Tests for Running/Flowing Sands 1

[Fackler]

Are there any tests or guidance documents to determine if a sand is a flowing sand?

I am being pressured to perform a soil excavation as part of a remediation project, but my geologist strongly believes the sands become flowing sands once we get below the groundwater table. I would like a way to prove that the sand will flow, potentially damaging a nearby highway or building, other than through my geologists visual observations, which unfortunately holds little weight with the ones pressuring for an excavation.

 

[GeoPaveTraffic]

Most sands will flow if excavated below the water table. The only way they will NOT flow is if there is a significant amoount of clay in the sand OR a very small amount of water available to move the sand.

Listen to your geologist.

 

[PEinc]

Remediation projects usually involve excavating contaminated soils and then replacing them with compacted clean or cleaned soils. This usually requires that the work site be dewatered and the water be treated. This should solve the running sand problem. If you are close to streets or structures, you may also need an excavation support system or underpinning.

It's hard to dig a hole in the sand at the beach unless you dewater to below the desired excavation depth.

 

[fattdad]

This is a flow net problem. You only get "flowing" sands if the water flow at the discharge point is at the "critical gradient". The critical gradient is a function of the bouyant unit weight of the soil and the unit weight of water. I cannot recall the math right now (just returning from my daughter's swim meet and all), but that's what the issue is (from my perspective). You can control the gradient at the discharge using a horizontal hydraulic barrier (e.g., sheet piles), well points or deep wells - for the latter two, you just would not have a discharge gradient from the sand mass to the excavation.

I do not agree it's a non issue if the sand mass has some magic amount of silt or clay. I've dealt with this issue in sandy-silty-clay soils, the mechanics of failure are different but the problem is the same, a really messy excavation.

f-d

¡papá gordo ain’t no madre flaca!

 

[DRC1]

The best way to prove may be to rent a small backhoe and dig test holes. Once you hit water, dig a little deeper and wait a few a little bit as the water flows in. Once it stablizes, try continuing to dig. Document the effects of the digging. It should be like trying to dig beach sand below thw tide line.

 

[Fackler]

Several of the items PEInc mentioned are the reasons I am not comfortable excavating this site. It is a relatively narrow area, maybe 50 feet from a Federal Highway to the business. I have also been burned twice in the past year with sheet piling causing or at least contributing to damage to buildings and roadways on other excavation projects. Plus we would have to dewater 10 feet or more of what can be described as beach sand. (Did I mention I am in Florida.)

I am looking for some scientific/engineering methods to aid in our attempt to persuade local regulators that excavation is not the best alternative for this site. Fatdad, I would appreciate those equations if you get a chance, they may be helpful.

 

[fattdad]

Fackler,

Several items to consider:

1) If you don't use sheeting and you just dig into the ground (i.e., well below the water table), you will have a transient condition where the sidewalls of the excavation are under their full hydrostatic condition (i.e., the flow net will not have had time to develop). As such, you will have much larger gradients at the discharge face then you would have if you dug more slowly.

2) If you dig more slowly (like who has the time. . . ) and use a sump, you can dig, dewater and establish the steady-state flow net concurrently.

3) I really don't know whether there is an overburden layer atop the clean sand layer. If so, there could be a phreatic surface in the overburden that could affect the calculation of hydrostatic pressures in the clean sand layer. While the overburden layer may not be yielding free water in the excavation sidewalls, it may be affecting the hydrostatic pressures in the clean sand.

4) After considering the dynamics of the problem, you need to determine the hydraulic gradient at the face (or toe, which may be where the highest gradient is located) of the excavation. For the case of the immediate time after the big cut into the sand layer, this gradient may be HUGE (i.e., well in excess of unity) - no doubt exceeding the "critical gradient" for initiating piping-type failure.

5) If you have any way of controlling the transient gradients there will nonetheless be a steady-state gradient where the phreatic surface is established toward the excavation. Calculate this exit gradient (i.e., deltaH/deltaL (change in head with respect to the change in travel distance). For the case of a "free-surface" flow net (i.e., where the phreatic surface is warped downward to the exit point) and you have a vertical discharge point, you can use the curve equation for a confocal parabola to establish the phreatic surface (you can also just draw it).

6) With the phreatic surface shown, draw you flow net "sqaures" and at the discharge point, determine the change in head for the change in distance over the final square. Compare this gradient to the ratio of GammaB/GammaW (bouyant unit weight divided by unit weight of water).

This is a theoritical outline of the dynamics that may or may not help. There is too much uncertainty in the transient period between the initial cut below the water table and the development of the flow net for the revised boundary conditions. There is no doubt that a quick cut into the sand will show instability. The question remains whether you want to first do some dewawtering to initiate the drawdown of the phreatic surface and then look at the discharge gradients at the subgrade after the excavation is complete.

Warning - I type faster then I do engineering - but this is what comes to mind when I try to assist.

Good luck.

f-d

¡papá gordo ain’t no madre flaca!