Questions about Ch4-steel-01

[zhwang]

Dear Seismosoft support,

I am trying to do some more analysis on Ch4-steel-01. However, from the references of Ch4-steel-01 in verification report (Page 93), I notice that the yield strength of beams and columns in experimental tests (388MPa, 393MPa, 398MPa, 286MPa and 347MPa) are lower than that you set in Ch4-steel-01. Besides, there is RC slab on the beams in the experiment references, but Ch4-steel-01 does not include the slab.

Although the above difference between the experiment and the modelling, the comparison of hysteretic curves on Page 91 of the report looks good. Could you please suggest me why you model this with different yield strength and not include the slab? But still get good simulation results for hysteretic curves? Do you think Ch4-steel-01 can well represent the real experimental frame?

Ch4-steel-01 is attached: https://www.dropbox.com/s/h866pk5t3dos2 ... 1.spf?dl=0

The references are attached: https://www.dropbox.com/s/k2f002elix07j ... 4.pdf?dl=0
https://www.dropbox.com/s/7p6eviyxb2um2 ... 6.pdf?dl=0
https://www.dropbox.com/s/f5b6bu7nu3vut ... g.pdf?dl=0


Thank you very much.

[z.gronti]

Dear Zhwang,

If you have read the attached reference of Nakashima M., Matsumiya T, Suita K, Liu D. [2006] you would have seen in page 14 "Numerical Analysis after test" that the measured material properties of Table 1, used in the experiment, are multiplied by some incremental factors. If you multiply the values of Table 1 with these factors you will get the yield strength assigned in the SeismoStruct model.

As far as the slab modelling is concerned you may see from the above reference, that in the SeismoStruct model was used the same modelling with the paper of Nakashima M., Matsumiya T, Suita K, Liu D. [2006].

[zhwang]

Dear Zoi,

Thank you very much for your kind reply.

In the reference of Nakashima M., Matsumiya T, Suita K, Liu D. [2006], I have seen in page 10 “During the first cycle in the positive loading of the 1/20 amplitude, the ‘North’ plane’s second floor beam was fractured from the beam bottom flange” and “loading to the maximum overall drift angle of 1/15. The first storey shear force decreased significantly because of a combination of strength deterioration of column bases and progress of column local buckling.” 1/20 multiplied by 8500mm is 425mm, and 1/15 multiplied by 8500mm is about 567mm. It means the fracture of beam happened at the displacement of 425mm, and the buckling of column happened at 567mm.

However, when I run the pushover analysis on Ch4-steel-01, I found that the fracture of the second floor beam (beam122 section a) is about 1298mm. Before 1298mm, only yielding happened but no damage.

Pushover for Ch4-steel-01 is attached: https://www.dropbox.com/s/m5vfsby39e59c ... 1.spf?dl=0

Could you please suggest why the model Ch4-steel-01 fail much later than experiment and how to modify the Ch4-steel-01 to make it have an accurate displacement of damage compared with the experiment test?

Thank you very much.

[z.gronti]

Dear zhwang,

Pushover analysis is a simplified method of analysis. I think that you should not compare the results of the pushover analysis with the more complex one of cyclic loading. Please, refer to adequate scientific sources about the application of the pushover analysis.