## response spectrum factor

### response spectrum factor

what is the basis of response spectrum factor?why do it important?

### Re: response spectrum factor

Hi sghosh. The response spectrum factor allows the user to adjust accelerograms to multiples or fractions of the same response spectrum without having to create an entirely new response spectrum.

For example, in some cases it may be desirable to adjust one set of accelerograms to 0.8 times a particular design spectrum, and adjust another set to 1.25 times the same design spectrum, just to give a numerical example. These could then be used as two perpendicular components in a multi-directional nonlinear analysis of a structure. The key point being that only a single response spectrum had to be read in, whereas had the factor ability not been incorporated the user would have to create three spectra really - the original design spectrum, a spectrum equal to 1.25 times the design spectrum, and a third spectrum equal to 0.8 times the design spectrum. Not that most codes produce a design spectrum which is geometric mean of two components in nature. So by adjusting to different multiples of the design spectrum, some measure of relative amplitude between components may be maintained while still producing a ground motion record pair which has the desired geometric mean spectrum.

I hope this makes some bit of sense sghosh.

Of course, if you simply want uni-directional records adjusted to a target spectrum, the the factor will be 1.0.

Best of luck sghosh.

For example, in some cases it may be desirable to adjust one set of accelerograms to 0.8 times a particular design spectrum, and adjust another set to 1.25 times the same design spectrum, just to give a numerical example. These could then be used as two perpendicular components in a multi-directional nonlinear analysis of a structure. The key point being that only a single response spectrum had to be read in, whereas had the factor ability not been incorporated the user would have to create three spectra really - the original design spectrum, a spectrum equal to 1.25 times the design spectrum, and a third spectrum equal to 0.8 times the design spectrum. Not that most codes produce a design spectrum which is geometric mean of two components in nature. So by adjusting to different multiples of the design spectrum, some measure of relative amplitude between components may be maintained while still producing a ground motion record pair which has the desired geometric mean spectrum.

I hope this makes some bit of sense sghosh.

Of course, if you simply want uni-directional records adjusted to a target spectrum, the the factor will be 1.0.

Best of luck sghosh.

Tim Huff

### Re: response spectrum factor

Thank you,huffte.

How can I specify PGA for a particular spectrum factor to achieve convergence?

How can I specify PGA for a particular spectrum factor to achieve convergence?

### Re: response spectrum factor

Ah, I think that there is no easy answer to your question sghosh. As noted in the Help System, if you check the "Specify PGA" box, then the "software automatically scales the generated PGA to match the target PGA."

So you might reason that if a scale factor of 1.5, just to use an example, is used on the spectrum factor, then one should multiple the PGA noted in the box by 1.5 as well. And if PGA were the sole factor in determining the compatibility of the response spectrum of the accelerogram to the target response spectrum, this would be difficult to argue against. But PGA is just a small piece of the puzzle when trying to adjust an accelerogram over a wide range of periods.

I would say start out without checking the Specify PGA box. Try to obtain convergence that way. If you are, in fact, using a spectrum factor other than 1.0 and you cannot get convergence, then you could try checking the Specify PGA box and changing the default value in the box to 1.5 times that initially shown. Try for convergence this way.

I will say that I have had excellent results in obtaining convergence using SeismoArtif.

Best of luck sghosh.

So you might reason that if a scale factor of 1.5, just to use an example, is used on the spectrum factor, then one should multiple the PGA noted in the box by 1.5 as well. And if PGA were the sole factor in determining the compatibility of the response spectrum of the accelerogram to the target response spectrum, this would be difficult to argue against. But PGA is just a small piece of the puzzle when trying to adjust an accelerogram over a wide range of periods.

I would say start out without checking the Specify PGA box. Try to obtain convergence that way. If you are, in fact, using a spectrum factor other than 1.0 and you cannot get convergence, then you could try checking the Specify PGA box and changing the default value in the box to 1.5 times that initially shown. Try for convergence this way.

I will say that I have had excellent results in obtaining convergence using SeismoArtif.

Best of luck sghosh.

Tim Huff