ASCE 7-16 Type 4 Horizontal Irregularity

[kjm93]

Hello everyone,

I'm designing the lateral system for a building in a high seismic area. 8 levels above grade and 2 levels of parking below. The main SFRS includes two concrete shear cores extending from the foundation to the roof. At the main floor at grade level, a lot of the seismic shear force wants to transfer through the P/T slab out to the perimeter basement walls. I am designing the connections between the walls and slab for overstrength forces into the diaphragm through direct shear transfer and collectors.

My question: is this considered a Type 4 irregularity given that there is a lot of transfer? Again, the core walls are continuous through the garage. If it is considered a Type 4 irregularity the transfer slab itself would need to be designed for the overstrength transfer loads as well, instead of just the collectors and connections from slab to wall.

For reference, here is the ASCE 7 definition of the irregularity:

"Out-of-Plane Offset Irregularity: Out-of-plane offset irregularity is defined to exist where there is a discontinuity in a lateral force-resistance path, such as an out-of-plane offset of at least one of the vertical elements."

Thanks!

 

[ProgrammingPE]

I wouldn't think that just because some of the load gets transferred to the other elements that this would trigger an Out-of-Plane Offsets Irregularity since your core walls are continuous to the foundation. The examples I see for this are for discontinuous elements like in the picture above.

You may have a vertical structural irregularity, though, for Type 3 Vertical Geometric Irregularity which occurs when a horizontal dimension of the SFRS is more than 130% of the adjacent story.

Structural Central

http://www.structuralcentral.com

http://www.youtube.com/channel/UCOj2mXXf3ZbZtgxTc6BtkGg

 

[kjm93]

ProgrammingPE,

Thank you. The part that makes me wonder is "discontinuity in lateral force-resistance path", because the transfer force path is discontinuous there, even though there is a continuous path through the core walls to the foundation. I've seen examples approaching this either way when considering these grade level podium slabs above garages.

The Type 3 Vertical Irregularity is OK as it only affects the analysis procedure and I'm using MRSA as it is.

 

[Deker]

I consider this condition to absolutely be an irregularity due to the large transfer force required at the grade level diaphragm. The practical impact this will have on the design is the requirement to amplify the diaphragm transfer forces by the overstrength factor per ASCE 7-16 Section 12.10.1.1. This amplified seismic force should be included in the design of the shear transfer dowels and collectors, which it sounds like you already have covered. Reference Section 12.3.3.3 is required to be satisfied for Type 4 irregularities but has no impact on your design since the shear walls are continuous to the foundation.

 

[kjm93]

Deker - would you, then, design your entire diaphragm for the overstrength transfer forces (plus it’s own inertial force)? I have basically checked that the overstrength collector force can develop into the diaphragm, but once it’s developed into the diaphragm I no longer have the omega in the transfer loads as the diaphragm takes it out to the perimeter walls. i.e. the beam-theory global shear and bending, openings, and shear connection to basement wall are all checked for non-omega loads.

 

[Deker]

Yes. Since diaphragms are generally expected to behave elastically, design the diaphragm for Ω[sub]0[/sub]V[sub]transfer[/sub]+V[sub]inertial[/sub] so as not to create a weak point in the load path. It's not uncommon for the transfer shear to be a critical factor in determining the thickness of podium level slabs. That said, you can greatly reduce the transfer shear demand by modelling the cracked stiffness of the podium slab. See below for recommendations from two sources (LATBSDC, PEER/ATC 72-1). Highly recommend reading Appendix A of the PEER/ATC document for additional guidance on dealing with the backstay effect.


LATBSDC

PEER/ATC 72-1

 

[kjm93]

Thanks for the references. I've seen the LATB document, but the PEER/ATC Appendix A is very informative. Seems I've modeled the stiffness conservatively so if I run into troubles designing the whole diaphragm for omega transfer forces I can look to refine a little.

 

[RobertHale]

I would think @ProgrammingPE is correct based on the ASCE 7 commentary C12.3.2.1, "Where there are discontinuities in the path of lateral force resistance, the structure cannot be considered regular. The most critical discontinuity defined is the out-of-plane offset of vertical elements of the seismic force-resisting system (Type 4). Such offsets impose vertical and lateral load effects on horizontal elements that are difficult to provide for adequately." Since your system is only has a high lateral load effect on the diaphragm, it wouldn't seem that the irregularity would apply.
Out of curiosity, are you modeling with a rigid or semi-rigid diaphragm?

Robert Hale, PE, SE

 

[kjm93]

Robert, I agree that it is not a Type 4 irregularity, and by code is not required to be designed globally for omega forces per 12.10.1.1. However, I am going to check the diaphragm for omega level forces and see how far off I am as a good engineering check. I am modelling the diaphragm as semi-rigid.

