Friction in Seismic Nonlinear Analysis

Hi All,

Autopipe has the option "Ignore Friction E1 to E10" for nonlinear analysis, does anyone out there know of any code references which give guidance as to if friction can or should not be considered in the earthquake cases (under B31.3)?

ASME B31.3 does not say, nor the new standard B31E, these both refer to ASCE 7. But ASCE 7 13.6.8.1, states "Pressure piping systems including their supports, designed and constructed in accordance with ASME B31 shall be deemed to meet the force,
displacement, and other requirements of this section". ASCE 7 section 13.4 NONSTRUCTURAL COMPONENT
ANCHORAGE excludes friction being considered, but is this applicable to pressure piping (or does compliance with B31.3 under clause 13.6.8.1 take precedence)? 

What's everyone's practice? Is this a case of either practice being acceptable?

FR

 

Parents
  • ASME B31 codes rely on ASCE 7, UBC and IBC codes for earthquake criteria.


    UBC 1997, section 1632.1 states:
    Friction resulting from gravity loads shall not be considered to provide resistance to seismic forces.

    Same statement is repeated in ASCE 7-2005 section 15.5.2.1 (Pipe Racks).

    Similar information can be seen in ASCE-7-2005 Chapter 13, e.g. Section 13.4 (nonstructural component anchorage). and 13.4.6 (Friction clips).
    It is true that Section 13.6.8.1 does not refer to 13.4 or 15.5, but I strongly recommend that friction is ignored for static seismic cases. Usually friction (because it is nonlinear) is ignored in all dynamic cases in AutoPIPE unless modeled as an additional spring and with the stiffness based on an assumed displacement (Stiffness = Normal_force / Assumed_Displacement).
    The reason behind ignoring friction in seismic cases is that the earthquake have both horizontal and vertical motions (usually vertical is about 2/3 of horizontal). The upward vertical acceleration will reduce the normal force which friction relies on. So unless you reduce the friction factors for seismic loads based on vertical acceleration, which cannot be done easily, it is best to ignore friction since that is likely to produce conservative results.


    Reduced friction factor can be calculated from:
                 Seismic friction factor = (1-Ey) * static friction factor
    Where:

           Ey = maximum vertical seismic acceleration in g's, for Ey > 1g use Ey=1g

    Regards,
    Karim Rinawi

     

     

     

     

Reply
  • ASME B31 codes rely on ASCE 7, UBC and IBC codes for earthquake criteria.


    UBC 1997, section 1632.1 states:
    Friction resulting from gravity loads shall not be considered to provide resistance to seismic forces.

    Same statement is repeated in ASCE 7-2005 section 15.5.2.1 (Pipe Racks).

    Similar information can be seen in ASCE-7-2005 Chapter 13, e.g. Section 13.4 (nonstructural component anchorage). and 13.4.6 (Friction clips).
    It is true that Section 13.6.8.1 does not refer to 13.4 or 15.5, but I strongly recommend that friction is ignored for static seismic cases. Usually friction (because it is nonlinear) is ignored in all dynamic cases in AutoPIPE unless modeled as an additional spring and with the stiffness based on an assumed displacement (Stiffness = Normal_force / Assumed_Displacement).
    The reason behind ignoring friction in seismic cases is that the earthquake have both horizontal and vertical motions (usually vertical is about 2/3 of horizontal). The upward vertical acceleration will reduce the normal force which friction relies on. So unless you reduce the friction factors for seismic loads based on vertical acceleration, which cannot be done easily, it is best to ignore friction since that is likely to produce conservative results.


    Reduced friction factor can be calculated from:
                 Seismic friction factor = (1-Ey) * static friction factor
    Where:

           Ey = maximum vertical seismic acceleration in g's, for Ey > 1g use Ey=1g

    Regards,
    Karim Rinawi

     

     

     

     

Children
No Data