Post Installed Tilt Panel to Slab Connection for a Seismic Retrofit

[palves]

I'm working on a seismic retrofit of an old building, and have no idea if the tilt panels are connected to the slab. The owner doesn't want to spend money scanning the concrete for this info so now I want to add this connection in to bring it up to code. My method would be to drill a hole from the the exterior, through the tilt panel, and then about 12" into the existing slab. Then I want to epoxy-grout lengths of #4 rebar into this hole to tie the panel and slab together. However, I am getting killed by Appendix D and provision D.3.3.4 since the panels are only 5.5" thick, and I'm only considering the slab to be 6" thick. This means that my brittle concrete breakout or pullout failure modes will govern every time over the ductile steel strength and I feel a little stuck because of it. Does anybody know another avenue that I can pursue or some provision that will help me out? Maybe an idea for an easier and better way to connect the panel and slab? This problem seems to a recurring theme is high seismic areas so send any and all creative solutions you have.

 

[msquared48]

What forces are you trying to resist?

Why not connect the panel to the strip footing at the outside face of the panel rather than to the slab?

Is there an interior finish/look/useage that prevents you from using an angle and afterset bolting to the panel and slab?

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[palves]

I'm trying to resist the shear forces from the in-plane loads and tension forces for the out-of-plane loads. The reason I'm analyzing the tension is because my analysis considers the panel to be braced at the slab, and I don't count on the exterior soil to provide adequate resistance for seismic loads in the outward direction.

I had thought about connecting the panel to the strip footing, but the owner is keeping costs low and want to excavate as little as possible. Even excavating down far enough to be level with the slab is more than he wants to do.

Adding an angle on the interior was another option we briefly considered, but I would have had to put the bolt into the slab at 8" back from the edge. This interfered with some of their access since they would need to be at 1'-0" o.c. around the entire perimeter.

Sorry to shoot down your ideas, but that's one of the reasons I'm feeling a bit stuck at the moment.

 

[hawkaz]

Do you know if there is slab reinforcing? If not (or if you don't know)- then dowels are useless.

If adding dowels to a reinforced slab, I would take the bars in far enough to develop them

msquared's suggestion may prove to be a cheaper method of anchorage than this.

 

[spats]

Tilt-up panels are almost always connected to the slab-on-grade, as a minimum. "The owner doesn't want to spend money scanning the concrete for this info"... I guess he'd rather pay to retrofit the entire building with connections. This is just plain foolishness.

I'm not an experienced seismic guy (wind is my thing)... I guess you can't use frictional sliding resistance?

 

[spats]

P.S. Not to insult you Gator fans, But I think the professor at UF who came up with the Appendix D provisions ought to be shot. Never has one subject turned the engineering community so upside down. People who live in the academic world should stick to academics. This Appendix D stuff being extrapolated to every real world anchorage problem is a lot of bull!

 

[palves]

Spats, unfortunately I can't count on frictional sliding resistance to create this connection for me. I'll probably use a little bit to reduce the shear load seen by the dowels which will in turn make Appendix D easier to satisfy.

And I agree that Appendix D is a big pain. I understand that they want the ductile failure to govern, but in my case there just isn't enough concrete there to make that feasible. The steel is ALWAYS going to be stronger than the concrete. My maximum failure mode is pullout and that is only at 46% of capacity so why can't we call that good? We're already adding extra safety factors (pullout is at 23% of capacity in a non-seismic region) to account for seismic regions so why isn't that good enough?