Changing structures over time 3

[Danielsp]

Hello folks

I am a civil Engineer, M.SC. in structures Engineering and I model and design structures, mostly using FEM.

Right now I am interested in static modeling structures that change over time. Some cases are really simple, for instance:

1) If you load a structure then remove some part of it while keeping the same load, then the new moments, shear forces and strains can be easily obtained by simply ignoring the original structure. You load the new structure and that's it.

2) If you apply Load 1 on Structure 1, then add a structural element to it and finally apply a new Load 2, then the final state of forces and displacements is obtained by adding L1 on Structure 1 to L2 on structure 2.

So far, no big deal.
But what if after the first load you add a new part to the structure while loaded and then remove an old one?

 

[phamENG]

No worries...I'm not from Brazil...different terminology and spelling is used in different places. Gets in the way a lot on an international forum.

 

[dik]

A couple of projects come to mind... one about 30 years back, the client had a large PEMB structure. He wanted to remove one of the outside supports for a new door. There was a lot of reinforcing required of the exterior bay and the third one in.

Just recently I looked at a PEMB added roof structure. It was a PEMB where the exterior supports were on a substantial masonry parapet. The parapet was about 5' high and the horizontal thrust was accommodated by using diagonal bracing angles to the floor. The client wanted to remove the diagonals. It was a 4 span structure and removing the diagonals would require reinforcing the entire PEMB structure.

So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Danielsp]

Brad805 and phamENG , I'm sorry... I just didnt see how a sketch would be relevant. It is a general problem, could be applied to any indeterminate structure you like.

phamENG, I have the proper software to solve indeterminate structures (

https://atirsoft.com/strap/).

But the solution always assumes that there are no locked in stresses.
I am just trying to figure out the general solution to problems like this and found nothing online.

 

[Settingsun]

Starting simple, consider a structure where you remove a support. This happens fairly frequently during construction. You can model this by applying loads to the 'new' structure (ie without the support in question), at the old support location, that are equal but opposite to the support reaction that previously existed. This gives the effect of removing the support. Add the results to the previous results from when the support was in place.

You can extend this principle to whatever you remove from the structure.

1) If you load a structure then remove some part of it while keeping the same load, then the new moments, shear forces and strains can be easily obtained by simply ignoring the original structure. You load the new structure and that's it.

Not the case when there are time-dependent effects like shrinkage and creep. What I described above is more general.

 

[Danielsp]

Yes, dik. That appears to be one instance of the problem at hand: a structure that you wish to reinforce before removing some part, and the load is present at all times.

 

[Danielsp]

steveh49, just removing is trivial. The thing is when you

1) Load the structure
2) ADD a new part without removing the load
3) Only then remove a support or whatever other structure part that you wish.

 

[phamENG]

As I said...terminology varies, but a sketch of the situation would have put us all on the right track from the get-go.

Does that software allow you to force a displacement of a support? If so, that's all you need. If it doesn't, then you'll have to do it by hand.

 

[MotorCity]

There seems to be alot of confusion in this thread in terms of what part of the structure is there, what part of the structure isn't there, what load is there, what load isn't there, etc.

At any point in time, there is a structure in place and a load in place. If there are phases of construction and/or demolition when any of the structure or the loading changes, you can simply create multiple structural models to reflect such conditions. Add a beam, create a new model. Remove a beam, create a new model. This approach will result in a snap shot of the structure at any given point in time. As you have probably realized from the confusion above, a picture is worth 1000 words, we cannot see what you are seeing.

 

[Danielsp]

rb1957, your take on determinate structures is right. But not on the indeterminate ones.
Fails-safe analysis refers to structures in which you only remove parts. How do you calculate stresses and displacements in the situation I described?

 

[Danielsp]

Yes, phamENG. I can force displacements. How would you proceed exactly?

 

[Settingsun]

steveh49, just removing is trivial. The thing is when you

1) Load the structure
2) ADD a new part without removing the load
3) Only then remove a support or whatever other structure part that you wish.

Apply the negative reaction load to whatever your new structure is, including your added part. That's the general principle.

Edit: I'll expand on how this works. Remove the support but apply the old reactions as external loads. You get the same results as when the support was in place. Now apply opposite loads. There is now zero net force at the old support location: it has been removed. But the negative loads were applied to the new structure with the added part, so the stresses and deflections due to the negative load are calculated from analysis of the new structure.

 

[phamENG]

I'm starting to feel like a broken record.

1)Analyze the original structure.
2)Determine the displacement at the point you plan to install the new support.
3)Run the model again with the support removed and a forced displacement at the new support equal to the deflection calculated in step 2.

 

[dik]

It's appreciated pham... the learning process requires redundancy.

So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[dik]

Another problem often encountered by engineers is changing load conditions... it's just part of the game we play.

So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Danielsp]

steveh49, by negative reaction load you mean minus all the loads exchanged between the removed part and the remaining structures?

 

[Danielsp]

phamENG, I'm not adding new supports, but new parts to the same structure. Would it work the same?

 

[rb1957]

"rb1957, your take on determinate structures is right. But not on the indeterminate ones." ... no, I understand indeterminate structures very well. The point to my "maybe" was in an indeterminate structure you can remove a structural element that can fail more than one loadpath, and the resulting structure could become a mechanism. I tried to cover al the "what ifs" (if I hadn't I'm sure someone would've pointed this out to me).

yes, fail safe is subtractive changes ... I used it as an example.

any changes to the structure mean rerunning the model. Ok, maybe not if you are just reducing the size of a member and reason that the change in loadpaths is negligible.

what is the big deal in rerunning the model ?

Something you can do is divide the model into "super-elements". These condense the structure to stiffnesses at the boundary. No change inside the super-element then no need to rerun.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[phamENG]

I'm done until there's a detailed problem statement and images/sketches to illustrate.

 

[Settingsun]

I'm not adding new supports, but new parts to the same structure. Would it work the same?

The principle is the same. If you removed a section of plate, you would apply loads opposite to the stresses on the boundary of the removed section of plate.

 

[rb1957]

like pham, I have no idea what the issue is.

I don't know why we're talking about removing elements are then applying their load.

Why would someone ...
1) have a structure and a set of internal loads, then
2) want to simulate removing an element by (with the unloaded structure) removing an element then applying the the reverse internal load of that element to get the effect of removing this element, then
3) adding the results of 2) to 1); instead of rerunning the external loads with on the structure without the element.

but OP says this is simple, and wants to know the impact of adding an element. Add your element and rerun your model.
You could maybe get somewhere by noting how the nodes (at the end of your new element) displace under your external loads and then apply this as enforced displacements on the model with the added element ??

But this is philosophy ...

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.