Documentation on HHT integrator (parameter Gamma for ex) ?

For posts concerning the documentation, errors, ommissions, general comments, etc.

Moderators: silvia, selimgunay, Moderators

Post Reply
Arnaud
Posts: 4
Joined: Fri Sep 29, 2006 1:31 pm
Location: University of British Columbia, Vancouver Canada

Documentation on HHT integrator (parameter Gamma for ex) ?

Post by Arnaud » Thu Apr 19, 2007 10:19 am

Hi,

I would like to use the HHT integrator for dynamic analysis.
However, no decent information is available in the user manual : in particular, for the main parameter Gamma, the only information is "Newmark parameter ?" (with the question mark...) (and normally the HHT method is also called Alpha-method)

I need to know what this Gamma parameter stands for and how it is used in this equations formulation, as well as what is the HHT formulation used in the code, and how to use this Gamma to specify the numerical damping associated with the method.

Thanks a lot for your help,
Arnaud Charlet
University of British Columbia, Vancouver Canada

silvia
Posts: 3909
Joined: Tue Jan 11, 2005 7:44 am
Location: Degenkolb Engineers
Contact:

Post by silvia » Thu Apr 19, 2007 11:10 am

i'll have fmk respond. in the meantime, though, you might want to look at the code directly.
Silvia Mazzoni, PhD
Structural Consultant
Degenkolb Engineers
235 Montgomery Street, Suite 500
San Francisco, CA. 94104

bkosbab
Posts: 30
Joined: Tue Nov 14, 2006 12:15 pm
Location: Georgia Institute of Technology

Post by bkosbab » Thu May 15, 2008 5:50 am

I know this is an old post, but if there has been a definitive answer, I would love to hear. Thanks

silvia
Posts: 3909
Joined: Tue Jan 11, 2005 7:44 am
Location: Degenkolb Engineers
Contact:

Post by silvia » Thu May 15, 2008 5:52 am

frank's away until may 25th.
you might want to research the method, and the best way to get the details is to look at the code.
Silvia Mazzoni, PhD
Structural Consultant
Degenkolb Engineers
235 Montgomery Street, Suite 500
San Francisco, CA. 94104

bkosbab
Posts: 30
Joined: Tue Nov 14, 2006 12:15 pm
Location: Georgia Institute of Technology

Post by bkosbab » Thu May 15, 2008 6:05 am

Thanks for the quick reply; I will continue to dig. However, it would seem that the gamma factor here is roughly equivalent to a Newmark gamma, and can be calculated from a desired alpha value. Hopefully this can be confirmed with further digging.

silvia
Posts: 3909
Joined: Tue Jan 11, 2005 7:44 am
Location: Degenkolb Engineers
Contact:

Post by silvia » Thu May 15, 2008 1:09 pm

i hate to cheeze out and say look at the source code, but you may get more out of it than if i do.
Silvia Mazzoni, PhD
Structural Consultant
Degenkolb Engineers
235 Montgomery Street, Suite 500
San Francisco, CA. 94104

sokito
Posts: 12
Joined: Sun Dec 09, 2007 10:29 am
Location: Atlanta

Post by sokito » Sun Nov 02, 2008 1:47 pm

For HHT you have to put a alpha value, not a gamma.
Originally alpha method is unconditionally accuracy and of second order accuracy with alpha (-0.3~0), but in OpenSees we put alpha (0.7~1).
So in opensees alpha=1 is equivalent to the newmark average acceleration method, and alpha=0.7 is maximum numerical dissipation without loosing the second order accuracy.
This is what I heard from Andreas Schellenberg.
Seokho Jeong
Geosystems
Georgia Institute of Technology

Post Reply