John,
If you wanted to prevent collapse limit states you would want an index
defined in terms of the shear stresses being mobilised. If the external
pressure is q_0 and the foundation pressure is q, the correct shear index is
|q - q_0|. If you exactly replace the external surcharge with foundation
contact pressure, there is no tendency to shear, the mobilised strength is
zero, and the safety factor is infinite.
But if you wanted to prevent excessive settlement, the situation is more
complex. If the calculation were for undrained settlement of clays, then the
only mode of deformation available is shear and the precepts of the previous
paragraph apply. But if there is some consolidation settlement, the absolute
effective stress would be of interest.
Since FoS is (often wrongly) fixed in relation to protection against
collapse, it must follow that net pressures are to be preferred - if by
"net" we mean the differential between the foundation pressure and the
external overburden pressure at the elevation of the contact plane.
Best wishes,
Malcolm Bolton
-----Original Message-----
From: Geotechnical Engineering Email List
[mailto:[log in to unmask]] On Behalf Of John McKinley
Sent: 11 May 2005 16:52
To: [log in to unmask]
Subject: Use of net bearing pressure for ultimate limit state design of
spread foundations
Dear colleagues,
To assist in curriculum development, I'm curious to find out whether there
is a general preference within the geotechnical community to define factors
of safety for spread foundations in terms of net pressures:
FOS = (q_ult - q_0) / (q_design - q_0)
or in terms of gross pressures:
FOS = q_ult / q_design
I'm aware that practice varies between organisations and companies. However,
I wonder whether there is a general preference for one or the other.
Philosophically, I think that it makes more sense to define an overall
factor of safety as the ratio of the load / pressure / action at the
ultimate limit state to that at the design state, rather than as the ratio
of the increase in load / pressure / action at the ultimate limit state to
the increase at the design state. It seems to me that using the net
pressures leads to odd results for special cases, such as buoyant
structures. Leaving aside the question of partial factors (as in, for
example, Eurocode 7), I also think that the gross pressure approach is more
consistent with what would usually be done for design of non-geotechnical
structures in civil engineering.
Yours sincerely,
John D. McKinley
------------------------------------------------------------------------
Dr. John D. McKinley +44 (0) 28 9097 4690
Lecturer in Environmental Engineering
School of Civil Engineering, Queen's University Belfast
www.prb-net.qub.ac.uk/eerg/People/Academic_staff/jmckinley/jmckinley.htm
------------------------------------------------------------------------
|