DCR check on concrete shearwalls

[lg1990]

Hello,

I am comparing shearwall results with ETABS and S-CONCRETE, and I came across this difference between how the flexural/axial D/C ratio check is calculated.

In S-CONCRETE, a horizontal line is projected to the P-M curve. The DCR is calculated by the ratio of horizontal length (shown below)
[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1639774034/tips/SCONCRETE_DCR_ovecu1.pdf[/url]

In ETABS, a projected vector is used to calculate the DCR. This in some cases would be conservative (with high axial loads), and unconservative (with low axial loads) compared to S-CONCRETE's DCR check.

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1639774140/tips/ETABS_DCR_js3egv.pdf[/url]

Is there a code reference that directs that a horizontal line should be used for the DCR check? What have you typically done? Both programs will determine if the wall fails (outside of the P-M curve), but a inaccurate D/C ratio can lead to issues.

Thanks,

Leo

 

[sticksandtriangles]

I have always used the projected vector to calculate DCRs for concrete PM diagrams. I feel this to be a more accurate representation of DCRs as compared to the horizontal line method; at the same ratio of axial load to moment, what is the DCR.

S&T -

http://www.re-tug.com

 

[KootK]

1) I don't know of a code reference for this. If one exists, my money's on Deker to produce it.

2) Traditionally, it has been the vector method. However, the utility of that method is somewhat predicated upon the bending and axial demand scaling up during loading in equal proportion-ish. This makes more sense for some types of loading than others.

3) Give that you're in California and enquiring about DCR, I'm guessing that your main concern is seismic. For seismic loading, you generally have a pretty static axial load and a highly variable flexural demand. For such scenarios, I can see an inherent logic in the straight line method because it matches the loading pattern.

4) Compared to columns, I find that most shear walls tend to be lightly loaded axially which may provide yet another justification for the straight line method.

5) I wasn't aware of this difference between ETABS and S-Concrete. Thanks for bringing it to our attention.

 

[Deker]

I suppose they are both correct, and you could also calculate a DCR for the vertical (axial) dimension. ACI 318-19 10.4.2 and 10.5.1 require that all three dimensions be satisfied. When you say that an inaccurate D/C ratio can lead to issues, what type of issues do you anticipate?

 

[chris3eb]

The thing that both methods always agree upon is whether the point is inside or outside the PM curve or PMM surface. It's kind of a moot point to know how much it doesn't work by, and there are too many variables for there to be a clean answer to that. Normally when you concern yourself with DCR values above 1, one of two things are happening: you are using ASCE 41 and planning on looking at DCR/m as your acceptance criteria, or you are trying to figure out how much you need to reduce loads to make something work.

If you are using ASCE 41 to check a column, be aware that you can't take a DCR from ETABS and simply divide it by an m-factor - refer to section 10.3.3 of ASCE 41-17. You are actually reducing your demands by m or j factors prior to doing the PMM check rather than calculating a DCR and then reducing that number.

If it's reason number 2, you want to consider which loads are locked-in and which things you have control to change. As KootK mentioned, the horizontal line method makes sense in cases where you have a low-rise structure and you are worried about seismic loads (in this case, the axial load would come primarily from gravity and thus would be "locked-in" relative to the seismic loading.) Some taller structures have a significant axial contribution from seismic overturning, so the vector method would make more sense because all loads change as a function of the seismic loading. Either way, it may be more useful to discount the locked-in loads which is to say that there is a big difference between 25% of your flexural capacity being used up by gravity and trying to play with seismic to make things work, and having 75% of your flexural capacity taken up by gravity where you can't control that much by playing around with seismic