Scaling

02-Analytical capabilities
Post Reply
aposhnejad
Posts: 1
Joined: 28 Apr 2009, 16:38

Scaling

Post by aposhnejad »

North American building codes require to match the ground motion to the target spectra in period range 0.2T to 1.5T in a way that the ordinate of matched spectra never be less than 1.4 times the target spectra.

Now my question is whether Seismomatch consider this "1.4" factor or I have to scale the target spectra?
User avatar
seismosoft
Posts: 1184
Joined: 06 Jul 2007, 04:55

Re: Scaling

Post by seismosoft »

Dear A.H. Poshnejad,

We are not sure we understand your query, since the code requirement you quote is not strictly related to the scaling/matching of a time-history (which is what SeismoMatch does), but rather to the combination of the two horizontals components of a given record (an issue that SeismoMatch does not address).

Indeed, such code provisions require that seven accelerograms pairs should be scaled such that the average value of their SRSS spectra is not less than 1.4 times the standard design-spectra for periods of 0.2T to 1.5T.

In SeismoMatch, issues of horizontal components combination are not considered, hence you will need to carry out such operations/checks outside of the program.

Seismosoft Support
Navid
Posts: 17
Joined: 18 Jun 2017, 02:09

Re: Scaling

Post by Navid »

Hi all. I have got two earthquake acceleration records from two assorted excitations, which are supposed to be scaled as per the Australian response spectrum for bedrock as a seed one, with the intention of harnessing them as the applied excitations at the base of a 3D soil-structure interaction (SSI) model. Could you please shed light on the appropriate values associated with the min-max period range inasmuch as the range, i.e. 0.2T to 1.5T, was posited by ASCE for scaling each pair of the horizontal ground motion components, culled from an individual record event. My gut feeling is that the 0.05-4-second period range would be acceptable in my project, isn't that bang on?


Thanks heaps!
Navid
huffte
Posts: 978
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Scaling

Post by huffte »

You lower bound of 0.05 is likely fine. The upper bound really depends on the first mode period and the level of in elasticity experienced by the structure. As an example, suppose a structure experienced a displacement ductility of 6. Further suppose that the structure were essentially elastic - perfectly-plastic. Then the effective period would be T*sqrt (6) = 2.45T. So you would likely want to extend the range to 2.45T, rather than 1.5T. So, if your fundamental period is less than or equal to 4/2.45=1.63 seconds, then your range should suffice for the example I pose above. Best of luck.
Tim Huff
Navid
Posts: 17
Joined: 18 Jun 2017, 02:09

Re: Scaling

Post by Navid »

Dear huffte,

You've got my big thank-you in light of such a peachy answer. Likewise, is there any reference positing a rough estimation of the inelastic deformation ratio, aka "ductility demand", which is usually equal to 2, 4, 6, or 8 in a pool of papers?

Regards,
Navid
huffte
Posts: 978
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Scaling

Post by huffte »

For a structure designed for collapse prevention through plastic hinging, a value of 6 might be a reasonable approximation. Of course, you can iterate and adjust if you first run the analysis and a higher value shows up in the results.

For a structure designed with isolation as the seismic strategy, much higher values can be realistic if the non-isolated period is very low (a very stiff non-isolated condition). Many isolation devices can operate at ductility levels of 10 or more.
Tim Huff
Navid
Posts: 17
Joined: 18 Jun 2017, 02:09

Re: Scaling

Post by Navid »

Hi huffte. Thanks a bunch. First and foremost, is it possible for me to have a reference for a rough estimation of the ductility demand ratio which is 6? Additionally, my understanding is that T', i.e. 1.5T1, could be named as the minimum possible value for Max Period, assuming the elastic behavior for a structure, whilst T'', equal to sqrt(nu).T1, is the minimum possible value for Max Period, supposing the elastic-perfectlyplastic behavior for a structure. Now, here is the example; the fundamental period of a superstructure is approximately 2 sec. Thus, T'' equals 4.9 sec; nonetheless, the maximum period for the seed response spectrum is in every turn fixed to 4 sec either making use of EU8 response spectrum or defining a response spectrum from other cods and standards. The thing which I have been striving to ask is how am I able to alter the said Max Period for the seed response spectrum such that sqrt(nu).T1 could be employed? Otherwise, T'' is equaling to 5, while the last value on the horizontal axis of the target response spectrum, considered in SeisoMatch would be always 4.


Kind Regards,
Navid
huffte
Posts: 978
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Scaling

Post by huffte »

You can define target spectra beyond 4 seconds. You may have to manually (spreadsheet) calculate the spectrum, save as a text file, and import from the file into SeismoMatch. I have defined target spectra up to 15 seconds.
Tim Huff
Post Reply

Return to “02-Analytical capabilities”