, at least for seismic analysis, PR or composite connections should be designed for a T ratio as close to 1 as possible to represent an FR connection. Otherwise, the lateral displacement failure criterion should also be checked for less than ideal FR connection conditions. Improvements in both the analysis and design of steel frames are necessary to make them more seismic load tolerant.

Original language | English (US) |
---|---|

Pages (from-to) | 37-59 |

Number of pages | 23 |

Journal | Journal of Constructional Steel Research |

Volume | 51 |

Issue number | 1 |

State | Published - Jul 1999 |

### Fingerprint

### Keywords

- Keyword
- Keyword
- Keyword
- Keyword
- Keyword
- Keyword

### ASJC Scopus subject areas

- Civil and Structural Engineering
- Materials Science(all)

### Cite this

*Journal of Constructional Steel Research*,

*51*(1), 37-59.

**Nonlinear seismic response of steel structures with semi-rigid and composite connections.** / Reyes-Salazar, Alfredo; Haldar, Achintya.

Research output: Contribution to journal › Article

*Journal of Constructional Steel Research*, vol. 51, no. 1, pp. 37-59.

}

TY - JOUR

T1 - Nonlinear seismic response of steel structures with semi-rigid and composite connections

AU - Reyes-Salazar, Alfredo

AU - Haldar, Achintya

PY - 1999/7

Y1 - 1999/7

N2 - The nonlinear seismic responses of steel frames with fully restrained, partially restrained and composite connections are evaluated and compared in terms of the maximum interstory and maximum top lateral displacements. Steel frames are usually analyzed assuming all the connections are fully restrained. However, considering the practical design aspects of connections, this is rarely true. This practice introduces unintended flexibility in the frame. Using a nonlinear time domain seismic analysis algorithm developed by the authors, three steel frames are excited by 13 earthquake time histories. Twelve of them were recorded during the Northridge earthquake of 1994. Any one of these 12 earthquake time histories can be used to represent the Northridge earthquake in future designs. To define the rigidity of a connection, a parameter called the T ratio is introduced. It is the ratio of the moment the connection would have to carry according to the beam line theory and the fixed end moment of the girder. Initially, the T ratio of all the connections is assumed to be 0.9, making them fully restrained. The results indicate that this assumption is inappropriate and gives unconservative responses depending upon which time history of the same earthquake is being used. Several frames with a T ratio of 0.95 developed very large lateral displacements causing instability, although they behaved properly when the ratio was assumed to be 1.0. For composite connections, slab steel has a significant beneficial effect on the overall structural response. It increases the T ratio, making it closer to the FR connection. Even for composite connections with a T ratio of 0.95, the frames developed large lateral displacement. Further parametric study indicates that, at least for seismic analysis, PR or composite connections should be designed for a T ratio as close to 1 as possible to represent an FR connection. Otherwise, the lateral displacement failure criterion should also be checked for less than ideal FR connection conditions. Improvements in both the analysis and design of steel frames are necessary to make them more seismic load tolerant.

AB - The nonlinear seismic responses of steel frames with fully restrained, partially restrained and composite connections are evaluated and compared in terms of the maximum interstory and maximum top lateral displacements. Steel frames are usually analyzed assuming all the connections are fully restrained. However, considering the practical design aspects of connections, this is rarely true. This practice introduces unintended flexibility in the frame. Using a nonlinear time domain seismic analysis algorithm developed by the authors, three steel frames are excited by 13 earthquake time histories. Twelve of them were recorded during the Northridge earthquake of 1994. Any one of these 12 earthquake time histories can be used to represent the Northridge earthquake in future designs. To define the rigidity of a connection, a parameter called the T ratio is introduced. It is the ratio of the moment the connection would have to carry according to the beam line theory and the fixed end moment of the girder. Initially, the T ratio of all the connections is assumed to be 0.9, making them fully restrained. The results indicate that this assumption is inappropriate and gives unconservative responses depending upon which time history of the same earthquake is being used. Several frames with a T ratio of 0.95 developed very large lateral displacements causing instability, although they behaved properly when the ratio was assumed to be 1.0. For composite connections, slab steel has a significant beneficial effect on the overall structural response. It increases the T ratio, making it closer to the FR connection. Even for composite connections with a T ratio of 0.95, the frames developed large lateral displacement. Further parametric study indicates that, at least for seismic analysis, PR or composite connections should be designed for a T ratio as close to 1 as possible to represent an FR connection. Otherwise, the lateral displacement failure criterion should also be checked for less than ideal FR connection conditions. Improvements in both the analysis and design of steel frames are necessary to make them more seismic load tolerant.

KW - Keyword

KW - Keyword

KW - Keyword

KW - Keyword

KW - Keyword

KW - Keyword

UR - http://www.scopus.com/inward/record.url?scp=0032623389&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032623389&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0032623389

VL - 51

SP - 37

EP - 59

JO - Journal of Constructional Steel Research

JF - Journal of Constructional Steel Research

SN - 0143-974X

IS - 1

ER -