steel truss strengethening

[eslah]

Dear all,
I'm currently working on an old 3d steel truss in order to strengthen it. Du to resent renovation, the DLL became heavier and deflections appears on the roof.
The idea is how to strenghen this steel structure? I was proposing a solution with steel cables.
Can somone please help or guide me in this reflection?
In advance thank you.

 

[Teguci]

Keep it simple.

Add depth to avoid increasing compression on the top chord.

 

[eslah]

Thanks Teguci for your answer ;-)
It's a visible structure inside an airport and the client won't see the reinforcement...
Doing so, you are not going to create a moment on the top of the concrete columns??

 

[phamENG]

eslah, I think Teguci's plan is probably best. You'd have to do it from each corner to support the 3D framework, and a good 3d model would help make sure it can be passed through the existing frame without interference, but it's the simplest and probably the most elegant. With that kind of system, you'd be hard pressed to modify the individual members without making it look bad. Is the architect still on board? If so, work with them to come up with an aesthetically acceptable solution. You could probably find a way to make it blend with what's there.

 

[eslah]

Thanks phamENG!!
As you can see on the section here,

We need two ropes a long the main truss!
And what about the added moment on the existing columns?

In my first proposal, the rope was fixed on the top of the RC column, passing through a polley which is fixed on the top chord and going down to the buttom chords to pull them up and cancel the created deflection..

 

[phamENG]

I don't see any added moment. It looks like your frame is pinned to the top of the column - as long as your reinforcement connects back to the top, you wont have an eccentricity to create the moment. The added dead load may induce additional secondary moments, and if seismic is the controlling lateral force you may have additional lateral load to the structure, but that's about it. If you disagree, draw a free body diagram of the column with the applied loads and we can discuss what you come up with.

 

[Celt83]

If this is actually a space frame/geodesic grid type structure as suggested by your sketch labels then strengthening in a single direction may not be appropriate.

Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

 

[klaus.]

looks like a (Mero) Space frame
be carefull applying and additional forces....

 

[ldeem]

I don't have a better idea on the overall approach but a couple things to keep in mind.

1. Wire rope can stretch quite a bit. Make sure you get a good wire rope design manual and account for the rope elongation.

2. If you plan to take the deflection out of the current roof you will probably need to jack up the roof and then install and tension the wire rope system. Depending on deflection the jacking may cause problems with the existing structure. Trying to pull the roof up with the wire rope seems like it would mean the wire rope will be taking all the load.

3. Analysis will be tricky. You have two overlapping structures with different modulus of elasticity. Steel and then wire rope due to its construction will not deflect like steel. It will be challenging to figure out how the load is shared between the two systems.

4. I am not sure where this is but when I hear large roof deflecting I think of rain water ponding. If you have accurate existing deflection you should be able to determine any added ponding load. You might want to consider this in design in case taking the existing defection out of the roof proves too difficult during construction.

 

[IRstuff]

Aren't you're essentially trying to substantially stiffen a truss with something that has less than 10% the cross-sectional thickness of the truss itself, and the cable is probably not even as thick as the cross-members in the original truss.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[human909]

Aren't you're essentially trying to substantially stiffen a truss with something that has less than 10% the cross-sectional thickness of the truss itself, and the cable is probably not even as thick as the cross-members in the original truss.

None of which is inherantly a problem if appropriate materials, design and installation is used. ldeem raises some pertainant challenges.

A vaguely simlar problem was faced on a concrete expressway in my town. It was found to be understrength. Wire rope tensiooned up underneath was the solution.

 

[hokie66]

I know the OP wants to strengthen the structure, but is it really understrength? Deflection does not necessary mean that there is a strength problem.

 

[dhengr]

Eslah:
Those types of space frame/geodesic grid structures are particularly difficult to deal with or to do any modifications on. They, in total, in every piece, are very sensitive to any changes or modifications. And, there have been some fairly spectacular failures too. You have not done a very good job at all in explaining what you think you are trying to do in your sketch or of the nature of the entire roof structure for that matter. You will not likely correct this roof load/deflection problem by messing around internally within the truss, which appears to be what you are trying to do in your sketch. You are not going to lift the bottom chords up by pulling down on the top chords. I think that Teguci’s approach is generally the right way to do this, but it needs much study and attention to the details. Unloading or reverse loading these space trusses can be as tricky as loading (over loading) them. You will have to jack the roof up, giving it some camber, in effect to unload (destress) members. Then you can install your wire ropes or better yet tension rods and tension them. Now, when you lower the jacks both the space truss and the new tension rods will share the total load. And, reduce the deflections because the stiffness of the entire system has been increased by the tension rods action on the entire roof structure. Given your question and your presentation of it, you should probably not be doing this problem without some senior engineering supervision. This is a complex problem in its details, and there are many of them.

 

[Aesur]

Be careful with using cables in tension as shown in your sketch. This will generate high tension forces, which appear to go through a pulley and then to the top of the column. When calculating the tension in the cable you will need to account for the pretension due to self weight + the tension needed to "lift" the truss back up (the added dead load) and hold it in place (with no downward deflection these forces will theoretically be infinite in magnitude). Imagine a guy walking on a tight rope or a zipline, even with a highly tensioned wire rope, the rope deflects under the load of the man. You also need to consider elongation and fatigue under constant tension. Additionally, this will impart additional forces on the truss joints nearest the column potentially causing the webs to buckle/fail. Another concern is the tension forces applied to the concrete column, is there adequate reinforcing an anchorage to handle these loads?

If the cable is designed to resist the additional deflection due to the added DL during the renovation it theoretically becomes possible to reduce your forces, however determining these forces is near impossible (at least I cannot think of a way) because when does the cable take over from the truss, can you say the truss will carry the load until it deflects to the allowed limit and then the cable kicks in and helps? In reality the cable will be constantly working to resist the loads on the truss, I cannot think of a feasible way to calculate the balance point between the two.

I have attached a quick freebody diagram below.

 

[MIStructE_IRE]

Can’t you simply reduce the dead load? I think you will have a very hard time justifying this approach with cable strengthening.

 

[eslah]

Aesur; Thanks a lot for these explanations! Great.
It's exactly what I was thinking to do in these nodes.
Today, the roof is still on scaffolding !! It's ugly to see inside a terminal.
What I'm working on it is to tight the wire ropes untill they start to pull up the truss and stop over there! Than release the jacks.

 

[klaus.]

I worked for 20 years on space Frames like that one
I must repeat myself...be careful doing modification on an old space frame without redesigning the old structure. It is not just stress in the members...you need to check the bolts and especially member buckling and node stability

 

[hokie66]

Listen to klaus. As well as his concerns, It will be nigh on impossible using your proposed scheme not to introduce bending in the space frame members, and they are not designed as bending members.

 

[MIStructE_IRE]

You also need to account for losses in the cables. In doing so, you’ll need to tension them further at the start - and in doing that I’m not convinced that you won’t end up with a pile of matchsticks on the floor!

 

[eslah]

Klaus,
Thanks for you advices.
I'm now thinking to do so :

Did you see this before?