RAM Frame Joint Code Check

I would like to add that AISC would generally say that rigid end zones should not be used in the analysis of steel frames.

www.aisc.org/.../f2da0d72-c80d-4c04-9726-c905d349615e.pdf

Not positive on Eric's original reference, but check this post too: there is a link at the bottom for NEHRP Seismic Design Technical Brief No. 2 which has similar recommendations. 

So if I'm understanding your link correctly, Bentley's recommendation and RAM Frame's default behavior is to ignore rigid end zones, correct? This would then have the program using using the higher column panel zone shear strengths in the RAM Frame joint check but at the cost of having more drift, correct? When I check my drift, with REZ ignored, it did come out higher (with the 413k PZS capacity in the attached screenshot). Then when I checked it with REZ and no reduction, my drift on my little 1 bay test frame was about 20% less (and with the 321k PZS capacity in the screenshot). So that seems to match up.

As far as which option was selected in RAM Frame, this information is then not brought into the RAM Connection module, and the user needs to check or uncheck the "Frame stability considered in analysis" option to match their REZ setting in RAM Frame (or possibly differently to explore effects of changing that, I suppose?). Having it unchecked by default in RAM Connection is the more conservative route, but I just want to make sure I'm understanding the interplay correctly. If I've already taken the hit on drift, then I might as well take advantage of the extra strength coming from that when I check it in RAM Connection. Whereas, if I'm doing delegated connection design for someone else, using RAM Connection as a standalone program, I would keep that unchecked in Connection unless I confirmed with them that they had not used REZ in their model. Am I understanding this functionality correctly?

One thing I'm still a little unclear about is the following NEHRP verbiage in section 5.4.3 of the 2016 2nd edition: "If the effect of panel zone deformations on frame stability is considered (see Section 4.2.1), AISC 360 §J10.6 permits an increase in the design shear strength beyond the level associated with global shear yielding of the panel zone. In this instance, Equations J10-11 and J10-12 may be used rather than Equations J10-9 and J10-10. If panel zones are explicitly modeled in the analysis, it is permissible to use these equations. However, as explained in Section 4.2.1, this is not common."

NEHRP section 4.2.1 that was referenced above says: "Most designers model moment frames with fully restrained connections using line representations of beams and columns, with the lines intersecting at dimensionless nodes. This accounts in an approximate manner for flexibility inherent in the panel zone. It is also possible to explicitly model panel zones using any of several 4-node scissor or 8-node quadrilateral elements. When this is done, AISC 360 §J10.6 permits consideration of increased strength for the panel zones. However, the benefits of doing this when fully restrained connections are employed are typically small; thus, this is not typically done. Some analysis software permits specification of a rigid end offset to simulate stiffness of the panel zone and to simplify calculation of forces at the column face. This is not an appropriate method to represent panel zones and often results in unconservative estimates of frame stiffness and underestimation of design drift." [underlining mine]

That makes it sound like RAM is not justified in using the higher capacity shear equations because RAM Frame is using centerline modeling. Or are you doing the explicit modeling internally?

Yes, we are now recommending that user ignore rigid end zones, using centerline analysis for the greatest flexibility in the model. We do not explicitly model the joint using 8-node quadrilateral elements.  

My reading of NEHPR 4.2.1 is that if an explicit joint is modeled or centerline analysis is done, either way you can use AISC 360-05 Eq J10-11 (and J10-12).

Elastic panel zone deformation contributions to story drift can be accounted for by either explicit modeling of panel zone shear behavior [we don’t do this] or by adjusting the lengths of beams and columns in a manner that accounts implicitly for the contributions of panel zone deformations to drift [this is accomplished using centerline analysis].

Ram Connection is not getting the information from Ram Frame about whether a rigid end zone was considered or not. The check box for “Frame stability considered in analysis” is unchecked initially, and it needs to be checked for models with no REZ used in the analysis so that the Connection program will also use J10-11.

I’ll log an enhancement for that workflow to be automated (it involves a change in the data exposed through Ram DataAccess and an update for Ram Connection).

Yes, the 2009 1st edition you're quoting from does seem to allow that, between 4.2.1 and 5.4.3. However, the 2nd edition that came out last May clarifies these 2 sections, and in the process, seems to eliminate that option when using centerline modeling. For instance, 4.2.1 now says that most designers typically use centerline modeling to account for panel zone flexibility approximately, while it is also possible to explicitly model the PZ with scissor or quadrilateral elements. It then states that when this is done, J10.6 allows the increased PZ strength. When what is done? Either centerline modeling or explicit modeling? I don't think so. It goes on to say that this is "not typically done", while it started out saying that centerline modeling was typically done. So I'd interpret that as meaning the explicit modeling is what allows the use of the higher PZ strength.

Also consider the updated language in 5.4.3. There it reiterates that J10.6 allows increased PZ shear strength if if PZ deformation effects on frame stability are considered, which can be approximate or explicit. And it specifies that this means using equations J10-11 & 12 in lieu of J10-9 & 10. But the next 2 sentences are, "If panel zones are explicitly modeled in the analysis, it is permissible to use these equations. However, as explained in Section 4.2.1, this is not common."

The authors seem to be tying Equations J10-11 & 12 specifically to explicit scissor or quadrilateral element panel zone modeling. And if that's the case, I don't think y'all (or most of us) would ever be able to use the higher strength equations.