Effective Wall Width - ACI vs PCI 6

[LearningAlways]

What effective width do you typically design a precast wall for? PCI recommends a width that is 12*thickness + bearing width. ACI recommends 4*thickness + bearing width which is three times as small as PCI's recommendation.

Maybe the difference comes in the first line of ACI's load distribution definition. From ACI318-14, Section 11.2.3.1: Unless otherwise demonstrated by an analysis, the horizontal length of wall considered as effective for resisting each concentrated load shall not exceed the less of the center-to-center distance between loads, and the bearing width plus four times the wall thickness.

Perhaps PCI has performed testing which increases their width to what they recommend but I haven't found much in PCI to justify their recommendation. Fellow engineers at my company stay with ACI's recommendation. We ran a finite analysis using mesh and the results are closer to ACI.

Does anyone know what study PCI did to justify this increase?

 

[r13]

I think the difference been in the method for embedment installation. In precast concrete, the embedment likely been pre-installed in a controlled environment. On the other hand, the embedment in CIP concrete are likely post installed, even pre-installed, the quality will be a concern. But you wouldn't make mistake by following the more restrictive rule though.

 

[JAE]

retired13. Why are you talking about embedments?
This question is about applied concentrated loads on a wall panel, in the plane of the wall, and how wide a portion of the wall can be counted on to resist the axial force.
Nothing to do with embedments.

LearningAlways - I checked with some of our precast boys here and they weren't sure why the difference. Neither ACI or PCI have any commentary on it.

 

[LearningAlways]

Thanks, JAE.

I noticed the missing commentary too, and it's a weird circumstance that neither has commentary, especially in PCI's regards given the large difference.

 

[r13]

Because more attention needs to be paid to the CIP embedment due to quality issue, thus the spread of the load ideally should be limited to a smaller area to ensure higher stress locally, thus require detailed design with potential for more reinforcement. This is less a concern for pre-cast panels/walls, since everything is complete in the plant. The above just a personal thinking.

 

[JAE]

retired13 - you make zero sense here. The question was about the global design of the wall panel. Read the first post.

 

[r13]

My understanding the question is why ACI allows a much smaller effective width for distributing bearing force than PCI. I don't see I am off the topic.

 

[Agent666]

I suspect with a design guidance produced by a particular facet of an industry, that there will be a propensity to favour that type of construction. So it could simply be a case of the precast boys pushing the limits of acceptability to make precast more attractive in a given application, whereas ACI take a potentially more conservative/theoretical view of this particular issue of load spread.

In New Zealand, our masonry and concrete standards are also based on the bearing area plus 4 times the wall thickness. Taking 12 times the wall thickness seems quite hopeful and would definitely require some further investigation on how the load might spread out locally to the point of applying the load. For example locally to the load, flexure in the wall from the load would obviously not occer over the full 12 times the thickness of the wall. It is perhaps for this reason ACI have simply taken a smaller value to more accurately capture the localised effects, rather than global behaviour at some point remote from the load application point.

I basically feel this ACI requirement is intended to ensure loading is dealt with more on a local scale than the PCI requirements suggest. No doubt the true behaviour lies somewhere between.

 

[JAE]

retired13:
No - not bearing force. This has nothing to do with bearing on the concrete, embedments, or bursting forces under a bearing surface.

ACI and PCI provide an "effective width" of WALL to use in determining the overall WALL's axial/bending capacity - in other words, how wide is the distribution of the load over the length of the wall where you can count on it like a beam-column in bending. If you read the provisions that the OP presented and understand them you'd see that your posts are way off topic.

 

[r13]

One has narrow distribution width, thus the flexural is concentrated in a smaller bend, vise visa.

I think you guys find a ease way to collecting the pink star, finally. I am sad this how this forum works.

 

[Brad805]

Here is an old link off the PCI page. Link A bit more searching may find a more recent discussion.

 

[r13]

Excellent post. Brad.

Here I was talking about. Not in line with PCI perspective, but..

 

[r13]

Further note: Both institutes are reside in the US and share the same talent pool. The studies, experiment, testing should not deviate too much as the code provisions indicated. There must be untold thoughts behind, just we don't know the exact reasons. The argument of PCI on the track record of its requirement/recommendation is very powerful and convincing, at least up to now.