ACI 318 Cracked vs uncracked concrete for anchor analysis 1

[maramos]

I'm in the process of designing a wedge anchor connection per appendix D of ACI 318-08. I'm trying to figure out when to assume cracked vs uncracked concrete.

When anchoring to a compression zone, one would assumed uncracked concrete and cracked for tension zone.

What happens when you anchor to a compression zone in a seismic area? Do you assume cracked? or can you still assume uncracked?

 

[Lion06]

I always assume cracked concrete.

 

[sdz]

You also have to consider that cracking can be caused by shrinkage as well as applied tansion. ACI App D is written so that cracked concrete is the default and a factor is allowed for uncracked concrete. I also would assume cracked concrete unless there were good reasons to use the factor for uncracked concrete.

 

[Lion06]

Another potential for cracking is the high localized stresses as a result of the anchor embedment. Even if it's a column in compression, the localized stresses surrounding the anchors are not representative of the typical column section.

 

[weab]

I agree, assume cracked if you can. If your design can only work with uncracked, you must consider seismic load reversals, etc. to make sure your uncracked concrete does not become cracked during seismic loading.

I don't have the code with me at the moment so check me on this. But I don't recall that the localized stresses due to the anchor loadings need to be considered. I believe that it is the initial condition that is at issue, that is, is the concrete already cracked. A bending compression zone in a thick beam or a compression only column should be acceptable for this. I'm going to check this though.

 

[Lion06]

I'll double check in the morning when I can look at ACI 318, but I don't believe that it specifies what types of loads to consider or not consider. It just says that if analysis shows no cracking at service loads, then you can use the 1.4 factor for uncracked concrete.

I've had to use rebar welded to embed plates on a number of occasions recently to get away from App. D, because I just couldn't get the anchor calcs to work.

 

[Lion06]

I was stopped at a traffic light this morning on my way in to work and I happened to be under and overpass. There were concrete wall abutments on either side with precast box girders spanning between them. The overpass was roughly 15' high. The abutments were doing 15' of retaining, but I'd be willing to bet that analysis would show no service level tension to exist on the exterior face of the walls (yes, this is the tension face for retaining, but I believe the DL of the bridge would more than offset this) let alone service level cracking. There were, however, significant and numerous cracks (both vertical and horizontal).

Just something to think about.

 

[StructJan]

I assume always "cracked" concrete in SDC C and higher.
All concrete is cracked, but cracked is wrong definition. It should be replaced by a word tension zone.