Hydro-Compaction of Sand

[eric1037]

OK. I have heard this from contractors time and time again. They claim that by totally saturating soil, then letting it drain, it will "compact" the soil.

In all of my years of experience I have never seen this as an effective technique. I can understand that it is possible for fines to migrate between the soil particles and settle at lower elevations, but it seems like this would have very limited effect and only on "dirty" sand overlying "clean" sand or gravel.

I have read about this technique being effective for collapsible soils that have a structure that is weakly cemented by a water-soluble chemical or small amount of clay.

Also, I would imagine that you would need a very large head of water to achieve this affect. And by that point it may be more the weight of water inducing the consolidation than the water itself.

I have reviewed the paper at the following link and it seems to indicate that even where there is improvement, it is spotty at best and the improvement level wasn't enough to achieve specified compaction requirements.

http://www.ejge.com/2004/Ppr0445/Ppr0445.htm

I am an open-minded engineer and I always like to consider cost-effective alternatives. So, what do you think?

 

[eric1037]

I just found a reference to this in the Peck, Hanson & Thornburn Foundation Engineering, 2nd Ed. On page 197 there is the following:

"Backfill in utility trenches and other spaces later to be covered by floor or pavements should preferably consist of the best available well-graded sand and gravel mixtures. Materials for this purpose are often dumped into place loosely and then flooded in an attempt to compact them. This procedure, although still widely used, should not be permitted. In cohesive backfills it inevitably leads to weakening and softening of the soil and to future loss of support and subsidence. In uniform or fine sands it can do no more than cause the collapse of the extremely unstable zones associated with bulking and leave the sand at a density index close to zero. If the backfill is a well-graded sand and gravel, the effect of bulking is negligible and no benefit is derived from flooding."

 

[GeoPaveTraffic]

Believe it or not our local Metropolitan Sewer Distric still allows this method of compaction (locally called "Jetting") in areas not under pavement. Most offten it results in settlements measured in feet over a time frame of several years. However, if done under the correct conditions the amount of settlement in a approximately 10 foot deep trench can be as little as 6 inches.

Please don't read the following as an endorsment of the method, I despise it. To do it well you must have a free draining backfill material, a lot of water and, most importantly, a way for the water to exit the backfill. I don't know if the fill needs to satuate then drain or if the compaction is more an effect of the flow of the water.

I would only consider allowing this method of compaction in areas where the fill will be setting for several years before additional improvements are constructed.

 

[oldestguy]

Interesting subject. Just because it is done a lot doesn't make it a good method.

I note that a city near here with mostly clean sand soils has done it for years (in sewer trenches) and seems to find little subsequent settlement. However when checked for relative density or percent compaction, in my experience, it did not meet our usual specs.

Another interesting use is in sand fill inside attached garages to new houses. The one case that sticks in my mind was a caved in basement wall that resulted. Contractor said it had been his practice for 25 years. Guess he won't do it any more.

So, like so many other "standard practices" sometimes they work and at other times they don't.

When professional liability is in the picture, it is better we use what works every time, generally because we know the mechanism present.

So, here is another question: Why is topsoil stripped before paving or a building on fill?. Are we prejudiced to black? I ask this a lot when this question comes up.

Lots of building (pavements and structures) has been done on topsoil and it works many times. I've also seen it fail (not my jobs).

Take the case of Pontiac,IL. Four feet of topsoil and it is not stripped usually. K-mart building and pavement sit on that. In that case it was my recommendation to leave it.

I have my reasons for and against. What are yours?

 

[BigH]

Hey all. This idea of compacting sand by flooding - well, the idea should be relagated to the waste bin (along with LRFD!! - just kidding, really guys!!.

I have run into this so many times myself. If the water can't flow, how can seepage forces (the only possible manner of compacting a sand deposit) develop? So backfilling and flooding sand in an excavation in clay is rediculous. In India, the contractor regularly tried to do this - this with a site ground gradiant of something like 1m drop in 3000m. If sand could be so compacted, why do we have loose alluvial sand deposits?

Now, I have done this once - in Vancouver to fill the gap behind a retaining wall and a vertical diamicton (till-like) excavated slope. But, we used round-grained river sand and had something like 10 to 15 m of pressure head for drainage - we filled in small lifts and flooded continually as the river sand really drains quickly. We never did any compaction testing - wouldn' thave wanted to go into the small space - but seemed to work okay. The telling point, in my view, though, was the very free draining nature, the rounded particle size and sizeable seepage pressure gradiant from 15m of head.

Seldom, though, do contractors have this situation. Interesting the 1983 (or so) City of Vancouver, British Columbia manual for sewers/backfilling disallows contractors to flood - yet, I watched countless times as city crews did it!!

 

[MRM]

We've recently witnessed yet another unsuccessful attempt at hydrocompaction...

We had one of our technicians head out to a site to do some DCP testing on a foundation subgrade. The tests indicated well compacted backfill near the surface, and very loose fill below that (no surprise there; the contractor thought he could compact 6 feet of fill at the surface using a plate compactor).

After we notified them of the results, without granting them our blessing on the subgrade, they called us out on the site again to recheck. They said they had flooded the areas over the weekend, and the subgrade should be "nice and tight now." We checked again using the DCP with the exact same results at lower levels.

It was a good day for two reasons. First, we were able to once again show this guy that flooding an area does not compact the loose soil to any meaningful degree. Second, we had yet another chance to show a contractor that you can not pull the wool over our eyes if we use a DCP to check your compaction at depth! It’s really hard for them to argue with you when the instrument sinks in under its own weight around 2.5 feet down.

 

[BigH]

MRM - good case history for all of us to remember.