Solved: Re: Rigid frame joint

Anonymous · ‎01-10-2016

Hi all,

I'm new to Robot and trying to model a concrete frame structure with some rigidity at the beam-column joints. By default It seems there is no rigidity assigned in the joint.

I can't find this feature in Robot, I'm afraid if I leave this, then I will underestimate the stiffness of the frame and since I'm doing earthquake design, the inertial forces will in turn be underestimated too. Even for gravity case, I think if the frame section dimension (joint dimension) is relatively large, then joint rigidity is essential to get the accurate hogging moment at the beam ends (at columns or walls face) and sagging moment at the beam mid point. Failing to model the joint rigidity may lead to underestimating hogging moment at beam ends and and overestimating sagging moment.

Anyone can help?

Thanks!

Artur.Kosakowski · ‎01-10-2016

HI yakso.sulistio,

I'm new to Robot and trying to model a concrete frame structure with some rigidity at the beam-column joints. By default It seems there is no rigidity assigned in the joint.

On the contrary (unless you defined a release or your structure is of a truss type) by default the connection is fully rigid.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

Artur Kosakowski

Anonymous · ‎01-11-2016

Hi Arthur, thanks for your reply.

Apologize if I didn't make myself really clear, what I meant was the beam-column joint panel rigidity (refer attached sketch).

As the joint rotates, if we have a FULLY RIGID joint panel, then everything within the panel will move or in this case rotate together, maintaining the original 90° angle between beam and column within the panel only (thus a0 = a1 = 90°). The a2 is.not.equal.to 90° due to the stiffness of the beam.

On the NON-RIGID joint panel (I think this is the Robot default), a0 may still be = 90° (when it is defined as the angle between beam and column at the joint or very close to the joint), but a1 and a2 is.not.equal.to 90°

(note a = alpha)

In other software like ETABS, we can specify the degree of rigidity of these joints panel to match the actual expected behaviour, I'm wondering (and really hope) if we can do this in Robot.

Thanks

Yakso

Artur.Kosakowski · ‎01-12-2016

In RSA you may define rigid links

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

Artur Kosakowski

Anonymous · ‎01-12-2016

Thanks Arthur,

This seems to be a solution, can we reduce the stiffness of the rigid link? In real design situation often we need to accomodate our joint panel not completely rigid and thus reducing joint panel is needed (e.g. steel joint panel). But if we can't, I think we can still use this and take the both extreme joint rigidity settings as our boundary conditions (thus upper and lower bound for sort of sensitivity check).

For the model I'm working on, I managed to apply length offset for the beam (applying 'reduced length' for the member length under relative offset) and it seems to work (I compared it with ETABS). Can I use this instead of rigid link as this is simpler?

Yakso

Artur.Kosakowski · ‎01-13-2016

The offset you used is in the principle equivalent to the rigid link solution I proposed. If the obtained results are as you need them then why not use it?

If you want to play with the stiffness you may divide a beam/column in some distance from the common node and apply a user section with appropriate value of IY on this short part

or define a realease with required relation between the rotation and bending moment

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

Artur Kosakowski

Anonymous · ‎01-13-2016

Thanks Arthur,

Yes this works, but if I may make a suggestion, why don't you add a feature to simplify the process (maybe something you can incorporate into member offset definition - reduce length (already there) and reduce stiffness maybe?) - just a suggestion, maybe something to be incorporated into React?

I think I can use the offset option for most of my cases.

Cheers,

Yakso

Artur.Kosakowski · ‎01-14-2016

If you mean reduce IY (AX etc.) for steel profiles then it is already on the wish list.

Artur Kosakowski

Anonymous · ‎01-14-2016

Thanks Artur, I wish we can have joint panel stiffness definition function (like in ETABS) for any type of frame structure (regardless the material).

Artur.Kosakowski · ‎01-14-2016

Could you attach the description of this option in ETABS please? Thank you.

Artur Kosakowski

Anonymous · ‎01-17-2016

Hi Artur

The following link gives explanation of the two features available in ETABS and how to use it:

https://wiki.csiamerica.com/display/etabs/Panel+zone+and+rigid+offset

Definition of end offset:

http://docs.csiamerica.com/help-files/etabs/Menus/Assign/Frame/End_Length_Offset.htm AND

http://docs.csiamerica.com/help-files/sap2000/Menus/Assign/Frame/Frame_End_Length_Offsets.htm

test case of end offset:

https://wiki.csiamerica.com/display/tp/End+offsets

Explanation on panel zone:

https://wiki.csiamerica.com/display/kb/Panel+zones

I really appreciate that you and Autodesk go this far to accomodate our wish. 🙂

Artur.Kosakowski · ‎01-19-2016

Thank you for these links. I assume that ETABS generates elastic releases but what I'd like to find out is:

1. How does it calculates stiffness of the spring (moment - rotation relation) for the two first options indicated on the attached picture?

2. Where are these releases then applied:

- at the beginning (intersection between column and beam) of the rigid link created for a column (1) ?

- at the beginning of the rigid link created for a beam (2) ?

- at (1) and (2) ?

- at the end of he rigid link created for a beam(3) ?

- at the end of he rigid link created for a column (4) ?

- at (3) and (4) ?

Thank you for your help.

Artur Kosakowski

Anonymous · ‎01-19-2016

Hi Artur,

I think your questions are explained in the 'CSI Analysis Reference Manual' which for copyright reason I cannot share it here. However if you somehow manage to get the document, Chapter VII The Frame Element - Section "End offsets" at page 111 to 115 explains about end offset.

Unfortunately I can't find any explanation on the Panel Zone other than from the links I have posted previously.

Yakso

Anonymous · ‎01-19-2016

I also found this, hope this helps:

http://docs.csiamerica.com/help-files/etabs/Menus/Define/Section_Properties/Panel_Zone/Panel_Zone_Da...

Yakso

Artur.Kosakowski · ‎01-20-2016

Thank you. This looks like an elastic release in (2) and with the stiffness of the spring for the two first options being calculated in the similar way to

http://help.autodesk.com/cloudhelp/2016/ENU/RSAPRO-UsersGuide/files/GUID-F2BB8C3F-062C-46D3-B3E3-DD3...

Am I correct?

Artur Kosakowski

Anonymous · ‎01-20-2016

Yes it looks similar to that.

Artur.Kosakowski · ‎01-25-2016

I have added both options to the wish list for next versions of RSA.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

Artur Kosakowski

Anonymous · ‎01-25-2016

Thanks Artur

Anonymous · ‎10-21-2016

Hello,

It is a fascinating subject! if I understood correctly, I have not found anything in Robot to take into consideration the effect of rigid ends.

In the ASCE/SEI 41-13 (Seismic Evaluation and Retrofit of Existing Buildings) Chapter 10 (Concrete Structures), it defined how to consider the joints between columns and beams.

Please take a look:

If we do not consider such effect the values of the internal forces will be different and, it could lead to mistakes.