performance criteria

[iman2008]

hello
I want you send me literature or paper for guidance on curvature/rotation values of steel frames to be employed in performance criteria option of the programs,exept Priestley,because i couldn't find it on the net and it is not available in our country .
thank you very much.

[Danae]

Excuse-me, the chord rotations mentioned as performance criteria, refer only to the plastic part of the rotation, or the rotation as a whole, with the yielding part? I'm asking that because FEMA-356 and ATC-40 give acceptance criteria as plastic rotations.

[seismosoft]

Danae: The chord rotations provided by SeismoStruct refer to the frame element chord rotations (see Help System), not necessarily to the structural member chord rotations requested in seismic design codes or guidelines. The latter can be obtained from the nodal displacements/rotations (see e.g. paper by Mpampatsikos etal. in the Journal of Earthquake Engineering, Vol. 12, Issue SP1, pp. 52-58, 2008).

Iman: If you carry out a thorough literature search, you will find plenty of papers, reports, books and design codes/guidelines that provide the information you are looking for.

Seismosoft Support

[Danae]

Thank you for the prompt response and for the article's name that you sent me. I tried to read it,especially the part of ductile mechanisms. I have three very important questions upon how Sesimostructs works, according to all the above :

1)Seismostruct automatically computes for chord rotations the sum of the drift of the shear span è1 and the nodal rotation è2 or it is only one of the two values?

2)What is checked and compared with the performance criteria value as chord rotation is the chord rotation of every element of which consist the members and this value is not the same as the chord rotation of the member as a whole?In that case which value would you propose as the chord rotation capacity is not a constant value which could be introduced?

3)So as to compare with the acceptance criteria of plastic hinges rotation that is requested in ATC-40 (èpl=0.010), do you mean that I have to check the rotation values of the nodes of each element + the drift of the calculated shear span and subtract the elastic part?

I'd like to thank you very much for all your help!

Please, I consider this post as urgent because I may need to change some things and I don't have much time!!

[seismosoft]

Dear Danae,

We will be repeating ourselves in answering your queries, but ok:

1) the element chord-rotation values output by SeismoStruct correspond to the rotations theta2 and theta3 indicated in the 'Geometric Nonlinearity' section of the Help System.

2) yes, as stated above already, and as described in the Help System, the chord rotation output by SeismoStruct is that of the frame element, not of the structural member (in case the latter is subdivided in more than one element).
The chord rotations of structural members can and should be obtained through the post-processing of nodal displacements and rotations, as mentioned already in our previous response, and as you yourself state in point (3).

3) yes.

Seismosoft Support

[Danae]

Thank you very much and excuse-me again for being tiring, but I wanted to be absoluteley sure that I have correctly understood. Thank you again for all your help!

[iman2008]

Hi
according to the last question,i could find Priestley M.J.N. [2003] Myths and Fallacies in Earthquake Engineering,but i couldn't find anything about curvature/rotation values of steel frames to be employed in performance criteria option of the program,can you tell me which chapter please?
Q:In IDA analysis , when i select a chord rotation performance criteria for some elements and choose stop option,why does the analysis stop fully when it reaches limit state,and don't continue analysis with the next scale factor?in fact, is it necessary to define performance criteria in IDA analysis or it's not? in the other word, in IDA analysis, isn't any problem if we let all of the Dynamic time history analysis run till the end of the record.
thank you very much.

[seismosoft]

Dear Iman,

1) your repeated question on steel performance criteria does not relate to SeismoStruct but rather to conceptual earthquake engineering issues. As stated before, we strongly advise you to carry out a proper literature review on this topic (visit your library, search the internet, read design codes or guidelines, etc).

2) as described in the Help System, the Performance Criteria function is simply informative, it has no influence in the analysis results. Hence, you may certainly run any type of analysis without having it activated.

Seismosoft Support

[amy]

Hello everyone,
I was reading the forum to understand how to calculate the chord rotation of beam/column when the member is divided into more than one elements. I also read the reference paper (V. MPAMPATSIKOS et al.) But I am still confused about the procedure.

I am analysing a steel frame using infrmDB elements. I divided each beam/column into three elements. So I know that I have to calculate the chord rotation as Seismostruct is giving me chord rotation for each element.

Is there anyone who can explain me the procedure to calculate the chord rotation or suggest me any reference with example?

Thanks
Amy

[huffte]

Hi amy.

The chord rotation may be calculated as:

(displacement at top of column - displacement at bottom of column)
divided by
(column height)

If your columns are divided into three equal elements, then it turns out that the overall chord rotation is the average of the chord rotations for the 3 segments. ( I recommend that you work the geometry our to convince yourself.) For example, suppose:

delta1 = 1
delta2 = 4
delta3 = 9
delta4 = 11
L=120

Then:

theta1 = (4-1)/(120/3) = 3/40
theta2 = (9-4)/(120/3) = 5/40
theta2 = (11-9)/(120/3) = 2/40

These would be the three values reported from the software. But the value you want is the overall chord rotation from top to bottom of column.

delta=(11-1)/120 = 10/120 = 1/12

delta=(1/3)(3/40 + 5/40 + 2/40) = 10/120 = 1/12

So, you could either calculate the chord rotation by subtracting displacements at the top and bottom of the column at each instant of time and dividing the maximum difference by the column height. Or, you could average the three reported values reported for the three subdivisions of the column. Note that:

1. The averaging procedure assumes that the maximum chord rotation occurs simultaneously with each subdivision chord rotation.
2. The calculation assumes displacements small enough so that tan(theta) and theta are essentially equal, a typically realistic assumption.
3. The calculations assume fixed base columns. You may need to make adjustment for non-zero slopes at the member ends (see FEMA 356, page 5-13).

Best of luck, amy.