Strange Problem 1

[geoffdale]

Hey all- here's a weird one (at least to me.) I have a client who is looking to install a 34" i.d. 38" o.d. plate (he's suggesting 1/2" thick as a starter), supported by tensioned cables supported by the tank of a storage tank (supports would go to grade at the outer support points). They want to use this as fall arrest, but also support some equipment from it. Has anyone got any data/examples of how to tackle this problem? I am thinking that the "ring" would be a flat plate beam supported by the cables at 1/4 points of the ring circumference- but this is coming up with a thick plate, and doesn't seem right in my head. See the attached chicken scratch....

Thoughts?

 

[paddingtongreen]

Before anyone can answer they need to know how the load is applied to the ring and the slope of the cables, vertical lift and horizontal run.

As a first shot at the horizontal pulls I would imagine four straight pieces, Then consider each bent to simulate the curve and see what moment comes up. Then you need to look at the vertical loads and bending.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[geoffdale]

The load is applied as a vertical load to any point on the ring as they hang the equipment from it. The cables are flat or very very minimal slope.

 

[paddingtongreen]

The cables cannot be flat. The starting angle is critical to finding the solution. The flatter they are, the greater are the forces in the cables. It looks as though an iterative solution may be necessary, placing the load at different positions on one of the segments. Our colleagues still practicing and using computer solutions may have a way around this iteration, or to automate it.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[MikeHalloran]

Tensioning the cables to 'nearly flat' and then hanging a weight from the central ring should be a fairly efficient way to buckle and collapse the tank. ... efficient in the sense that a smallish load can induce a great deal of tension. I have used the basic principle to pull a car out of a ditch by pulling up on the center of a rope tied between the car and a fortuitously located tree.

Find some possibly illuminative equations near the bottom of this page:

http://en.wikipedia.org/wiki/Tensile_structure

Mike Halloran
Pembroke Pines, FL, USA

 

[Triangled]

"Fall arrest"... In agreed addition to all of the foregoing, what's or who's, "fall" is to be arrested?

 

[geoffdale]

Thanks all. Spent some time going through this tonight and found that Roarcks is the closest analysis for the ring.
The essentially flat comment is related to a 2-3' sag on a 100' or so diameter tank due to restrictions. We know this will put a large load on the tank-hence my previous comment on adding reinforcing angles (or something at the connection points to the tank.)
Fall arrest was for people.

 

[graybeach]

Why not have the cables be continuous across the tank? If the ring is truly needed, it could be attached via u-bolts to the cables at their intersection.

 

[dhengr]

Geoffdale:
Graybeach’s approach is the best solution for the cables, since it takes the nasty issue of the ring stresses out of the problem, and saves you a bunch of cable hardware. But, you will still have much trouble making this cable arrangement work because the 100' dia. tank is just not made, nor can it be easily strengthened, to take the large concentrated reactions that the cables will induce on it. Can you strengthen the roof structure and hang this load off of it?

 

[paddingtongreen]

At 2'-3" and 50", the 4.9k load would induce a tension of about 110k in the cable. The cable would sag! the top of the tank would deflect inward and 2'-3" would increase considerably, maybe dropping the person on the ground with a bump.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[Ron]

For fall arrest you'll have to add another 5kips.

 

[BAretired]

The concept is wrong. Try another approach.

BA

 

[rb1957]

"Wrong" ? ... really inefficient certainly, fatally flawed probably ...

to get an idea of the cable loads, assume they deflect 0.6deg, ie sin = 0.01.
so to carry 5,000 lbs vertically, the 4 cables will need to develop 500,000 lbs ... does that scare you ?

Quando Omni Flunkus Moritati

 

[geoffdale]

Thanks all for the replies and they are confirming our gut feel. We are pushing our client in another direction as this was an idea that they suggested and we ran the #s, and came up with the same conclusion.