Increasing Ductility Demand

[mindlessflip]

Hi dear all,
I just wanted to ask if anyone could give me some advice in order to increase the ductility demand in my RC-structure using the SeismoStruct software:

It´s a 3-storey frame with 3-bays in the x-direction.
Column sections are modelled 30x30cm with C30/37 (con_ma) and reinf. is modelled with S500 (stl_mp):
- longitudinal reinforcement is 8x16mm,
- transverse reinforcement is 2 and 2 with diamond stirrups
- hoops 12mm-0,1m.
The beam sections have dimension of 30x40cm with C25/30 and are also reinforced with S500.
- longitudinal reinf. is 4x12mm,
- transverse reinf. is 2 and 2
- hoops 10mm-0,1m.

Masses are modelled with lumped masses (8t/edge). Rigid diapgrams are used in every floor. Applied loads for the pushover analysis are 5kN for the first, 10kN for the second and 15kN for the third storey. Maximum displacement =0,5m in order to reach umax, maybe to much? Target displacement is calculated based on EC8.

When I run now the static pushover analysis the structure reaches a umax of 0,3350m (3,7% of total height). How do I define the uy (yield displacement). Should I use the uy of the idealized curve or from the real one? When I do this I have 0,3350/0,0826 = 4,06. When I use the real curve I have 0,3350/0,1350= 2,48.

PS. When I have a look on the deformed shape, it looks like there is only hinges on the column ends of the first floor. There are no plastic hinges on the beams. Thank you very much, I would very appreciate some advice!

[huffte]

If the first floor columns hinge prior to any beam hinges, it would seem that the first floor columns are under-designed. You can examine hysteresis curves for beams in the post-processor to see if, indeed, they remain elastic for the entire pushover analysis.

Regarding whether to use first yield or the idealized yield, it depends on what you wish to use the value for. Are you wanting to know the instance of first yield? Then use the real value. Or are you wanting to develop a bi-linear force-displacement curve which would presumably give similar results to the detailed analysis you have performed? Then use the idealized value. Or is it something entirely different from either of these? This is perhaps a matter best discussed with your supervisor to determine what you wish to accomplish with your analysis.

Best of luck with your work.

[mindlessflip]

Thank you very much for your quick response @huffte!

The aim of my study is to check the ductility demand of the RC structure for a few time histories. Maybe then I should use the yield value of the real curve?
Furthermore I wanted to ask you if there is any possibility to obtain the participation factor for example in the form 1,24 for the first mode?

Greetings

[huffte]

I think the most important concept will be to be explicit in reporting your results regarding how you established the yield displacement. I would say that a large number of studies have historically relied upon the idealized value. But, as I say, just be precise and specific about the results you report. You and your supervisor must get together and make that decision yourselves, I believe.

As for participation factors, you could certainly perform an eigenvalue analysis on the model to get modal participation data.