Risa 3D Stair Modeling 2

[McMichael05]

Anyone have experience modeling stairs in Risa3D? Was having convergence issues with this stair because of the unique hanger situation (Typ C12x20.7 stringers, C7x9.8 Flight headers, 1" dia. hanger rods)at the 4th landing, back left corner. Because of the supports it was twisting and pushing the 4th landing out causing a divergence. Anyway, i tried modeling the concrete landings as concrete plates within Risa but im not sure if this is an accurate model now. It solves without any issues with the original shapes and the deflected shape looks correct, but i dont have much experience with modeling plates. Just thought id see if anyone has any advice on this topic. Thanks!
[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1626095249/tips/Stair_2_Graphic_qwdc0n.pdf[/url]

 

[JLNJ]

Maybe RISA is telling you that your stair has lateral stability issues.

 

[Aesur]

Please share an image of the model using lines with local pin/fixed locations. I am leaning toward stability as JLNJ says above.

As for the plates, they do look odd. Here is a link to meshing in RISA which will help you with recommended plate sizes, etc: Link

Please note that when modeling many items in RISA it's rare to need to model plates for landings; therefore I expect stability is the issue.

Another nice trick is turn off P-Delta and see if the model runs (sometimes still doesn't run), then look at the deflected shape to see what is going on, this can many times give you a good starting point for where your issue is.

 

[McMichael05]

I agree with both of you, there is an instability issue because of the hanger system with this stair. My main concern is why when i add in the concrete plates in the model, it stiffens it up enough where i dont have an instability issue and all the original shapes work. So im trying to determine if the model is accurate with the plates modeled. See the attached graphic of the pin/fixed locations.
[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1626182723/tips/Stair_2_Graphic_with_pin-fixed_u6ntpw.pdf[/url]

 

[JoshPlumSE]

If I look at the 2nd level up, I don't see any resistance in the lateral direction. The hangers are pinned the stringers are pinned. I an probably say the same thing about the landing directly above it. So, both the levels that are restrained only by hangers are unstable laterally.

Note: Try applying self weight in the two lateral directions and see what happens to your deflections. I think those two levels with fly off into space.

 

[McMichael05]

Yeah so the only way i can get the model to work is to increase the size of the stringers to C15's to increase stiffness. The issue then is a headroom issue with the now bigger stringer. Unfortunately the stair is kind of out in the open so connecting the landings to get some resistance in the X and Z direction isnt really an option.

 

[Aesur]

I believe the reason you are getting the model to work with the concrete is because it's acting as a diaphragm with fixed connections at the perimeter members. Try removing the plates and fixing the ends of the stair header beam and see if that gives the same results as with the plates.

As others have said, laterally the model is unstable, can you add x-bracing up from the 2nd to 4th landing and then 4th to roof on the side opposite of the stairs?

Let us know what happens when you try that.

 

[JoshPlumSE]

The issue is lateral stiffness at the hanging landings. This is probably an issue whether or not the model converges. This is probably an issue no matter what size your stringers are.

At least that's how I interepret what's happening.

 

[McMichael05]

So i ended up not modeling the concrete deck using plates because when i released the moments at the corners i ended up getting the same divergence issue as before. While the concrete deck will brace the landings somewhat, i dont think i could jsutify them being moment connected to the stringers.
But Aesur, great suggestion! I have updated the model using tension only X bracing at the 2 hanging landings and we are good to go. Just an FYI, when you do X bracing in Risa, they recommend you set the density of the material the X bracing uses to 0 to prevent the sagging issue under self weight.
Attached is a graphic of the final model. Thank you guys for your help!
[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1626199042/tips/Stair_1_Final_Graphic_erhtvt.pdf[/url]

 

[JLNJ]

??

An x-brace tension-only structure with no compression elements? Doesn’t look right.

 

[JLSE]

I dont like the tension only either. I think you were on the right path with the plate modeling. Represent the landings and the stair treads in the model.... stairs can act like deep beams, adding stability.

 

[McMichael05]

Why do you guys think that? The X bracing ties the landings together and helps stiffen the stair in the X direction. This is shown in the model. Obviously not saying that just because the model says its right that it is right, just curious why you dont like it.

 

[JLNJ]

It just doesn’t look right. You can’t have all tension elements in a bracing system.

That said, your verticals may be “pretensioned” enough from gravity loads to keep them from going into compression if your lateral loads are small enough.

Maybe there’s enough diaphragm action in the stringers and landings to keep it from racking.

 

[JLSE]

I dont like the lack of redundancy and I dont like relying purely on tension bracing... because dont they see compression in reality? and when they buckle, how will they buckle? will it affect the braces tensile capacity? Will the very long braces elongate? Will elongation affect the function of the stairs? How much will it deflect, and how will that affect the function of the stairs.

.... people need to get out of the building after an event. Stairs need to be able to resist that event's loading, and still be safe enough for a flood of people trying to evacuate in a hurry.... scrambling around fire-fighers, perhaps.

The braces might make good "suspenders", but I'd want the "belt" to be diaphragm action... Id recommend you look at using the deep beam stairs and landing diaphragms together, and if it still moves too much... but its close, and its stable, then add the x bracing... but consider the compression bracing because of the redundancy and the stiffness it provides... or at least consider the failure mechanisms of a tension only brace and design it with additional safety factors.

 

[JoshPlumSE]

I've seen tension only lateral bracing relatively frequently small structures like decks and such. I've also seen it with metal buildings like smaller warehouses and such.

Granted, I don't like it either, especially here in seismic country. I don't have the exact code reference (because I've never been directly involved in a new design with them), but I remember another engineer telling me that they're not allowed for some higher seismic design categories.

Doing them for the main means of egress of a building might raise some red flags. But, if you're not in seismic country, I think it would be allowed by code. Maybe not a great idea. But, probably allowed by code.

I wonder about whether it's possible to have a horizontal strut that connects back to the main structure in that direction.

 

[Aesur]

@JLNJ - I was thinking the same thing, however under seismic loading it may still see slight compression in the rods; maybe the stringers are stiff enough to resist this. I agree it is odd overall, however how many things do engineers hang from ceilings daily, including mechanical, art, etc.. without compression elements. As @JLSE said, the concern then becomes life safety during an event in which people need to use the stairs.