Here is a nice link for barrier cables for vehicles.

http://pti.mountaininternet.com/reportFiles/28-Technote14.pdf

If I were to guess, it is going to be hard to meet the defelction requirements.

 

If you push your 4" sphere hard enough, any cable, regardless of size or pretension, will eventually separate enough to allow it to pass through.

"The biggest problem is that all this tension is loading a new steel beam"

How is the cable anchored at the ends?

Posts/verticals should be spaced a maximum of 3 feet apart. Cables should be spaced a maximum of 3 inches apart, straight runs and no more then 70 feet. Tension the cables till you can not deflect them more than 4 inches apart.  Consider using Quick-Connect® SS fitting:

http://www.cablerail.com/standard/partslist.shtml

Consider a tensioning sequence, start at mid (center) of post work to top and bottom as shown in:

http://www.cablerail.com/standard/installation.shtml#tension

I don't think you could limit deflection to 1/4" in a 7' span with any reasonable cable tension- something needs to change there.

notice boo's addition of verticals ... 7' sounds like a long way ... the guard would morph into a sling-shot

shobroco:
That is not a particularly good design solution, except to pull the 2nd fl. beam down, or the 1st fl. beam up.  The forces don't have to go to found., they are all resolved internally in that frame, as long as the two beams and their columns or bearings can take the loads.

However cute or clever the Arch. thinks it looks, it just doesn't work well.  You can't possibly tension the 3/16" cable enough to prevent the 1/4" lateral deflection.  Do the statics: one leg of the right triangle is .25"; the other is (3.5' x 12 or 42"); that's 168 to 1, and the forces in the cable and laterally will be in these same proportions.  A 2kip cable force will resist about 12 pounds lateral, and in the process the cable force will increase to (168.003/168)( 2k) or 2.04kips.  You can't practically find a stiff enough beam and cable system to make this guard system work.  Think about the cable for a minute, it just will not carry a lateral load until it is deflected a significant amount in proportion to its span length, and then it still takes a substantial force and reaction along its length.

That will make one hell of a harp though, if you tune it correctly.

Agreed,

If deflection is critical then cables are not the way to go.

A good Senior engineer would steer the architect away from this type of solution before it was fixed in their vision.

Propose a couple of solutions.  

Maybe thread the cables through a couple nice 2x6 railings, drilled at 3.5" o/c, and held at 2' and 4', or some such, from the 1st fl.  At least then you are limiting cable deflection along the length of the 2x6's (but, the 4" ball may still pass through) and starting to make the whole system act as a cable net, which will act better with less cable tension.  The railings will bring 5 or 10 cables (you determine) into play against the primary loading concern, but req'r. a lot of hardware.  There are several railing systems on the market, with hardware for this type of application.  You do the design and digging.

Most of highway stuff talks about energy absorption and a lot of cable stretch and deflection, and that's good for them because it does absorb energy, without a solid (brick wall) stop, and that's what they want to save people.  That may teach you something about cable structures, but doesn't fit your needs very well.  There is some literature out there on cable structures.  

I told the architect right away that it was going to be pretty difficult to do, but you know architects.  I can't oppose the forces between the top & bottom beams because he doesn't want to see the posts at the top & bottom of the stairs; they have to be hidden in walls further away.  The cables are threaded through holes in the bottom beam flange & held by swaged fittings, and tightened by turnbuckles connected to the top beam & hidden in the bulkhead.

I told him he was going to have to tune it, since the first inclination of anyone using the stairs will be to strum it.

Thanks for the links boo, those fittings are much better than what the architect proposed, but it think nutte is right: a 4" sphere is going through regardless of the tension.

I guess I have to sell some aesthetic modifications to the architect.

nutte is exactly right.  The spacing between the cables is the key, not the tension.  You need closer spacing, so that your tension is reasonable.

If you go to a 2-1/2" spacing, you can then have an 11/16" deflection in the cable.  At present, you can only have 3/16" deflection, so you can cut your pre-tension by a factor of more than 3 if you change the spacing.

I have been involved with a few of these in the past. Generally what we have ended up using is a relatively stiff 3/16 dia 1X19 cable with intermediate posts spaced about 32" o.c. If the posts don't fit in your arch plan you can also use thin 3/4 x 1/8 flat bar spaces 32" to distribute loads/deflections over the set of cables.

