Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: pier design

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Well, I think that you CAN come up with "springs" as supports using the P-Y Method (which LPILE does), but the trick with the springs is that they are not linear.

In the past, I have been concerned with interaction not only of the PILE with the soil, but the entire structural system (including the columns, superstructure, AND the pile/soil).  So I set up a spreadsheet which used P-Y curve construction equations, which I got from the book "Pile Foundations in Engineering Practice" by Prakash and Sharma.

I assumed initial spring constants, ran a STAAD-III model, took the displacements at the springs, and used them to obtain NEW spring constants, reanalyzed, ad nauseum.

You converge to a "solution" fairly quickly in that case, and the spring constants are "linear," but you end up with a "secant modulus" that yields the same deflected shape of the pile (and, coincidentally, also takes into account the interaction of the structure as well.

-----Original Message-----
From:	BVeit [SMTP:BVeit(--nospam--at)]
Sent:	Friday, March 06, 1998 11:22 AM
To:	seaoc(--nospam--at)
Subject:	Re: pier design

In a message dated 98-03-05 21:07:12 EST, arvelw(--nospam--at) writes in
response to Martha Alic's pier design question:

<< Also check out Bowle Fifth Edition of his text.  It is usually straight
 and has alternative methods. >>

I use Bowles also.  Because of the way he sets his method up,  if you model
the pier using springs with stiffnesses based on the soil stiffness and the
area of the pile, you can get answers identical to Bowles.

 I use Visual Analysis, but I'm sure any package would work.  It is a good way
to go because you can model constraint (say through a collar) with a stiffer
spring.  It shows you the moment and shear throughout the pier, just as Bowles
does.  This method also allows some flexibility for stratified soils.

The only problem with it:  It's like pulling teeth to get a soils engineer to
give a realistic soil elasticity modulus (aka spring constant.)

Short of an actual pile test, I don't know how to estimate one either, so the
modulus numbers tend to be the loosest part of the whole process.  Can anyone
comment on how L-Pile gets around this difficulty?  Doesn't the finite
difference method used in L-pile do basically the same thing as Bowles?