Appendix D anchorage loads - clarified

[ampersand]

Every engineer I know has struggled with determining the design loads for high-seismic anchors, specifically, applying multiple load-increase and strength-reduction factors until the anchorage is ridiculously overdesigned. Plan checkers often don't understand the larger design process, and will require that all of these increase/reductions be applied simultaneously. One very important point to clarify is as follows:

Although the element you are anchoring to concrete (i.e., OMF column) may have been designed for overstrength-level loads (aka omega-naught, or "amplified seismic loads), you are not required to design the anchors for overstrength-level loads. ACI 318 App. D Sec. D.3.2 instructs you to use the load combinations from ACI 318 section 9.2 or C.9.2, which are not the overstrength-level load combinations.

If you choose to design your anchors per App. D Sec. D.3.3.6, you will apply a strength reduction factor of 0.4 to your anchors. The intent of this factor is to ensure ductility. Please note that the intent of the overstrength-level loads, which you may have used to design the element you are anchoring, is also to ensure ductility. To apply these overstrength-level loads to the anchors, and simultaneously use the 0.4 strength reduction factor, is extremely conservative, and is often very difficult to design.

Attached are five pages from ACI 318, with text highlighted, to clearly show what I have tried to explain above. Feel free to keep this document handy in case you need to demonstrate this to a plan checker. I hope this is helpful to some engineers out there.

Cheers!

 

[wannabeSE]

ASCE 7-05 eq 12.4-7 defines the horizontal load with over strength as E_mh = [Ω]₀ Q_E. The ACI combinations do not use Q for seismic loads, they use E. So, some may say that overstrength cannot be neglected.

However, the steel seismic provision, AISC 341-05, addresses this. Section 8.5 of AISC 341-05 states "The special requirements in ACI 318, Appendix D, for 'regions of moderate or high seismic risk, or structures assigned to intermediate or high seismic performance categories need not apply." It may be easier to justify including the overstrength factor on the demand side and neglecting ACI's 0.75 and 0.4 strength reduction factors on the capacity side when anchoring baseplate on steel lateral frames in high seismic areas.

 

[JAE]

I think if you read through ACI 318-11, section D.3.3.4.3, they have extensively revised this section and do include an OPTION to use the [Ω]_o factor as one possible way to deal with seismic.
There are other non-[Ω]_o options so ampersand's suggestion that using [Ω]_o is not always necessary is valid.

 

[ampersand]

Actually, I don't have a copy of ACI 318-11, but I'm glad to hear they may be trying to clear up the issue. But we seem to all agree that the intent of the 0.4 strength reduction factor always was the same as the intent of a seismic system Ωo factor- to maintain ductility.

 

[Archie264]

As usual on issues such as this, it's as clear as mud.

The various code writers have created a situation that poses a hazard in that very few people can understand the code anymore, much less the intent behind it. And whether those who do understand it are the day-to-day practicioners is another question. And since most practioners can't afford to take the time away from running their business that would be required to sit on the code writing committees said committees seem to be slanted in favor of theoreticians and practioners from very large companies who can afford to send people to sit on them.

...so...

To address this the software companies write programs to translate all the Greek letters into "usable" format, which people then use without much idea of what they've just done. I often hear "gray hairs" quite properly express concern about over-reliance on computers but I don't hear much push-back in the way of pleas for simplification of the codes. If it were just a source of frustration that would be one thing. But, unfortunately, if it hasn't already, it will at some point become a safety issue, as engineers struggle to decipher the code.

And unfortunately, as each new issue of the code books increase in thickness by about 30%, there doesn't seem to be any relief in sight...