Uplift on Column Anchor Bolt

[tw]

We've got a 10 story building with 3 steel x-braced frames one direction and 110 mph wind. The situation we have is uplift on the foundation. Analysis for wind using 60% DL & 160% WL put a factored tensile load of 900 kips on column base. Using ACI 318-02 (Appendix D) requires a massive pedestal (large edge distances required) and deep, deep embedment of anchor bolts. We can bury the column base and add concrete weight to it to add DL but we still have quite a number to deal with.

My question is how have others solved this problem? Are there anchor bolts that are deformed like rebar so behavior would be like development length rather than failure cone of headed anchor bolt?

Seems like rebar could be threaded like anchor bolt, is this done? That seems the cheapest solution but have not seen anything written about it.

tw

 

[egc]

TW,

I would assume that the failure mode for deformed rebar and design basis is similar to headed ABs. If your question deals with the attachment of the column to the foundation, I would look at column base (on a steel beam grid) encasement. This value (900 kips) sounds high to me for a net uplift but hey, thats why they need engineers <)).

I just hang here (occasionally)

 

[dik]

as egc says, the load seems awfully high... you may want to re-figger... or, something. You may want to consider relocating bracing to columns with a greater dead load; it sounds like you are bracing to a lightly loaded column.

Failing that, anchorage for that magnitude of uplift will likely require rock anchors or something of a similar ilk. You may want to look at helical anchors, check with someone local that can install them... but, I've never heard them used for that magnitude of load.

For large uplift AB's I've used concrete encased 'Dywidag' threadbars as rock anchors. These have a slightly higher price for normal grades of reinforcing and accept a proprietary threaded nut and coupler (great for AB's with large tension). They are also available in high strengths, suitable for post-tensioning.

 

[tw]

Actually my initial post should read 775 kips factored uplift from load case of 90%DL & 1.6WL.

Still a mother.

tw

 

[ishvaaag]

If I had such uplift load and weight to balance was not available I immediately would start to think on the lift capacity of pile or piles uner the column. Respect passing the load yes, a transversal insert from which to start on the pilecap or even &quot;under&quot; it may do. Upon such tensile load I wouldn't care much on whether the details are a bit expensive, but on sound constructability and reliability for the intent.

Other alternative depending upon placement would be if I could take the lift in flexure by some foundation wall under the column, a case when the reinforcement would become manageable, and of course checking that the general stability and allowable pressures on the soil and so on are met.

 

[tw]

I've got the weight and resistance in the foundation, I am only concerned with the transfer from steel base to concrete foundation.

tw

 

[egc]

TW,

I remember seeing a condition like this in either ASCE or the ER magazine a number of years ago. They (the authors) used a steel beam grillage to transfer the uplift to the concrete foundation but as I recall they used grouted rock anchors to tie the foundation down (they had severe space limitations).

On the other hand since I am over fifty, maybe my memory matches my vision (and they don't have corrective devices for the mind yet!).

I just hang here (occasionally)

 

[rowe]

From your original post, you asked about deformed anchor bolt/bars being able to achieve some capacity by development length vs. strength of concrete cone failure.

PCA has a publication by Dr. Cook addressing &quot;anchorage to concrete&quot; (I thought that ACI was suppose to incorporate Dr. Cook's research, but I don't have a copy of the ACI -02 so I don't know). Anyway, the concrete cone failure can be reinforced by stirrups/ties and if you can satisfy the &quot;cone reinforcement&quot; requirements, you could then base your available strength on the anchor bolt capacity.

If you're interested:

PCA Publication &quot;Strength Design of Anchorage to Concrete&quot; by Ronald A. Cook, PhD. (it's only about $15-$20 and it's well written with several examples).

good luck

 

[tw]

Looking at the ACI 318-02 Appendix D, I don't see where the grillage would help you numbers-wise when you are limited by concrete breakout strength. It's the c1 and c2 numbers that kill you.

Plus you get little benefit from reinforcement accross the failure plane.

tw

 

[egc]

I may be missing something here but I though your problem was in &quot;increasing&quot; the area of the &quot;failure plane&quot;. Activate enough area and your stresses should be okay. May be you need to draw me a picture <).

