for a cantilever wall use active, for a propped wall use passive, and for resitance to sliding e.t.c. use passive.

Suggest to use "at rest" pressure for both cantilever & propped walls.

The suggestion for the latter case was well established. The argument for the former is that though it only takes slight deflection on top of the wall to allow the development of active pressure, however, the soil will reach an new equilibrium back to the at rest state. Then, the wall deflects further more,...active pressure...so on so forth, until the soil reaches its final equilibrium state. At that stage, the total deflection could be significant. or the wall failed before reaching the final. Thus, it is suggested to use at rest for the cantilever wall as well.

For passive pressure to occur, it requires the wall move into the soil, such as pressure in front of a free moving retaining wall, or a deadman in tie back system.

In most occasions, we apply active pressure on those temporary, or less important, free standing sheet piling walls.

An easy explanation supplied by the Geotechnical Engineers in the reports is to use at-rest for braced or supported walls and active for walls without top supports.  To develop active pressures you need about .001H of movement, where H is the wall height.
Passive is a resisting force only activated by large movements in the order of .01H.  You wouldn't want that much movement except for retaining walls where movements are not critical.

My criterion is

- Active: general walls (for phi=30º -> k=0.33)
- Rest: bridge abutments and important walls that cannot have great displacements (for phi=30º -> k=0.50)
- Passive: I usually model the soil with springs and ballast coefficient and the passive means the maximum possible reaction of the soil in piling walls (for phi=30º -> k=1.33)

http://NotOnlyBridges.blogspot.com

Ok, that's what I thought. I usually design with at-rest for retaining walls acting as foundation walls at the same time (my case).

JedClampett: Do you have a (book/article) reference for the 0.001H criteria? That doesn't seem so bad, 4 mm on a 4 m wall..?

To fully develop active earth pressure, you need to allow 1-in of movement on a 10-ft tall wall (think rotational movement).  If you are not willing to allow that movement (or if the wall is braced at the top and it just can't happen), then you should use at-rest earth pressure, which is about 1.5 times greater than active earth pressure.

Bear in mind, there also has to be movement at the toe to mobilize passive earth pressure.

Bear in mind also, if you are designing a sheeting system, Terzaghi and Peck found that the earth pressure envelop is either rectangular or trapazodal, depending on whether you are restraining fine- or coarse-grained soils.

Some folks design cantilevered retatining walls using at-rest earth pressures on both the "active" and "passive" sides as they don't want to count on any movement.  This would then require stability from base shear or a gravity mass above the wall footing.

I'd hope that a geotechnical engineer would be able to explain this to a civil/structural designer, but then again. . . .

f-d

¡papá gordo ain't no madre flaca!

I was afraid you'd ask that:
In Winterkorn and Fang, "Foundation Engineering Handbook" Table 12.1
Sand:
Active Pressure: Parallel to Wall .001H
Active Pressure: Rotation About Base .001H
Passive Pressure: Parallel to Wall .05H
Passive Pressure: Rotation About Base >.1H
Clay:
Active Pressure: Parallel to Wall .004H
Active Pressure: Rotation About Base .001H
Passive (No values given)
My favorite Geotechnical reference, but I'm sure it's been out of print for a while.  My copy is dated 1975.

fattdad: Thanks for the reply. I know this may sound dumb but you actually cleared the doubts in my mind when you said:

"Bear in mind, there also has to be movement at the toe to mobilize passive earth pressure."

That actually can't happen for the wall I'm designing at the moment.  

http://www.vulcanhammer.net/utc/ence461/461-sl20.pdf

Contains a table of minimum movements.