Lateral earth pressure 3

[Rancorinco]

On the design of basement foundation wall of span "L" between the basement slab-on-grade horiz diaphragm and the first-floor horiz diaphragm, with soil backfill against one side up to a height "h" from the wall base, what exactly is the lateral load against the wall??

I've asked the problem to several geotechs, who all recite the faith: soil is a substance between water [K = 1] and solid [K = 0], where the number K is the ratio of horizontal to applied vertical pressure. So by this, the "Fluid Analogy": given a soil of unit weight U, the lateral pressure at a depth d from the top of soil is H = K*U*d, whose load distribution is a triangle, with the apex "0" located at top-of-soil.

The problem is: what is the value of "K"? Before he can answer, a Geotech will want to know also if the soil is "active" or "at-rest". Herein lies the rub. The "at-rest" value of K is higher that the "active" K, therefore it requires a heavier and more costlier foundation.

Yet I've been on jobsites where the soil has been cut vertically 8' or higher, standing like the face of a cliff. The geotech gave 60 pcf at-rest design lateral pressure. Yet, as we can see, in this "service" state, there is no lateral pressure, "at-rest" or otherwise.

Does common sense need to kick in? What is a reasonable design parameter with a wall? Can you backfill with gravel and then design 40 pcf fluid pressure?

Also, when you are designing concrete, concrete is designed at the ultimate state. Seems to me if the soil is "active", then it is failing, then the soil is at ultimate, too. So why does everyone factor the lateral soil earth pressure [see ACI 318 Chapter 9 load factors]?

So if the K can be reduced, and the load factor eliminated, what is there to design? Answer: minimum concrete reinforcing ratio. Most house foundations don't even have that.

 

[cdh61]

Rancorinco,

Just a note, do not think that the lateral soil stress will not "kick in". There are lots of people who are killed every year due to standing soil cuts and think it will be OK to work beside. Without looking I can recall a handful of people killed while working adjacent to over steepened cuts.

In addition, it is not as simple to say just add a layer of granular behind the wall and then you can decrease the lateral stress based on a lower K value.

Worst house case I have see - Owner informs operator to backfill around house foundation, no lateral soil support for footings or joist support for walls. Footings and walls move into the basement area over 1.5 ft during backfilling. Combination or errors, but cut slope fails during backfilling, large excavator on cut slope and no support for the footings or walls.

regards

 

[BigH]

cdh61 - nice post!
This is the "rub" of earth pressures - if you have rotation, then you can use the "lower" value - if you restrain rotation, it will be more - makes sense to me. The standing analogy doesn't work in that in many cases it is the negative porewater pressure giving it "stability" but when wetted, . . . Stay out of the excavation.

 

[DRC1]

This is the difference between soils and steel or concrete design. Soil properties are generally not well known with out considerable testing so geotechs tend to use more conservative values than structural engineers. So 40 psf may be okay, but without testng of the actual backfill amterial,you can't tell for sure. A lot of house foundations are un reinforced, But most are only about six and a half to seven feet below grade to the top of the slab, which doesn't really generate large earth pressures, especially when you consider the florr slab and the ceiling brace the floor. As you get deeper, the loads get larger quickly. The change in load is equal to the square of the change in depth.
There are a lot of reasons that a cut that should fail will stand straight up. One is that moisture in the soil creates a temporary cohesion. Another is that the soil just doesn't know any better. Eventually time and gravity will catch up. Usually suddenly.
Another point is when you backfill the building, you will require compaction which induce soil stress at least equal to the at rest pressure. Once the compaction is complete, you will probably have a pressure distribution close to the tradional triangular assumption.

 

[cap4000]

Soils have a partial arching self standing capacity through its internal friction and or cohesion. As long as the soil tension forces equal the cohesive strength of the soil you are temporarily safe. Although your cut is vertical and unbraced, rain, snow, groundwater and or gravity will eventually change its properties drastically. For rigid bridge abutments I would use at rest, for more flexible retaining and basement foundation walls I always use the active pressure. I have always used the 1.7 live load factor over the past 30 years mainly because no 2 geotechs ever exactly agree as its not a perfect science. I find its that what makes it extremely interesting. Good Luck.

 

[mapelCE]

I'm sure you've got this sorted out by now but I'll throw in my two cents.

K values for walls: Most geotech authors will cover this in depth and even Lindbergh discuss it in his review manuals for the CE. At rest is best used for walls that can't move. i.e. basement walls supported at the top, walls sitting on hard rock that don't allow rotation, bridge abutments, etc. I think the rule of thumb for movement is H/200. If you can allow that you can use Ka in determining active pressure. You can use a triangular distribution for a backfilled wall no problem. Things change if you talking about shoring excavations so this only applies to walls backfilled after construction.

SAFETY: Having been in heavy construction for over ten years now I can tell you that working 400' in the air is no where near as nerve racking as working 40' in the ground. Excavations are dangerous, bottom line, because of all the assumptions and unknowns. Get buried in a trench up to your armpits and you will most likely suffocate and die before anyone can free your chest to breath. Just be cautious. Cohesive soils can change relatively quickly. OSHA rule, 4' max vert. They allow 1/2:1 in type A soil short term but our standing rule is anything steeper than 1:1 gets a slope stability check automatically. One thin lense of sand can ruin your day.

Good Luck with your project.

 

[PMR06]

Interesting replies... though Rancorinco's question regarding WHY we factor lateral earth pressures has not been answered. Indeed, if the geotech specified values are conservative, why is the load factor not smaller... say 1.2 as for fluid? Obviously soils are not a "well defined" fluid, but if the design values provided by a geotech are already "factored" for conservatism, why factor again? There is no doubt that soil excavations are not to be taken lightly and that safety should a top goal of the design, but is conservatism on top of conservatism a logical way to design?

 

[BigH]

PMR06 - isn't this due to LRFD vs traditional soil engineering using working stress design?

 

[DRC1]

Another thing to point out for a fluid, the actual load distribution is triangular. The soil loads against the wall are not triangular, and vary due to several reasons, wall siffnes and rotation, soil compostion and permiability,degree of compaction and how that compaction varies through the soil. The actual distribution is actually most likely parabolic, although this an ongoing debate to which much study research and beer has been devoted. In the end, the triangular distribution works and is probably conservative, but by how much we really don't know. The other reason is that unlike structural material, soils are highly varible, with very small samples representing large ares with generally less than sufficent laboratory testing. So you have considerable uncertinty to which the current systems seem to have adapted .