ACI 350.3-06: What about dynamic earth pressure?

[ARS97]

I'm designing a rectangular concrete tank and I'm working through the seismic loads per ACI 350.3-06. I have the various dynamic forces calculated from Section 4.4.1.....Pw', Pr, Pi and Pc. What is stumping me is the consideration of "Peg" in the base shear calculation. "Peg" is the lateral force on the buried portion of the tank due to the dynamic earth and groundwater pressure. Chapter 8 is referenced for Peg, but it doesn't appear to include any further details, except for resultant locations.

I know the static lateral pressure from the soil and groundwater. In this design, the tank is being designed for groundwater at the surface (21' depth). The lateral pressure will be due to the buoyant weight of the soil (120 pcf - 62.4 pcf = 57.6 pcf = 57.6 psf/ft) and the groundwater (62.4 pcf = 62.4 psf/ft). This will be assumed to produce a simple triangular pressure distribution (total of 120 psf/ft) over the 21' depth. This equates to a static pressure of 2,520 psf at the wall base.

Is Chapter 8 simply stating that you are to use the full static load as the dynamic load (Peg)?

 

[Lomarandil]

I don't have any experience in buried tanks or that code, but from what I know of soil dynamics, I'd not consider the static and dynamic soil pressures to be equivalent.

That said, there are a variety of methods and considerations to be taken into account for soil dynamics problems -- you probably want to consult a geotech.

----
The name is a long story -- just call me Lo.

 

[WARose]

In general, you figure the bearing pressure as being equal to static + dynamic (from lateral events or whatever). However, some soils (especially loose, sandy soils) cannot take a sudden increase in bearing pressure from a dynamic load. This can result in unwanted settlements and/or liquefaction. As Lomarandil said: you may want to get a Geotechnical Engineer involved in this if you have concerns.

 

[ARS97]

I guess my question is, are dynamic lateral soil pressures typically ignored in these designs? It seems like the overwhelming majority of the lateral force would be generated through the effects of the contained liquid and concrete mass.

 

[Lomarandil]

It's hard to know without knowing more details of your project. Three criteria that immediately come to mind:

What is a quick characterization of your soils (e.g. risk of liquifaction as WARose mentioned)

What proportion and depth of the tank is buried? (e.g. what is the rough magnitude of lateral pressures from outside the tank vs inside?)

Where is the project located? (e.g. are dynamic effects due to earthquakes likely to be significant?)(I can't think of any other plausible dynamic drivers for most situations, but that's not to say there couldn't be others)



----
The name is a long story -- just call me Lo.

 

[WARose]

I guess my question is, are dynamic lateral soil pressures typically ignored in these designs? It seems like the overwhelming majority of the lateral force would be generated through the effects of the contained liquid and concrete mass.

It's been so long since I've used ACI 350.3-06, I cannot say. However, you do not ignore earth pressures in any design (static or dynamic).

 

[ARS97]

The tank is buried it's entire depth (20'). Soil is generally a silty sand (SM). Attached are my WIP calcs for the ACI 350.3 procedure. It seems odd that ACI 350.3-06 doesn't address it directly, at least from what I can tell.

 

[Lomarandil]

We're getting to the point where without digging into ACI 530 (which I don't have), I won't be able to help you further.

But based on your first two data points -- buried full depth and silty sand -- my judgement would be that you absolutely should not neglect dynamic soil pressure.

The only other tidbit I can provide is to steer you toward a few methods (ideally, performed with the guidance of your friendly local geotech). The Mononobe-Okabe method is probably the most widely known and used (it's codified in AASHTO). Last I heard though, it was considered to be lagging behind the bleeding edge methods like Steedman-Zeng, Newmark(?) or FEA methods.

----
The name is a long story -- just call me Lo.

 

[bones206]

I have used the Elastic Solution Method per ASCE 4-98 to determine dynamic earth pressure distribution for large junction chamber walls (applied in a FEA model). I come from the nuclear power industry, so it was a method I was already familiar with. It is more appropriate for non-yielding walls than the M-O method, which is based on active pressure conditions. When I first started using ACI 350.3-06, I had the same question about the dynamic earth pressure and was met with shrugs and non-answers from the experienced water structure engineers in my company. I think they traditionally ignored sloshing forces and seismic earth pressures for buried tanks.

 

[ATSE]

Engineers sometimes incorrectly ignore actions they don't understand.
Sitar (UCB) has argued for years that for low walls and moderate seismic, static design will govern. However, don't ignore dynamic forces. The dynamic event will of course create more demand on the retaining wall vs static only. If the static ground water is above the footing base (no drainage condition), then the lateral seismic pressure for saturated soil will be even greater.
Seismic wall pressure is very complex. The geo and structural community does not agree on how to calc seismic pressure distributions, but Seed-Whitman is probably the most accepted. Use y = 0.6*h per ACI 350.3 ch 8. [Sitar probably argues for y = 0.33*h or 0.5*h].

 

[ARS97]

The project specifications (which are based on ACI 350 largely) require a static design condition where the groundwater level is assumed at the top of the tank (1' above final grade). This requirement is for both buoyancy requirements and strength checks. For such a condition, you're essentially considering the buoyant weight of the soil (dry wt - 62.4 pcf) and the water pressure (62.4 pcf) acting as a fluid (no internal friction) over the full height.

I do have access to a geo report that was originally completed about 10 years ago for this site. In that report, it is noted that groundwater was not encountered during the investigation, however, due to observations of wet areas within the boring and typical seasonal variations, the groundwater should be assumed to be 12' below grade for design. Obviously though,the project specifications, and ACI 350, override this recommendation......at least for the static condition.

The dynamic/seismic condition is a separate consideration in my mind. It seems completely unreasonable to assume the static flooded condition can occur simultaneously with a seismic event, especially considering the location (Ss = 0.139 & S1 = 0.052). To me, for the dynamic/seismic condition, I think that the groundwater level should be assumed as noted in the geo report (12' below grade). This produces a MUCH different, and MUCH lower magnitude, static lateral pressure distribution. The dynamic/seismic loads would add to this revised soil condition, however, based on what I've calculated so far in ACI 350.3, it seems very unlikely that it would be enough to approach the same magnitude of forces as the static condition, simply because of the groundwater level assumptions.

There is nothing of substance in the geo report dealing with the dynamic lateral soil load during seismic events, so maybe I can touch base with the geotech engineer and see if they could possibly add an addendum to it. I'll also have to look into the various methods listed above and see what I can track down. I just was a bit surprised at the lack of detail within ACI 350.3 on that subject....you're just left hanging.

 

[ATSE]

For pseudo acceleration parameters Ss = 0.139, I wouldn't be too worried about seismic action.
Since seismic capacities don't include an environmental durability factor (Sd) penalty, static loads (hydro on the inside, soil on the outside) will govern your wall thickness and reinforcing design.