topic Help with BURGER support in Robot Structural Analysis Forum

Help with BURGER support

hvelasquezmast — Thu, 20 Jan 2022 18:51:13 GMT

Dear Colleagues, I hope you are fine.

I am trying to model a support of the BURGER kind (see diagram) by using an eastic support in series with 2 releases-in-chain, to evaluate foundation settlements, but I am running into trouble on the following 2 topics.

I'd be glad if someone could give me a hint so I could progress here.

i) Topic #1: I think I am doing fine by emulating the phases Epsilon 1 and Epsilon 3 (see diagram) in my .rtd file attached, since there is "springs-alone" or "springs-dampers-in-paralel" only, which allow me to define the corresponding realeases properly (in terms of their compatibility, elasticity and damping).

But I am having a hard time figuring out how to define the release at phase Epsilon 2, where there is a damper alone, with no spring to take control of elasticity. Program notifies an invalid calculation argument if elaticity k = 0 is used; and if elasticity k = high is used then the efect of the damper is nullified.

ii) Topic #2: Testing the releases-in-chain in phases Epsilon 1, Epsilon 2 and Epsilon 3, this time with springs only, it all works fine, but it still produces though certain notifications of node instability at some places.

Truely yours, Hernan

Re: Help with BURGER support

kevinj.888 — Fri, 21 Jan 2022 06:25:16 GMT

You could try to put a minimal restraint in the unsupported direction, correspondingly.

For example, if the instability is in the z-direction because your spring is free to move in z-direction, then set kz to 1kN/m/m or 0.1kN/m/m instead of leaving it as zero.

This will not provide significant restraint but it helps to 'stabilize' the support and prevent instability error.

Betreff: Help with BURGER support

hvelasquezmast — Fri, 21 Jan 2022 10:37:10 GMT

Thank you very much for your time, Kevin.

I have tried to follow the way you suggest, but I still keep having an instability Ty 2 in one of the nodes, which doesnt allow me the modal analysis (Error 2083).

I leave you a link to my .rtd file below, in case youd wish to examine.

https://drive.google.com/file/d/1VX72K-NhM0mmKBYvAejubnU_i8y1vL3q/view?usp=sharing

Thanks to you again. Hernan

Betreff: Help with BURGER support

Krzysztof_Wasik — Sat, 22 Jan 2022 16:09:40 GMT

Hi @hvelasquezmast

I have modified connection for nodes 1, 4 and 7 by copying them from other corener.

Please check model.

Betreff: Help with BURGER support

hvelasquezmast — Mon, 24 Jan 2022 11:59:41 GMT

Thank you very much, Krysztof. Very an intuitive solution. I'm sorry I did not think of it before asking.

Betreff: Continuing with the BURGER support

hvelasquezmast — Mon, 24 Jan 2022 13:09:38 GMT

Continuing with the BURGER MODEL elastic-damped supports you so genty corrected:

https://drive.google.com/file/d/1T2Gkd8OapR8y5RnaPqZvQpf3pv6MSPfz/view?usp=sharing

i) I run a Time Analysis on them, simulating the loading (and unloading) effect of the foundation soil, and the calculation algorithm notifies me that the damping coefficients defined within suports and nodes in the structure, will not be regarded.

I kind of expected this, since on prior steps towards calculation, the Modal Analysis Setup and the Time History Setup required both entering their specific damping coefficients. How would you suggest to deal with this "dupicity" of damping definitions?

ii) Diagrams in the Time History report are default-given in Seconds. Settement time of foundation soil is very long, and usualy measured in Days rather than Seconds. Can Units, and the general display of the diagrams, be adjusted?

Thank in advance. Hernan

Betreff: Continuing with the BURGER support

Krzysztof_Wasik — Mon, 24 Jan 2022 15:15:11 GMT

HI @hvelasquezmast

Time history analysis when Newmark method is used takes into account dampers defined in supports, releases and compatible nodes.

I do not see any warning in your model.

Default time unit (second) cannot be changed in Robot.

Betreff: Continuing with the BURGER support

hvelasquezmast — Tue, 25 Jan 2022 12:02:30 GMT

Thank you again: acknowledged.

I tried to upgrade the support (as shown on the pictures) to behave as a non-linear support with a direccional release U+ and instabilities/non-convergences started to be notified.

The Newmark (Acceleration) algorithm and P-Delta selection are being used.

Also, the time interval measured in seconds was grown, in order to deploy graphically slower settement behaviours.

Please, if you have the chance, could you check within the linked model, what appears to be generating those instabilities/non-convergences?

https://drive.google.com/file/d/1IoXqZ4GG6iikMzcVkl8-yeYIhKL9_rhJ/view?usp=sharing

Betreff: Continuing with the BURGER support

Simau — Tue, 25 Jan 2022 12:24:51 GMT

@hvelasquezmast

Look at Nonlinear Analysis Parameters

Betreff: Continuing with the BURGER support

hvelasquezmast — Thu, 27 Jan 2022 16:47:53 GMT

Thank you so much.