Steel cables for lateral stability - pre-tension force? 1

[nivoo_boss]

Hey everyone!

Let's say I design a steel frame building and I would like to use cables for lateral stability. How should I determine the required pre-tension in the cables?

 

[human909]

Not enough information...

At the very least I'd expect the cables to remain in tension for serviceability loads. Consider keeping them in tension for ultimate strength loads but this really depends on the design. High preloads mean greater loads on the compression members for non lateral load combinations.

 

[nivoo_boss]

I'll just put some random steel building here (I googled it). Let's say I want to use steel cables/wires in the roof and walls for stability (cross-braced).

 

[centondollar]

The easiest way is to install cross-bracing bars with a connection (slot-type) that only allows tension to develop in one of the two bars at a time, depending on the direction of the lateral force and deflection. In theory, it is doable with cables, but it would probably be more expensive to install and detail the connection for them.

In short, it is not useful to add cables into a one-storey steel frame. Guyed masts (used for e.g., power cable towers) are more typical applications of tension cables, as are bridges of various types. Stick with the ordinary tension/compression (preferably only tension) bars.

 

[nivoo_boss]

"The easiest way is to install cross-bracing bars with a connection (slot-type) that only allows tension to develop in one of the two bars at a time, depending on the direction of the lateral force and deflection. In theory, it is doable with cables, but it would probably be more expensive to install and detail the connection for them.

In short, it is not useful to add cables into a one-storey steel frame. Guyed masts (used for e.g., power cable towers) are more typical applications of tension cables, as are bridges of various types. Stick with the ordinary tension/compression (preferably only tension) bars."

I agree with you. But still, I would like to know the design process when using cables.

 

[STrctPono]

I was part of a pre-engineered steel building project almost exactly like the one you posted several years back. The manufacturer never pre-tensioned the threaded rod lateral bracing. They just snug tightened it and had a note on their drawings that the building owner would need to come back in a few years and re-snug the rods since they would loosen. During our final inspection, many of the rods were sloppy and loose which we asked them to go back and retighten.

In regards to your general question. The answer depends on the industry you are working on and the importance of the structure. For instance, why pretension the bracing rods in a pre-engineered metal building? They are tightened just enough to remove the slack so that the rod can engage when the building displaces out of shape. If you were to pre-tension the rods to levels that were above the expected tension loads due to your design wind event, your tension rod would theoretically not elongate and the building would not displace. But this is overkill for this type of structure. Other structures that utilize "pre-tensioning" of the cables/rods for stability would be those that cannot tolerate the expected levels of displacement.

Does anyone know if you can "pretension" cable wire rope? I am not aware of cable being pretensioned in any sort of quantitative way. I may be wrong. My experience is only with stressing PT strands/tendons or rods. I know you can torque tension rod as well without a hydraulic jack.

 

[BridgeSmith]

I wouldn't think you'd want to pre-tension the cables. It would decrease their design capacity, since you'd have to subtract the pretension from their ultimate stress capacity to get the design capacity for applied loads. As human909 pointed out, it would also reduce the usable capacity of the compression members that have to resist the preload force in addition to the applied loads.

I would think you'd want them snug, with no slack, when the building is erected, and all significant dead loads are in place, but not under any significant tension. You could try to figure out the amount of tension required during erection to accomplish this, but it seems to me it would be impossible to get it right, considering the number of variables involved. I would think it would be much better to adjust them to snug-tight at the latest point in the construction process that they are still accessible.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[azcats]

Cables on guyed towers are tensioned. The code give a few different methods for measuring tension, one of which is a shunt dynamometer.

[URL unfurl="true"]https://www.straightpoint.com/clamp-on-line-tensiometer[/url]

Another method involves jerking on the cable and measuring the time for the wave to travel up and down.

 

[bones206]

I used to build Butler PEMB's as an ironworker. We would tension the rod bracing as needed to square up the building and remove any sag. There was no explicit consideration of pretension force.

 

[Settingsun]

You would need to assume a stiffness of the cables then have the cable size and pretension set to match that. You'd also need to consider temperature change to make sure the stiffness isn't significantly changed or account for it.

Here's a link with equations and graphs for effective stiffness. I've never used these particular equations (just first search result and assumed correct) but the graphs tell the general story at least.

https://www.google.com/url?sa=t&sou...8QFnoECAQQAQ&usg=AOvVaw26aE7pZDr5nj5voNrp74fd

 

[nivoo_boss]

"You would need to assume a stiffness of the cables then have the cable size and pretension set to match that. "

What do you mean by that? Let's say I know A and E of the cable, then what? EN 1993-1-11 actually has equations for effective E for tension members, including cables.

 

[rb1957]

why pre-tension the cables (beyond hand tight) ? one diagonal works for shear in one direction, the other for the other ... both don't work (or have to work) at the same time.

another day in paradise, or is paradise one day closer ?

 

[Settingsun]

Thanks for the pointer to Eurocode.

For building braces (<20m), you don't need much pretension to remove sag but it isn't zero. See the graph below. But, if not controlled (as Bones206 says), you can get high force. At the top of the image, there's a few words about the outcome of a university test where they got their lab technicians to tighten up bracing rods.

