Weird results of calculating arc constructions (innormal bending moment)

[Anonymous]

Hello!

Sorry for my English, it's not very good 🙂

I'm very fresh with using Autodesk Robot program. I'm now finishing University in specialisation of civil engineering.

I have a Diploma research about an 100-years old bridge. I tried to make a flat construction in Robot which is adequate to real one and i have two big problems here.

One is unnormous huge bending moment, few times more value than a longitudinal N force, which is very weird in arc constructions (it's missing its puropse ;). I don't know how to interpretate this because it's absolutely weird to me.

And second one is the tension from bending moment is completely wrong calculated from bending moments, cause with normal calculations it should be about 15 MPa , not 0,85 🙂 (Simple division of bending moment by strenght ratio of the profile). Am I interpreting results wrong, or it's some issue?

I'm attaching rtd file with my case,

Thank You very much for further help!

Kind regards to You all,
Maciek

PS. Sorry for bad placing the loads, just use first case, it's all made in "own weight" because i program was showing me only results from 1 section of loads, i am very fresh in this and learning this program only by myself.

[Rafacascudo]

I think the moments are normal. You have fixed supports and the arc is not very high.

About the bending stresses , I agree , it seems very wrong.

[Rafacascudo]

Problem is with tapered section . Setting a constant 67cm height section , then everything is calculated normally;

And it seems you are also right about the bending moment. How can it go from 4248 to 228 Kn.m???

It shouldn´t be so different the results!!!

Let´s wait for Artur , Rafal and Pawel!

[Artur.Kosakowski]

Hi @Anonymous and @Rafacascudo

I have made a few quick tests and it seems that for the chain of RC tapered sections Robot 'rotates' the sections by 90 degrees. The solution is to either add' artificial' 90 deg gamma angle (the very bottom mode or (IMHO better) to use the steel section type and change bars material to concrete (the middle one).

If I managed to answer your question(s) press the Accept as Solution button please. This will help other users to find solution(s) much faster. Thank you.

[Rafacascudo]

Hi @Artur.Kosakowski ,

    So , you are saying that when using RC beam tapered option , Robot displays correct values on Properties table , but when doing the math internally during analysis ,it takes Iz instead of Iy, making a terrible mistake???

Is that what is happenning???

So ,from this point on , everything else is compromised?? Analysis Results , RC design ,etc???

When doing the trick( steel bar type) , will it be possible to still proceed to RC design??

[Artur.Kosakowski]

Hi @Rafacascudo

So , you are saying that when using RC beam tapered option , Robot displays correct values on Properties table , but when doing the math internally during analysis ,it takes Iz instead of Iy, making a terrible mistake???

The values displayed in the table are as used in the calculations. For the components of a superber (RC tapered section assigned on a chain) they are 'reversed' for the same section assigned to list of 'simple' bars they will be as defined in the section label.

When doing the trick( steel bar type) , will it be possible to still proceed to RC design??

If you need to run RC design you should create number of tapered sections corresponding to each of the bar of the arch rather than apply it to the chain of bars.

If one or more of these posts answered your question, please click Accept as Solution on the posts that helped you so others in the community can find them easily.

[StefanoPasquini6790]

Hi everybody,

I'm really curious to see where this topic will lead us....

[Artur.Kosakowski]

Hi @StefanoPasquini6790

I do apologize if I hasn't been clear enough so far. The situation is that when you assign RC beam tapered section to a chain of bars the section properties are incorrectly assigned to the bars which create this chain/superber. For now there are two solutions I can think of:

1. Assign the equivalent steel tapered section to this chain and then change the material assigned to the bars of this chain back to concrete

2. Do not use the chain definition but assign individual RC tapered sections to each of the bar one by one.

If one or more of these posts answered your question, please click Accept as Solution on the posts that helped you so others in the community can find them easily.

[Anonymous]

Hello again and thank for You replies!

I wonder how did You changed the bending to the 228 kNm?

Because i afraid that the 1800 kNm in the middle of the arc would be never carried by RC section with Ned approx 200 kN and fi8mm bars each 10 cm up and down... And construction is standing over 100 years 😉 I've tried many combinations, and always Med is far too great to be carried by construction like this with that low Ned (i'm calculating it like normal compressed-bended construction, pole for example). I've tried various variations, even much higher arcs with less span, no car weight, and always there was something wrong with bending. Definitely too high. For me that's now important to only find the most stressed section of this arc, and i'll calculate its RC calculations by hand 🙂

The main reason to build arc structures is to minimalise effects of bending, it must be hardly compressed, and here i can't see it any way 😉 This is no real that arcs built of stone or bricks are standing when bending moments of that values are affecting them (far greater tension than materials streching resist...)

And another qestion, can i know local values in each bar of horizontal/vertical forces in one graph, not in global X/Z ? because these aren't accurate values to calculate the forces in the section. 

