Second order analysis, rf-plasticity and 7gdl

[Helena Goni]

If I’m studying a frame (it already exists), and critical factor is 1.60 in the first bucking mode (global buckling), I need to calculate in second order with imperfections. The first question is if it’s better to perform the analysis with several load increments.

Then I can check in RF-Steel EC-3.

This is the model.



If I study the top member in RF-Steel EC3 checking stability with general method (6.3.4) only in minor z-axis (because I’ve already introduced imperfections) and lateral buckling, as it has an I section with a lower cut I-section, RFEM classifies it as class 3 (in the next image, stability is not activated jet):



So, if I want to use plasticity capacity, I have to use add-on RF-Plasticity that I have recently bought. But if I do that, I can’t use general method (6.3.4):



So, I have to use 7GDL:



But, if I’m using a second order analysis in RFEM, load combinations are selected to design in RF-Steel EC3. If I select second order also in warping analysis, is this correct 2º order + 2 order?

What I would like to do is to check with general method but using plasticity capacities, but I don’t know if that’s possible. How should we proceed if we want to use plasticity capacities?

I’m sorry for the extension of the question. I’m a bit lost.

[Oliver Metzkes]

Hello Helena,
thank you for contacting us.

Regarding your first question concerning the load increments, I would like to recommend the following Knowledge Base article on our website: https://www.dlubal.com/en/support-an...ge-base/001558

As described there, it can be helpful to use more load increments to avoid convergence problems for nonlinear calculation, but it is not necessary to use them to get plausible results.

For standard I-shaped cross-sections, the plastic section resistance can also be used within the General Method depending on the cross-sections classification. The shape that you have chosen is unfortunately only implemented as 'General' section type and we can only classify them as class 3 or 4. Its not possible to use the plastic resistance within the Equivalent Member Method or the General Method then.



One easy approximation to use plastic cross-section resistance would be to create a parametric I-shaped section by omitting the inner flange and to use it for the analysis instead of the real cross-section. No additional Add-On module would be needed for that.

With the RF-STEEL Plasticity Add-On module you can check general cross-sections by using the plastic resistance. To consider stability issues, you would need to use the RF-STEEL Warping Torsion Add-On then, as you already know. Within the warping torsion analysis it is necessary to calculate according to 2nd order theorie and to set the correct imperfection magnitude. As this is a local solution for the selected beam only, any global 2nd order effects should already been considered within the internal forces in the main model - that means also the global model needs to be calculated with global imperfections and according to 2nd order theorie.

I can therefore also recommend the following webinars (one from 2016 when the 2 add-on modulkes were introduced, but still helpful and a newer one about stability design methods in RF-STEEL EC3):

https://www.dlubal.com/en/support-an...ebinars/002068

https://www.dlubal.com/en/support-an...ebinars/002138

I hope that my answers were helpful to you.

[Helena Goni]

Thank you, Oliver.

Your answer and webinars have been very helpful. I didn’t know I could study a member of the frame in Steel EC3, but I haven’t been able to do that in my file. Has this option any limitations?

This is my frame:




And this is what happens when I try to study it in Steel EC3:

[Oliver Metzkes]

Hello Helena,
I see from your screenshot that member set 11 is defined as 'Member Group' - but only member sets with continuous members can be designed within STEEL EC3. So you need to create member sets as continuous members for the analysis, then they can be selected.

Best regards,
Oliver Metzkes

[Helena Goni]

Hi Oliver, If I do that I get this error that I don't understand:

[Oliver Metzkes]

Hi Helena,
I see that you selected the whole Frame with all 3 spans including the columns for the member set. Unfortunately this is not included in the definition of the continuous member set, as we need to have one definite x-direction and therefore the member set is not allowed to have branches (like the columns in your case). In your case you can only analyze the beams without the columns as subsystems within RF-STEEL EC3 add-on module, assuming fork supports at the beam ends.

Then the definition can be followed as explained in the webinar I already linked before:

https://www.dlubal.com/en/support-an...ebinars/002138

I hope that my answer was helpful for you.

Best regards,
Oliver Metzkes

[Helena Goni]

Ok Oliver, thanks, it's been helpul.

Helena