 

[Deker]

You all seem to be focusing on the walls being continuous to the foundation rather than the major discontinuity in the load path that occurs at the podium level. For those who don't believe this is an irregularity, is it your position that Horizontal Type 4 irregularities always occur simultaneously with Vertical Type 4 irregularities? And at what percentage of the global shear required to be transferred through the diaphragm would you consider this to be a "discontinuity in the lateral force-resistance path"? For what it's worth, I've attached a couple figures below that show walls being continuous to the foundation but still indicate that irregularities / discontinuities exist.


Jack Moehle, Seismic Design of Reinforced Concrete Buildings

John Hooper, Presentation on ASCE 7-16 Seismic Provisions

 

[kjm93]

My understanding is that the difference between Horizontal Type 4 and Vertical Type 4 is whether the offset in the wall is perpendicular to the wall (Horizontal Type 4) or parallel to the wall (Vertical Type 4). Both would cause the diaphragm / transfer element to carry significant overturning forces in addition to the shear forces. The diagrams in the ASCE 7-16 Commentary for both of those irregularities show the walls being discontinuous at the level of the irregularity.

Like I said, I'm going to check my diaphragm for overstrength to see how close I am, although I think I will be over capacity by about 25%-50% in shear without specifying additional reinforcement.

 

[kjm93]

A recent StructureMag article about when overstrength is used in diaphragm design: Link

 

[Deker]

Unfortunately your article doesn't address amplification of transfer forces per 12.10.1.1 which is really the heart of the issue we're discussing.

Yes, a VT4 irregularity occurs when there is an in-plane offset in the wall. And yes, an out-of-plane offset in the wall is an example of an HT4 irregularity, but it is not the only condition where it occurs.

From Table 12.3-1, Out-of-Plane Offset Irregularity: Out-of-plane offset irregularity is defined to exist where there is a discontinuity in a lateral force-resistance path, such as an out-of-plane offset of at least one of the vertical elements.

Other conditions include setback slabs, podium slabs, slabs at the top of partial height vertical SFRS elements, etc. If it were only meant to apply to discontinuous offset walls, why say "such as" in the definition? And why would the figures I posted above show full-height walls?

An HT4 irregularity is the analogue to the VT4 irregularity. Any element involved in the transfer of forces out of the wall and into another part of the structure is required to resist amplified seismic loads. In a VT4 irregularity, the transfer occurs through beams and columns. In an HT4 irregularity, the transfer occurs through the diaphragms (and, in the case where VT4 also exists, beams and columns). If not for this requirement, you could potentially have a 30 story tower transferring 50% of its entire base shear out through a podium slab that's not been designed for any specific level of ductility. Not great.

I'm glad that you are checking your diaphragm for overstrength, but my goal is to convince you and others here that this isn't just one of those good practice checks. It's a requirement.

 

[kjm93]

If it is a requirement and the intent is for transfer podiums to be considered Horizontal Type 4, it seems that it should just be explicitly addressed in the code since it is a very common occurrence. And if it was a deliberate change in 7-16, changing the Commentary of Horizontal Type 4 seems necessary as one could easily interpret it as being defined as having BOTH vertical and lateral load effects on the horizontal member (i.e. diaphragm), right?

 

[Deker]

An offset wall is an extreme example used to illustrate the intent of the provision. If it was meant to to apply only to offset walls, the provision would be written that way, as it is with VT4 which applies only to in-plane offsets. Unfortunately codes and commentaries can't consider every possible scenario. The intent is clearly defined, and once you understand that the intent is to flag discontinuities in the lateral load path, it easily follows that podium slabs trigger this requirement.


I would have thought that the intent was clear, but seeing the responses in this thread has shown me that it's not being interpreted consistently among practicing engineers.

From commentary section C12.3.2.1, "Where there are discontinuities in the path of lateral force resistance, the structure cannot be considered regular. The most critical discontinuity defined is the out-of-plane offset of vertical elements of the seismic force-resisting system (Type 4). Such offsets impose vertical and lateral load effects on horizontal elements that are difficult to provide for adequately."

Offset walls are mentioned because they are the most critical type of HT4 irregularity, not because they are the only type. As explained above, HT4 and VT4 are two separate requirements with two separate criteria, and having one does not necessarily trigger the other.

Nobody has addressed the points I've made in my posts above, but for those who have been interpreting this differently I'd like to know your justification.

 

[kjm93]

I'm not sure which points you are referring to not being addressed.

Question - in the Moehle diagram above, the setback irregularity mid-height. If those two walls align, it would seem odd to me to consider that an "Out-of-Plane Offset Irregularity". Maybe it could be named something less specific like "Lateral Force Transfer Irregularity" if its not intended to only account for out-of-plane offsets.