[Artur.Kosakowski]

Hi @Anonymous

Try to check this model.

If one or more of these posts answered your question, please click Accept as Solution on the posts that helped you so others in the community can find them easily.

[Anonymous]

Yes!

Now it's looking much more real. And what did You do with the model to acheive that?

And can I have another request to You, because i have not any idea how to make that, to place moving two point forces (2 metres span) car presentation and to find the most adverse position of it to the construction? Is this possible, or with that short bar elements of the arc it's only possible to do that with one-point force? 

And how can i make more than one of cases affecting simultanousely the construction to calculate? Because when i take in the window the numers of cases , example 1 3 5, program is only calculating one of them (that's why i placed all weights - construction weight, snow, and car in "self weight" position.) 

[Rafacascudo · Accepted solution]

@Artur.Kosakowski

I think this is a very serious issue that can lead to terrible consequences for us ,end-users , but also to autodesk , as it is an admitted mistake . As @Anonymous likes to say , It´s a disaster!!!

Easy to see the severity of the mistake looking at the support reactions.

2 models . 1st on top ,to the left ,the original of this thread ,with the correct model with steel type bars to its right side.

Below Them  is the 2nd model having a beam in which Iy is the stronger axis.RC tapered to the left and correct steel type model to the right.

In both cases , diferrences can be enormous !!!

Horizontal reaction Fx for the correct 1st model on top is about 12X greater than on the wrong RC tapered model!!!!

In the 2nd model The correct Moment My reaction is 31X greater than the given My reaction on the wrong model!!!!

   Of course,in more than 17 years , I used  the RC tapered beam applied to a chain of bars several times ,as it is a lot faster than creating different sections for each bar of the chain.

   So I will be checking most of my past important models ,seeing if there are any problems with any of them.

    That said , the minimum I expect from Autodesk is a fast action correcting this issue on a HF or a service pack , or at least greying the "tapered" option on the corresponding dialog window.

[Anonymous]

Ah, i see now 🙂

It's all about tapered shape. Because i tried it more times with new models, but i forgot that to new model it's automatically given material i've used last time and it was always tapered, i didn't change materials because i was looking only at the inside N,T,M forces, not tensions graphs - i thought that program is making these calculations lineary, without depending on material and its shape, then calculates tensions after it using material shapes - at least in 2d mode 🙂 So results were enormous anyway each next time - i didn't understand correctly Your answers because i was thinking it isn't only about tensions and tapered materials but always. But i thought "whutta!, it's impossible in program like that to make as big mistakes which can result very fatal to people who use it", but when it's only in certain cases now i understand 😉 And now i finally understood which was made with my arc to be corrected results. It's weird i'm the first person to discover this issue after all those years of program existence, do the engineers blindly trust computer programs now and don't try to interpretate of results logically? I don't even imagine how much $ more were the costs of making constructions from projects made using this function 😉

[Rafacascudo]

@Anonymous  , thank you very much to bring this issue for us!!!!

RSA is still a good software  . Maybe a little "cloudy" right now but we all hope that it will find a good way to regain its glorious past.

[Artur.Kosakowski]

Hi @Rafacascudo

I have notified the RSA development team about this program behavior and I hope (as all of you) for a quick fix but please be informed that it is not going to be my decision. 

@Anonymous

Please mind that if you use a tapered steel section type assigned to a chain of bars (superbar) and you want to assign concrete material to it you need to first create a new steel type material with concrete properties and then assign it to such created superbar. 

[Anonymous]

Thank You, i'm fresh in this program and i don't think that i'll manage to calculate the RC in it for now 🙂 it's important for me now that i have correct NTM values and i can find the most stressed section - i'll calculate its strenght by hand - it's simple math 😉 Can You help me with placing correctly two-points 2m span moving car force onto this arc? Because i don't know it's possible to do it on construction of 50 short bars ...

[Artur.Kosakowski]

Hi @Anonymous

Could you attach a sketch showing both the vehicle and its intended route?

[Anonymous]

Hello!

It's like that. I want to find the positions for worst cases to a upper section and a lower section, near the support. 

[Artur.Kosakowski]

Hi @Anonymous

The route should be defined as polyline following the nodes of the bars and you have to set the 'right' tolerance. Check the redefined cases 4 and 5.

If one or more of these posts answered your question, please click Accept as Solution on the posts that helped you so others in the community can find them easily.

[Anonymous]

Thank You, it works 🙂

But i still don't know how to combine two cases into one result, because i can see the NTM for case 1 or case 4 etc, but when i check in the cases list for ex. 1 4 , it's still results of one of them, and i can't see combined NTM. Must i sum them, or there is some way to activate them simultanousely?

Because i don't see any difference in N force depending of variant of load (2x240 kN or 1x320 kN). Moments are changing when i choose for ex. 1 4 or 1 5, but N force - not.