Need a book?
Engineering books recommendations...
Return to index:
Pile Cap Design--Strut and Tie method
- To: seaint(--nospam--at)seaint.org
- Subject: Pile Cap Design--Strut and Tie method
- From: Michael Hemstad <mhemstad(--nospam--at)mbjeng.com>
- Date: Mon, 19 Sep 2005 15:51:41 -0500
I am attempting to design some large caps carrying loads from single
concrete columns into pairs of drilled piers constructed with about 2 feet clear
between them. In one example, the column is 40 inches square and the caps
are 84 inches wide and about 16 feet long. I am caught between what seem
to be conflicting Code clauses in ACI 318-02.
Our columns are designed with fc' = 12,000 psi concrete. Where they
land on the pile caps, we invoked ACI 10.17, which allows bearing stresses
up to double the allowable of (phi x 0.85 fc') considering the confinement
offered by a bearing area larger than the column area. This, we thought,
allowed us to specify pile caps with fc' = 6000 psi. The column load Pu =
8620 kips, resulting in a bearing stress of 5390 psi. The allowable
factored stress as I calculate it is 2.0 x (phi = 0.65) x 0.85 x fc' = 6000), or
6630 psi. So far, so good.
However, when designing the caps and their reinforcing using the Strut and
Tie provisions of Appendix A, I can find nothing which would allow us to double
the stress. So, the allowable (factored) stress in my struts is only (phi
x Beta-s x 0.85 fc'). It seems that I therefore can't carry the
12,000 psi column on a 6000 psi cap.
Does anyone know if the doubling provision of 10.17 can be applied to
struts in the Strut and Tie model? Section A.5.2 may allow some increase
in stress for "confining reinforcing," but seems to exclude confining
Another, perhaps related question, is when the "bottle-shaped" struts come
into play. When can the struts be considered prismatic, and when will
they swell out to the bottle shape? The Commentary to the definition says
that "a bottle shaped strut is a strut located in a part of the member where the
width of the compressed concrete at midlength of the strut can spread
laterally." When is this not true? It seems that all
struts will meet this definition in at least one dimension.
This is problematic because if the swelled shape is assumed then confining
reinforcing is mandated by A.3.3 to get up to a Beta-s value of 0.75. If
instead a prismatic strut can be assumed (by some unstated justification) then
Beta-s can be taken as 1.0. Yet the allowable stress for the bottle shaped
strut is still taken at the ends, before it has swelled.
If anyone can shed some light on these questions, I'd appreciate it.
Mike Hemstad, P.E.