topic Time History Analysis results in Robot Structural Analysis Forum

Time History Analysis results

Anonymous — Mon, 25 Mar 2019 10:49:21 GMT

I am running a THA on a bridge to simulate a Train going over. The bridge is simply supported and 50m long. When i get my results they dont make any sence as the moment is way too small, i dont understand why. I should get something around 40 000 kNm, but end up with 8000 kNm. I am linking a Picture to it aswell. Thank you in advance for Your help.

Re: Time History Analysis results

Rafacascudo — Mon, 25 Mar 2019 11:56:53 GMT

By your picture ,it seems you are using bar offsets on the main girders to model them on their real position . In these cases , the bigger part of the additional moment is "hidden" on the couple formed by the girders and the slab normal forces . A smaller part is resisted by the slab itself .

If you want only the girder resisting almost all the load applied , delete the bar offsets .A small part of the acting forces will still be resisted by your slab.

If you want ALL the loads to be resisted only by the 2 girders ,then use cladding instead of a shell slab and make sure the concentrated loads are applied directly on the girder bars . Claddings can be very bad for distributing concentrated loads or distributed contour loads . Use them only when you have uniform distributed load on the whole cladding area.

Re: Time History Analysis results

Anonymous — Mon, 25 Mar 2019 12:56:51 GMT

Thank you for replying. It worked as i got 16000kNm now, but thats still very far from the 40 000kNm i should get. Do you have any other suggestions for what it can be?

Re: Time History Analysis results

Rafacascudo — Mon, 25 Mar 2019 13:12:06 GMT

can you send the model? Zip and attach it

Re: Time History Analysis results

Anonymous — Tue, 26 Mar 2019 08:35:03 GMT

Sorry my file was too big to zip. I made another that i have With here, both a text file and a regular .rtd file. hope you manage to open them. Thank you for Your time.

Re: Time History Analysis results

Anonymous — Tue, 26 Mar 2019 08:46:05 GMT

Sorry about the mess. Its probably easier if you put in the THA analysis yourself as the file gets too big after i run the calculations. I Have ziped the file, the function (as a text file for robot to read) and the parameters i want to use. Thank you very much.

Re: Time History Analysis results

Rafacascudo — Tue, 26 Mar 2019 12:19:33 GMT

You don´t need to send the file with THA results.

"Save as" "Robot - without results" with a different name for the file ( Bridge 1_NR ) .

File will be a lot smaller.

Re: Time History Analysis results

Anonymous — Wed, 27 Mar 2019 08:25:24 GMT

There we go, now i know this for next time!

Re: Time History Analysis results

Rafacascudo — Wed, 27 Mar 2019 18:38:54 GMT

I am not an expert on dynamic analysis but I noticed a few things in your model.

1- THA uses modal analysis in its method to achive dynamic responses . They can be greater than those found in regular static analysis for the same loading. Your modal analysis only considers the generated self-weight as contributing mass. You should also include all the dead loads "attached " to the bridge . You can do that converting the corresponding dead load case to mass using the below tab on analysis type dialog window

2- The 10 modes you set to be found(and considered in THA) mobilized only around 55% of the bridge mass in the vertical direction. It is desirable that you have over 90% of mass to have reliable results. Also , there are 2 separate structures on your model. Avoid that when you have modal and THA analyses to be done.

3- So I deleted the isolated beam structure . Also deleted all the beam offsets. I Set the modal analysis to find 50 modes only in the Z direction(reached 96,29% of the mass contribution) . I also refined the mesh to 50cm size but that didn´t make a much difference comparing to 100cm size you used. Higher beam moment was at 4,75s .

For the static case of a 3205,8 Kn concentrated load at midspan ,overall bending modent found would be 40072,5 Kn.m which is exactly found on a test load case . Moments of 19865,19Kn.m for each beam and 342,17 for the slab which sums 40072,55Kn.m . See below

For the THA component at 4,75s(highest beam bending moment) Sum of the moments reach 39252,82Kn.m (2 x 19466,32 + 320,18) which is only 2% below the test static load case. see below

With 100 modes 98,05% of the mass is contributing and Sum of moments go up to 40006,19 kn.m .

Are you following any recomendations for your THA analysis ?

I cannot see how a single variable with time point load on midspan can simulate the passing of a train.

Test model without results attached

Re: Time History Analysis results

Anonymous — Thu, 28 Mar 2019 13:56:09 GMT

Now i understand it better, thank you very much, great help! The point load is a equivalent force to the moment created by the train at all times in the mid span. Its taken from M=PL/4 -> P=4M/L. I use this since it is a simply supported beam and may be applied to the single point force "dynamisk last" in my file.