Helical Steel Stairs

[EngrRC]

Has anyone designed a helical steel stairs? Attached is a 3d model i am working on. It has a 7.3m horizontal span and 4.5m floor to floor height. There is an intermediate landing. The top and bottom connection points are somewhat aligned so you get like a 2D-truss. Any tips for what kind of section is most efficient and constructable? I am also getting high horizontal reaction forces due to the truss action.

Regards,
RC

 

[racookpe1978]

Any vertical supports for the flat platform between the two stair sections?

You're using tube steel, is it mitered (as shown) or rolled smoothly where there are no vertical displacements? Much more expensive to do the multiple all-around welds and cuts than rolling, it would appear. Better looking rolled as well.

Note, rolled helical sections on the spirals are a bit tricky though: You have to twist the tube steel as it goes down, which you do not have to do with simpler round helical sections. You'd have to twist the more common 6 inch C-channels as well.

Any architectural covers or wood or drywall over the bare steel?

You have a 3D model, can you convert it to a FEA assembly?

 

[EngrRC]

Thanks for the response racookpe1978!

There are no supports on the landing. The only vertical supports are existing PT beams at the floor levels (think directly above each other in plan). That picture already shows my STAAD model.

I am looking at various framing arrangements, hoping to arrive at the most efficient one. From what you see in the picture, there is one outer curve stringer and one inner curve stringer. I am getting high torsional moments on the outer stringer; hence, I added transverse beams to connect the two stringers. Thoughts on the generally efficient type of framing for this type of stairs?

Stringers are RHS 200x400mm. From the contractor's input, they will be using cold bending for the steel sections. I am not really familiar with the constructability of all this. Steel sections will be clad but I am restricted with 200mm wide stringers more or less, which is why I am using tubes instead of round.

I am modelling the supports as pinned as it will be very difficult to design post-fixing with extremely high bending moments. Note that the supports will still have moments due to resolving the torsion of the stringers. I am also analyzing this for vibration as per SCI P354. With pinned supports, I am getting around 5 Hz of frequency and this is quite low to satisfy the vibration criteria. Some references say that for vibration analysis, you may used fixed supports even if they are detailed as pinned, in turn you will get higher frequencies and better chance of satisfying the criteria. I am not quite convinced that you can treat supports as fixed though. Has anyone done this?

 

[hokie66]

This looks like the stairs you see on a cruise ship. They are fabricated as big tube sections, welded plate like the rest of the ship. That's the way to deal with the torsion, but I think connections to a concrete floor will be cumbersome.

 

[EngrRC]

The initial concept of the architect was to use two flat metal stringers which to me was absurd considering the size of this staircase. Or is it?

 

[hokie66]

Yes, the architect's concept is absurd.

 

[EngrRC]

Going back to my previous message regarding torsion in the stringers. Is there a way to detail the supports (tube to end plate) so that no torsion from the stringer is transferred to it? As I said, it is tough to design post-fixing with moments. I want all torsion to be resisted by transverse beams, i.e. compatibility torsion.

Will fin plates welded to end plate and through bolts to HSS be a good solution?

 

[JStephen]

Some of the helical stairways on tanks use two flat metal stringers (8" or 10" flat bar, 1/4" to 3/8" thick). However, those stairways will have knee braces supporting them every few feet and are just designed as simply supported beams. I can't imagine that system working here.

I've seen ads from some of the metal fabricators concerning forming helical shapes, and it might be worthwhile talking to one of them to see how practical the finished design is, how smoothly those pieces can be formed, etc.

 

[hokie66]

EngrRC,
Compatibility torsion depends on there being a backstay to provide that compatibility. In your case, both stringers would be torsion members, so no compatibility available.

 

[EngrRC]

hokie66, interesting point. However, are we not using compatibility torsion for edge beams in general building design? These perimeter beams have transverse beams framing on only one side, much like the stringers in my stairs scenario.

 

[hokie66]

Yes, but the transverse beams frame between two edge beams, not between an edge and interior. Both edge beams are torsional members, and the old analogy of two drunks leaning on each other holds here.

 

[Agent666]

Because the movements/rotations are so small for vibrations, it is usual to treat 'nominally pinned' structural connections as being fixed for the purposes of a vibration analysis. I believe SCI's P354 guide and AISC Design Guide 11 recommend this approach.

 

[EngrRC]

Agent666, thanks for the response regarding my question on vibration.
You are right, SCI and AISC both recommend doing so. I agree that the 700N walker weight is very small, but isn't the structure weight+SDL enough to force the supports to behave as pinned, even before the walker comes in? I am also confused why they are making simplified vibration formula considering pinned support when the fixed support-formula should be just a factor away, and that they recommend using fixed.

Hokie66, I get your point now. I will be using torsional resistance for the bottom stinger connection and no torsional resistance to the top stringer connection (connecting to side face of PT beam).

 

[hokie66]

I would be reluctant to take that approach, but let us know how it works in practice. There is a lot many of us don't understand about spiral stairs.

https://www.google.com/search?q=spi...ome..69i57.14399j0j8&sourceid=chrome&ie=UTF-8

 

[Agent666]

The friction in bolted connections is enough for them to act as fixed under the small strains that are mobilised during vibrations.

 

[EngrRC]

For seismic load on the staircase (as per ASCE 7 Seismic load on non-structural components), should I use the Omega factor of 2.0 in designing the anchorage to concrete?