The vertical spacing never exceeds 3" such that under reasonable conditions a 4" sphere will not pass through. We usually spec 300 to 400 lb cable tension each but I have no idea if anyone has ever read the note much less used it. End post moments from cable tension sometimes control. After pushing and climbing on them in the field I am much more comfortable with the concept. This is really an easy problem to over analyze (as I have in the past).

where did 1/4" deflection come from ?

i'd set up a test for the arch to look at ... show him how easy it is to deflect a 7' span cable.  let him wind up the tension, up a fish scale on it so he doesn't overload it.

would a clear plastic post be sufficiently invisible ??

"This means they must be tight enough that they can't be pushed more than 1/4" sideways"
This is open ended, what force must be resisted?

Think here about the practicality of tightening the cables; tighten one to a design tension, tighten a second one and deflects the members and reduces the tension in the first one, etc. etc.

Think of the stringer size, Using dhengr's 2k, even though it looks inadequate, you have almost 7k/ft upward preload on the stringer.

The good part about this preload is that a lateral force on a couple of wires would shorten them, increasing the uplift, causing a deflection which would ease the preload in the adjacent cables. That is, any cable that tried to shorten would tend to pick up increasing amounts of the total preload.

Michael.
Timing has a lot to do with the outcome of a rain dance.

The 7' post O.C. is excessive, the spacing between posts and or braces should not exceed 42" max.  Using a maximum vertical wire rope spacing of 3" free opening between cables when they are installed will allow you to met the code requirement that a "4" sphere shall not pass through any portion of a barrier on a guardrail/stair rail".

Attached is cable deflection and loading info

• http://files.engineering.com/getfile.aspx?folder=f2d86a2d-5be7-4504-af40-75

other reference

• http://files.engineering.com/getfile.aspx?folder=e36ef90a-78fb-4b62-a5c1-d7

third reference

• http://files.engineering.com/getfile.aspx?folder=ecfafe82-1e31-441c-95b0-b5

Page 13 of Boo1's first reference gets to the heart of what I was saying earlier.

The 2006 International Building Code (IBC) and 2007 California Building Code (CBC) require that guardrail intermediate railings be spaced so as to prevent a 4" diameter sphere from passing between the rails (IBC/CBC 1013.3). However, the code does not state what load is to be applied to the 4" diameter sphere. While such an oversight is not critical to solid railing members, it is of utmost importance in flexible railing systems such as wire-rope cables. In the absence of Code guidelines, a rational load must be developed.

They proceed to apply the 50 psf infill load to the area of a 4" circle, and double it for impact, resulting in 8.7 pounds force trying to push this 4" sphere between the cables.  This is a rational approach, but the end result, 8.7 pounds, seems low to me.

My understanding is the 4-inch opening, is based on anthropometric data showing that 95% of all children 4 months of age and older would be prevented from completely passing through a 4-inch opening.

to keep people off the cables, i suggest barbed wire (maybe razor wire) ... it'll also help in keeping people from going through the wire (creating several hanging up opportunities) !

We set up a test cable today & at 500lb pre-tension 10lb lateral force deflects a 7' cable about 3/4".  3/16" stainless rod was almost exactly the same.

I just talked to the architect again, he really doesn't want visible posts, so closer spacing seems to be our best option.  The 4" spacing is to prevent children getting through it or stuck in it, the load that the guard has to carry is unrelated to the force required to push a 4" sphere between the cables.  The cables will easily carry the required lateral load directly, although our code (Ontario) like IBC does not have any value for the force on the sphere.  

For rb1957, the 1/4" deflection is the amount that adjacent cables have to deflect to pass the 4" sphere.  Maybe your razor wire is best.  I don't think I'd push on it with 8.7 lb force.



 

Boo: the links are great, & will probably lead to over-analyzing as flgulfcoasteng says he has done.  Thanks to everyone for their input.

Since we cant find any code requirement for the force on the 4" opening, ducument you engineering judgement using an industry recommended force as your standard.  I would use the documents attached as a guide for the engineering file.   

One option that hasn't been considered is that you don't necessarily need posts to keep the cables from spreading apart. You only need links between cables. These could be aesthetically designed.

Compositepro:  Thanks for the good suggestion, I'll try that on him.

It didn't take long to get a response: our code does not permit any horizontal member between 140 and 900 above the walking surface that will facilitate climbing.  If links at 900 vetical spacing won't work I guess razor wire would still meet that requirement.