Engineered fill specifications

[MRM]

Hi everyone,
In my area of Michigan, we usually call for "end product" specifications regarding engineered fill placement. That is, we say what we what in terms of relative compaction and the contractor obtains the specified percent relative compaction using as much time as required, in any way they think would be the most cost effective for them depending on the project, soils, etc.

I'm curious to hear if you have any examples of projects of where the engineer uses a "method" specification, calling for a specific fill placement method based on the fill compaction characteristics determined in the lab and the contractor performs that specific fill placement technique. Whether the fill is compacted adequately in the field after the technique is not a concern to the contrator. If it needs to be recompacted, it would be at the owners (or engineer's) cost.

Also, are you aware of any projects were a specific relative density is required for engineered fill as opposed to relative compaction? I assume we use a relative compaction specification system since it is so much easier to obtain a maximum density in the field, rather than perform the min/max density tests in the lab required to develop a relative density specification.

Thanks for your comments!

 

[Focht3]

Performance specifications, rather than specifying the contractor's "means and methods", for typical construction projects became commonplace when geotechnical engineers became able to predict field density (using "Proctors&quot and measure the contractor's performance (using sand cones, balloon densities, drive cylinders, and nuclear density gages.)

My granddad was a highway engineer - graduated from college in 1914. I heard lots of stories about what it was like to be a field engineer under those circumstances - and I don't want to return to those approaches.

"Means and methods" may be appropriate for landfill liner construction where liner permeability is a big concern. (See David Daniel's excellent papers (various publications since about 1980) for more on the construction of landfill liners.)

 

[SoilRocks]

Reading your question, two examples of a method spec come to mind - refuse compaction and subgrade stabilization. For refuse regrading, we specify lift thickness and number of passes with certain equipment - no testing. For subgrade stabilization (i.e. pumping soils), we typically specify the materials, thickness, and number of passes with certain equipment - again, we require stability, but usually don't test. For most other earthwork instances, testing for relative compaction is typically just a better way to document the constructed project.

 

[MRM]

Thank you for your responses.
Mark

 

[BigH]

In India, the MOST highway specification has an intersting twist. They specify the percent relative compaction to be achieved for a fill that is considered "suitable" by means of Atterbergs, gradations, etc. Then they also go on to specify a minimum MDD - such as 15.7kN/m3 for subgrade.

This causes some consternation in that a fine to medium sand (trace silt) avaible for subgrade fill has very good CBR values (>10) - but the MDD is only 15kN/m3. Indian engineers are very reluctant to use it since it doesn't meet the minimum. Way I see it is that it is sand, no organics, compacts great (neglecting upper 100mm after it dries, of course), has very good CBR value compared to minimum - so am I going to use marginal clay fill with CBR 6, high LL/PI just so I can get >15.7kN/m3??? Doubt it - I'd go with the sand notwithstanding the "low" MDD.

Has anyone else run into such minimum MDD values specified?

Regards

 

[dirtman]

For filter sands we have found a good correlation exists between a one point D698A test performed on a sample of oven dry sand to the value about equal to 70 percent relative denstiy. It is much easier to perform the one point dry Proctor test than the vibrating table test.

I wrote an article published in the October, 1996 issue of ASCE geotechnical magazine on this subject (McCook)