Does anyone else think...soil pressure

[dougantholz]

That soil pressures are already really high to be in the factored load combination as 1.2D + 1.6L + 1.6H?
When I was in school a decade ago we used 45 pcf for at rest pressure, now I'm seeing 65 pcf in a lot of soils reports for cohesive soils. It seems like this already has a healthy factor of safety on it, and should get a 1.0 or maybe a 1.2 load factor.
Anyone else think I'm right, or do I alone carry the banner of crazy...

 

[DaveAtkins]

Yes, I agree lateral soil pressures are most likely conservative, BEFORE the load factor increases them even more. I see almost vertical excavation cuts at many jobsites, which seems to indicate lateral soil pressure is zero!

However, you must follow your geotechnical report and the governing code.

DaveAtkins

 

[dougantholz]

Agreed I have to follow the law, I'm just not happy about it. I mean has anyone here ever heard of a concrete retaining wall, built according to plans failing because it was under-reinforced? I've seen enough overturning and in-adequate dowel walls, but never an actual concrete failure.

 

[Lion06]

I am thinking that if you have seen retaining walls fail by overturning and it was designed properly to begin with that the soil pressure was actually UNDERestimated, right?
I think it much more likely that it is designed closer to the OT limits than to the concrete strength limits for varioius reasons (not the least of which is that if you need 0.47 in^2/ft, you are probably going to put in #7 @ 12" and call it a day. You just increased your capacity by almost 30%.

 

[Lion06]

I wasn't presuming to know the relative conservativeness of todays lateral earth pressures, just making an observation.

 

[LobstaEata]

Hi Doug

There are enough of us old guys in the engineering community to help you carry the banner of crazy. You can't work in this business for too long with all the code and design changes before you start second guessing yourself at every turn. Anyway, here's my take on your subject.

Rankine's theory states that the pressure exerted by an earth backfill against a retaining wall can be computed with reasonable accuracy on the basis of theory only for conditions rarely encountered in practice. That said, theoretical earth pressure calculations can rarely be justified for a particular retaining wall because the physical characteristics of the backfill are usually not sufficiently known. Regardless of that, Rankine's theory has been widely used to design most retaining wall structures in the US. Go figure.

Assuming you are using Rankine's theory, 45 pcf for at-rest pressure suggests that the soil density would only be 90 pcf, considering that Ko conservatively runs about 1/2 the overall soil density for a cohesionless soil. Typical granular soil densities are about 130 pcf, which would explain the 65 pcf at-rest pressure you've been seeing in your geotech reports and would be appropriate for designing walls on unyielding foundations (ie. mats or pile supported).

The 45 psf pressure appears more appropriate as an active earth pressure, which would be about 1/3 the overall soil density for cohesionless granular soils. Perhaps this pressure is what you remember from school?

If you are into cohesive soils however, you definitely will not want to not use a cohesive soil backfill and should advise your client to replace it with a clean sand and gravel. You also would be wise to review long term settlement estimates with your geotech based on the intended earth surcharge. If the expected settlement cannot be tolerated by the wall (or mat), then a deep seated foundation might have to be considered.

I hope this has been helpful

PE in New England

 

[dougantholz]

Lobsta -
Thanks. That was pretty good looking back at Rankine. Don't you think that's pretty well known quantity and that the service load already basically gives an expected upper bound.
My problem is not the 65, but rather the 1.6 LF in ASCE. I think it should be closer to 1.2. Those load combinations just seem to be as much "feel" as science.

 

[msucog]

not to stir up the conversation here, but as a geotech myself, i'd argue the term "conservative" when dealing with a material that can vary dramatically shovel full to shovel fulls, comprises most of the material quantities that exist on a project, can be dramatically effected (by exposure to weather, mixing, compaction, pressure, etc) and where the engineer has very little actual control over the material. concrete and steel are highly "controlled" and tested materials. soils are by far many times more complicated a material than anything else on a jobsite...and it is usually the most misunderstood and problematic issue on construction projects too. you're not "building a watch"...your playing in the dirt. i've seen some ugly things happen in my short career due to soil conditions...sometimes due to something obvious and sometimes due to the crazy hand mother nature deals.

however if you substitue in graded aggregate base for foundation or backfill materials, i suppose you'll get a much more "desireable" response from the geotech as far as recommendations go. but, i could see where even that cannot fix all issues on a jobsite. and the cost to use all GAB will also be shocking to the owner. and believe it or not, i've seen contractors screw even that up (building with GAB)...they are amazing creatures (those funny contractors that is).

 

[hokie66]

The problem with reducing load factors or safety factors for soil pressure is the pressure can change a lot over time. Cracks occur, get filled with debris, expansion takes place, etc. The other thing is that the hydrostatic component is often wrongly neglected. I think the practice by some geotechs of giving "equivalent hydrostatic pressure" is partly to blame for this.

But most problems with retaining walls occur because of skimpy design and construction errors. There is another current thread here about a leaning wall. Turns out the bars are in the wrong face. So msucog's acceptance of concrete and steel as "highly controlled" is not always appropriate either.

 

[BigH]

The problem with the concrete and steel as "highly controlled" is not the material itself but bloody poor workmanship and not following the design - as you correctly pointed out. What is the coefficient of variation for steel? For concrete? Almost infitesimal for the first and low for the later - whereas for soils, it may range from 10 to 40% on the strength values - so as msucog points out, the soils can be much more problematic given a very good standard care of workmanship.

 

[msucog]

hokie66, you are absolutely correct about the term "highly controlled" not always being appropriate. i was referring more to something like bigh's point about shotty workmanship but even just the control of the material is tough. most of us have run across projects where there have been problems with concrete strengths. now imagine if concrete was made up something other than "controlled" materials at very specific proportions...just imagine the headache that would cause...now maybe you see the internal struggle of the geotech. and as one other point to bigh, the variation can occur over very short distances (i've personally seen such variation within one shelby tube sample--so, within about a foot or so).

but back to hokie's point: i can assure you that no matter how much you inspect something, the contractor will quite often do his best to screw it up after you walk away (even if it's perfect when you looked at it). this is why i fully support a thorough special inspecdtions program...at least it provides some confidence that things are close to what it should be at that point in time...but again, once you walk away, it's a coin toss.