Complex System Boundary Conditions

[George Zucas]

Edit: Sorry about the vague title, it was intended to be complex beam system boundary conditions but somehow it turned out like this.

Hello,

I am trying to learn complex beam system designs and I sometimes struggle to assign boundary conditions. For example I am trying to design the lifting system in the picture (only one of them). I made it the same for the most part; 12 I-beams attached to each other and diagonal pipes inbetween at the sides. Now I'm defining boundary conditions. There are a few options in my mind and they all give vastly different results. Here are some of them:

I made it such that the I-beams under the structure carry all the load (distributed over the length) and fixed the upper four corners. While this give plausible results, this creates only vertical forces and therefore the diagonals carry little to no load. This doesn't take into account the forces created by the angled ropes. So I don't think this is a good model.

Then I modeled the ropes as beams, disregarding their stress values since in this model their job is to transfer the load only. I don't worry about the ropes at the moment. Then I fixed the system in the hook of the crane. Since the hook only prevents translation I made this end as free in terms of moments. No translation allowed but that is all right? The ropes are free to rotate. This gives me some insane values for combined stress in the beams, hundreds of thousands of MPas. If I fix the top end completely ( no translation, no rotation) then it gives plausible results. Though I don't know how correct this is, there shouldn't be any moment resistance there.

Third, I only designed the beams and instead of ropes, I inserted appropriate forces instead ( removed the distributed load). But then where to fix the system? If I fix the lower corners then the beams under the huge structure carry barely any load since the system is supported at the corners.

As you see, I am at a loss. How do you approach to assigning these conditions? How should I proceed?

Any clue to lead me to the right path is appreciated. Also is there any good book that goes into detail about these systems? In school we pretty much only analyzed very simple 1-2 beam systems with pinned supports. My textbook is the same, same exercises over and over.

 

[SteamKing]

Edit: Sorry about the vague title, it was intended to be complex beam system boundary conditions but somehow it turned out like this.

Hello,

I am trying to learn complex beam system designs and I sometimes struggle to assign boundary conditions. For example I am trying to design the lifting system in the picture (only one of them). I made it the same for the most part; 12 I-beams attached to each other and diagonal pipes inbetween at the sides. Now I'm defining boundary conditions. There are a few options in my mind and they all give vastly different results. Here are some of them:

I made it such that the I-beams under the structure carry all the load (distributed over the length) and fixed the upper four corners. While this give plausible results, this creates only vertical forces and therefore the diagonals carry little to no load. This doesn't take into account the forces created by the angled ropes. So I don't think this is a good model.

Then I modeled the ropes as beams, disregarding their stress values since in this model their job is to transfer the load only. I don't worry about the ropes at the moment. Then I fixed the system in the hook of the crane. Since the hook only prevents translation I made this end as free in terms of moments. No translation allowed but that is all right? The ropes are free to rotate. This gives me some insane values for combined stress in the beams, hundreds of thousands of MPas. If I fix the top end completely ( no translation, no rotation) then it gives plausible results. Though I don't know how correct this is, there shouldn't be any moment resistance there.

Third, I only designed the beams and instead of ropes, I inserted appropriate forces instead ( removed the distributed load). But then where to fix the system? If I fix the lower corners then the beams under the huge structure carry barely any load since the system is supported at the corners.

As you see, I am at a loss. How do you approach to assigning these conditions? How should I proceed?

Any clue to lead me to the right path is appreciated. Also is there any good book that goes into detail about these systems? In school we pretty much only analyzed very simple 1-2 beam systems with pinned supports. My textbook is the same, same exercises over and over.

First of all, designing a complex lifting system involves more than just looking at the load support beams. You should also be knowledgeable about the rope slings and how they are attached to the beams.

It's not clear what tools you are using to analyze your lift; i.e., are you trying to do these calculations by hand or are you using software. If you are using software, you should try to make a rough calculation by hand in order to check what comes out of the computer.

Wire Ropes should not be modeled as beams in any event, since they are capable of supporting only tension loads. Due to their built-up construction, it is important also to know their max. working load and correct Young's modulus, since the latter may be different from ordinary steel.

It's hard to tell from the image attached, but the diagonal braces on the vertical sides of the basket which supports the load appear to be there to provide stability and to absorb any side loads applied by the angles in the wire rope slings. You can't ignore the angles which the slings make as they run from the basket up to the crane hook, an it is important that the attachment points for the slings on the basket are designed properly so they do not collapse under any bending moments produced by the tensions in the rope slings.

These are just some of the areas of investigation off the top of my head. Without looking at more than just a single photo taken from very far away, it's hard to say if this is everything. We haven't even touched on things like welding and such, which are just as critical as properly sizing the members in the baskets.

 

[George Zucas]

Hello Steamking,

I know that it is more complex but I am progressing slowly to understand everything correctly (I've very recently started my first job and I am basically in a learning phase). The reason I've selected something like this is that is a good example of what I'm struggling about. We are doing similar designs at work a lot (not the same use case but we also use beams in various configurations and do calculations on them).

I am actually using FEM software to find stresses and deflections. I try to do hand calculations as much as possible and if there is a solution by somebody else (I sometimes study older projects if the solutions are available) I do cross-checking.

I don't know how to model ropes in the program (or even if there exists such an option, it is a very specialized program with little to no documentation so it is a bit difficult to use) so I used beams thinking it would be OK but thinking after your comment you are right, beams also translate bending moments although there shouldn't be any due to the rope. So my best bet would be, instead of ropes, inserting forces which I also tried. However if I insert rope forces, where is the support for the system is going to be? It is basically standing in the air. I mean obviously the reaction will be on the big structure but it is not like it is welded there the beam under the structure will bent as well, How would it be modeled? I am thinking but haven't come up with something yet. It may be that actual 3d cad programs like Solidworks etc. may let you design the system 1 to 1 so that applying FEM would be easier ( the one I use is a bit different, you can only attach the beams to each other from the ends, the one like in the picture is basically surface contact only).

Anyway, thanks for the help. Now I realized that instead of using ropes, inserting rope forces would be a better option (at least with my tools). I need to find a way to model the bottom.

 

[SteamKing]

This is a pretty serious and complicated project with which to start your professional learning curve.

Are there any other senior engineers to whom you can go for advice? Who can review and critique your calculations and analysis?

 

[George Zucas]

Well it is not something given to me from the company, I've chosen it myself. Though I did a few designs myself, some of them more complicated than this. Though they were easier in one aspect, they all had pretty clear support points so it was much easier to model in both my mind and the FEM software. This type of problem has been in my mind for some time now. The problems in which the main structure is not really fixed to any place for support but supported by surfaces. I wouldn't be able to do analysis on a single bolt either.

There is only one engineer and he is the general manager and while he always tries to help, he is busy most of the time and I don't want to go to him for ever single problem. Though my reports are both reviewed by him and the company we are working for so they are checked at least twice, so no risky business there :). Though learning has been a bit difficult for this reason.