I just hang here (occasionally)

 

[structuresguy]

Another option might be to bury the column deep into the concrete, with only enough concrete under the column for punching shear. The column would have shear stud connectors welded to the flanges and webs. Use enough studs to develop the required tension. Then figuring on minimum stud spacings, come up with a depth of concrete required to achieve the weight needed to conteract the uplift. Combine this with horizontal and vertical reinforcing in the concrete, particularly near the column, and you will have &quot;tied&quot; the concrete to the column. Now the classical pull-out would not exist.

Also, I disagree with your statement that you get &quot;little benefit&quot; from reinforcing steel. The cone fails in shear exactly the way a beam fails in shear. Beam stirrups greatly increase the shear capacity of the beam. Well the same is true for a cone-type pullout failure. Without having run any numbers, I would figure on at least doubling the pull-out capacity with proper reinforcing. If what you said was true, elevated flat plat floors would never work. they fail by tension cone the same as anchor pull-out.

Also I question your load combination for concrete design. I gather you are using IBC load factor for wind under LRFD. However, my interpretation is that for concrete, IBC specifies that the combinations of 318 shall apply, with the exception of earthquake. So this would mean 0.9D + 1.3W, not 1.6W as you used. I would say you are over-designing the uplift requirement by about 25%.

IBC 1901.2 states that concrete shall be designed and constructed in accordance with IBC chapter 19 and ACI 318, as amended in Section 1908 of the IBC. Nowhere in Section 1908 does it require you to use the IBC load combinations. In fact, only for seismic does it tell you to not use the ACI load combs.

You still would have one heck of a load regardlees, I figure maybe 625kip. Just out of curiosity, what column section are you planning on using?

Don't want to step on any toes, just telling it the way I see it.

 

[structuresguy]

Just found this, in fact, IBC specifically tells you to use the load combi's from ACI Section 9.2 in IBC 1913.3.2, again for all but seismic.

One comment on my above statement, shear studs would have to develop both the tensile and compressive loads I would think, as I would say the studs would engage under compression loads before end bearing on the base plate would be fully effective. Maybe some ratio could be used between shear stud and end bearing, but I wouldn't know what it would be.

 

[msajjadh_imported]

I suggest you to get ASCE Report &quot;Design of Anchor Bolts for Petrochemical Facilities&quot;. It suggests the way dowel bars are considered in design. Because Dowel bars are in the failure plane of concrete. ACI 318-02 does not address this issue.

 

[tw]

Structuresguy,

ACI 318-02 does use 90% DL combined with 1.6 WL. If wind load has NOT been reduced for directionality then you would use the 1.3 factor instead of 1.6. My forces are correct according to code.

What I mean by little benefit from the reinforcing steel in the anchor bolt cone failure is that it is not included in the calculation for concrete breakout strength in ACI, other than a slight increase in a factor. Obviously, properly developed rebar across a failure plane will do its job and I'll have plenty of it.

I think a careful reading of the ACI 318-02 commentary RD.4.2.1 tells me what we all would intuitively know and probably answers my original question at least in some respects. I just wanted to hear how others handled the situation. Appendix D is new so I'm just getting used to it.

My first floor columns in these frames are W14x283's with W10x77 braces and a 27' stance.

tw

 

[structuresguy]

Well, I stand corrected on the load factors. I have not seen ACi 318-02. I am still using 318-99. I was not aware that load combination factors had changed. I guess I will have to get a copy of the 318-02 code.

 

[ick2]

tw SOS-I agree with the suggestion of high strength bolts or dywidag rods. Usually the high strength bolts are tied together with a plate embedded in concrete. What you really have to be carefull here is the SHEAR from the bracing. When your bolts are in tension there is little or no friction between the bearing plate and the foundation. In this case the bolts will not fail in shear but in BENDING!! (which they are not designed for). In order to avoid this you should provide a shear key.
In the past I was part of a group investigating a failure that happened because the engineer forgot to provide a shear transfer mechanism other than the bolts.