To your point, it seems that 12.3.3.3 would have always required the overstrength for diaphragms/elements supporting a discontinuous wall. So the logic follows that the updated 12.10.1.1 would be somewhat redundant if it weren't meant to cover other HT4 irregularities, possibly including a podium slab. So that makes sense to me.

To play devil's advocate, I still think the commentary could still easily lead you to think the definition of HT4 is out-of-plane offset SFRS elements.
[ul]
[li]First sentence: is there a difference in load path vs. path of resistance? We know walls continuing to the foundation do not necessarily have a discontinuity of path of resistance because some load is still going down those walls. However, we know from analysis that there is a discontinuity in the path of some load as it transfers out.[/li]
[/ul]
[ul]
[li]Second sentence: Where they put the (Type 4) at the end could be seen as defining the irregularity. It also doesn't help that it is the name of the irregularity as well, as a mentioned above.[/li]
[/ul]

I'm just here trying to understand the code better so thanks for the discussion.

 

[ProgrammingPE]

I agree now that this should be considered an HL4 irregularity. The wording differences between HL4 and VL4 are pretty subtle, but the differences are intentional.
[ul]
[li]VL4 includes "Discontinuity" in its title and HL4 does not.[/li]
[li]VL4 says "seismic force-resisting element" and HL4 says "seismic force-resisting path"[/li]
[li]VL4 says "an in-plane offset" and HL4 says "such as an out-of-plane offset"[/li]
[/ul]
The code gives the example for HL4 of out-of-plane offsets of the vertical elements in multiple locations which is an uncommon and extreme case (100% of its shear is transferred through the diaphragm). The introduction of a lateral element at a lower level is common and much less extreme (it looks like this irregularity would be triggered if any of the lateral element's shear is transferred through the diaphragm even if 99% still goes down to the foundation).

It would probably be beneficial for them to mention this case as an example as well since the discontinuous element case is presented so strongly and apparently being discontinuous is only a requirement for VL4 and not HL4. I also agree that the name probably shouldn't be Out-of-Plane Offset if you want to include cases other than that.

Thanks for pointing this out!

Structural Central

http://www.structuralcentral.com

http://www.youtube.com/channel/UCOj2mXXf3ZbZtgxTc6BtkGg

 

[Deker]

Benny, I agree with you and ProgrammingPE that calling it an "Out-of-Plane Offset Irregularity" and then using "such as an out-of-plane offset" in the definition is inconsistent. I've attached some more snippets for consideration, this time from ASCE 7-10 and the 2015 NEHRP Recommended Provisions. The two sections are in agreement, with the exception that the NEHRP provisions require amplification of transfer forces. Note the definition of transfer diaphragms in the NEHRP provisions includes conditions due to change in lateral stiffness of the vertical elements.

Also note that while the the wording is crystal clear in the NEHRP provisions, the revision used in the ASCE 7-16 adoption to require amplification only in structures with an HT4 irregularity is confusing (mainly due to the sloppy wording described above). Did the ASCE 7 committee think a reference to HT4 was simpler than adding a definition for "transfer diaphragm" to the standard? Or did they want to exempt regular structure from requiring amplification? Or did they intentionally mean to require amplification only where there is an actual offset in the vertical elements? I can speculate, but who knows...

The provision seems to be ambiguous enough to justify either interpretation, but knowing the intent of the NEHRP provisions I'm inclined to lean towards amplifying the transfer forces. Full disclosure: I amplify all transfer forces for diaphragm design as a matter of course. If they're small, great. No impact on the design. If they're large, it's likely due to a condition that would trigger an HT4 irregularity. This abides by the basic principles of seismic design and keeps things simple for us.

This has been an enlightening discussion. Thanks for engaging with me.

ASCE 7-10 Section 12.10.1.1

2015 NEHRP Recommended Provisions

 

[kjm93]

I like the NEHRP version of that paragraph, it is clear it is intended for any transfer diaphragm. The 7-16 12.10.1.1 paragraph seems to imply by the first two sentences that there are instances where there are transfer forces but no HT4: "Always design diaphragms for transfer forces plus inertial forces, and if there is an HT4 Irregularity apply omega to the transfer forces". I'm trying to think of an example of that given our discussion. Maybe that's just leaving some room for engineering judgement if transfer forces are low.

Anyways, thanks Deker and others for the insight. I'm not sure other engineers in my office approach it this way so it would be good to bring up.

Cheers!

 

[Deker]

I'm trying to think of an example of that given our discussion. Maybe that's just leaving some room for engineering judgement if transfer forces are low.

Some examples of regular structures that require consideration of transfer forces but would not require amplification per ASCE 7:

• Dual systems.
• Full-height shear walls of different lengths.
• Full-height shear walls with different opening layouts.