You also don't want to be on the cusp of the brace becoming floppy on a hot day.

 

[MIStructE_IRE]

I would have an adjustable turnbuckle on there and have the cables hand tight at time of installation. Then retightened once the structure has been plumbed/lined. Then tightened every 2 years as part of ongoing maintenance - but always just hand tight.

Pretensioning beyond hand tight in my view is unnecessary. It also pre compresses your columns and introduces all sorts of knock on effects. Why bother? It does nothing for you anyway.

The much bigger issue you’ll have with cables is elongation. PL/EA.. I’d consider the use of solid bar tendons before cables.

 

[STrctPono]

Thanks azcat. I learned something new today.

I wouldn't think you'd want to pre-tension the cables. It would decrease their design capacity, since you'd have to subtract the pretension from their ultimate stress capacity to get the design capacity for applied loads.

I may be misunderstanding you but I don't see the pre-tensioning force reducing the ultimate strength of the brace. I think of it like this. A brace member pre-tensioned to 50 kips would see no additional stress from an externally applied load beyond it's pre-tensioning stress until that external load exceeds the pretension force. Even then, they are not additive.

 

[dhengr]

Nivoo_boss:
Stl. bars or rods have a regular P/A stress and strain component making up their stretch, and can be made up in most any fab. shop. Whereas, cables/wire rope have a P/A stretch component and a mechanical stretch component caused by the various stands tightening around each other, plus any sag straightening. This leads to much more elongation of a cable with the same stl. area. This is not a good situation in a bldg. where you are trying to limit the drift or movement at the roof. Do the trig. for a bracing triangle and the extra elongation of the cable diag. becomes significant in the amount of bldg. movement under lateral load. This leads to the thinking of pre-tensioning the cables, and additionally it takes a significant amount of cable tension to even start to take the visible sag out of the cables. Read the paper mentioned above by Steveh49 or look at some good Strength of Materials textbooks which covers basic wire rope design. Finally, good cable hardware and fixings to allow connections and length adjustment are expensive and can’t usually be done in just any small shop. The solid bars or rods are just easier to deal with all the way around. They should all be tightened to plumb and square the bldg. up. Then, when the bldg. is complete and settled-in, they should all be re-tightened so that they work properly for the lateral loading and bracing. A reasonable gage/measure of tightness/tension after basic tightening, might be some turn-of-the-nut method of measurement.

 

[Ingenuity]

A brace member pre-tensioned to 50 kips would see no additional stress from an externally applied load beyond it's pre-tensioning stress until that external load exceeds the pretension force. Even then, they are not additive.

I agree with STrctPono. Somewhat analogous to a pre-tensioned anchor rod or bolt, where you do not subtract the pretension from the capacity of the rod/bolt.

P/A stretch component and a mechanical stretch component caused by the various stands tightening around each other,

The "mechanical sketch component" is usually dealt with via a modified (reduced) elastic modulus that accounts for the 'stranding' effect; eg: solid steel bar: E=29000 ksi; 7-wire strand: E=28,500 ksi

 

[human909]

I wouldn't think you'd want to pre-tension the cables. It would decrease their design capacity, since you'd have to subtract the pretension from their ultimate stress capacity to get the design capacity for applied loads.

Most tension structures don't operate like this.

I may be misunderstanding you but I don't see the pre-tensioning force reducing the ultimate strength of the brace. I think of it like this. A brace member pre-tensioned to 50 kips would see no additional stress from an externally applied load beyond it's pre-tensioning stress until that external load exceeds the pretension force. Even then, they are not additive.

Exactly.

It is kinda counterintuitive in many circumstances. I recently designed something that structurally was very similar to a bicycle wheel. Compression forces on a bicycle wheel are resolved by a REDUCTION in tensions. You'll find negligible increase in the pretensioned spokes. Though this bicycle wheel though needed a 200T crane to lift...

Likewise when you have a tightly pretensioned crossbrace system then your lateral loads are actually resolved by a REDUCTION in tension stress in the tension members and normally a REDUCTION in compression stress in one of the compression members.

 

[Settingsun]

Tensioned rods/cables do have a stress increase due to external loads even before the stress due to the external load exceeds the pretension. The situation isn't the same as tensioned bolts because the bolts have plates completely filling the gap between the two ends of the bolt. There's effectively a parallel compression member which causes the net-zero force.

The easiest way IMO to understand tensioned members is to consider them as typical tension-compression members under the external loading (up to the point they approach/go into compression) then add the pretension force. For tension-only cross bracing, you get a 'compression' force in one of the diagonals equal to the tension force in the other. When the pretension force is added, you see the reduction of pretension under load. However, the other diagonal's load did increase unlike the tensioned bolt.

For the original question, the cable braces will act fundamentally the same as rod bracing provided there's the same pretension as in rods.

 

[Lomarandil]

I'll also second Steve's point about sag. In typical building applications and scales, this isn't usually a major factor, as most sag is removed by the time the cables are "hand tight". In larger scale applications (tower and temporary guying for construction are the two I've faced), it's possible that the cable has no effective bracing stiffness over the first few inches/cm of lateral movement, as the sag has to be pulled out first before any substantial brace reaction force develops.

----
just call me Lo.