Why 95% Compaction 11

[1967pradyot]

Hi

Why for any compaction of soil, it is stated that 95% of Modified Proctor or Standard Proctor dry density is to be obtained ?
If we specify 85% or 90% and do obtain the required strength, whether it is OK? I want to know what exactly the justification of specifying that specfic ( generally 95%)degree of compaction.

 

[Rjeffery]

The short answer is that compaction is a way of trying to produce a uniformly dense mass of earth. it produces strength as a byproduct of the densification. If the mass of earth us uniformly compact then any settlement will be uniform as well (we hope). by compacting to a lower percentage the settelment will be greater. conversly, by compacting to a higher percentage the settlement will be less. 95% is chosen, I suspect, as a matter of engineering economy.

 

[MRM]

The "95%" spec (usually specified when speaking in terms of the modified Proctor test) is somewhat arbitrary. It has been proven to work in most common situations where an engineered backfill is required and most contractors expect that particular relative compaction in the spec so there is less chance that a required relative compaction in a spec will be confused.

In addition, 95% of the maximum Modified Proctor value usually corresponds to a relative density between about 65% and 75% depending on soil gradation, particle shape, angularity, etc. A soil with a relative density in this range is considered "dense" by most geotechnical engineers. Typically, a dense soil has good engineering properties under most applications; beneath foundations or behind retaining walls, for example.

 

[BigH]

MRM - yes, 95% is typically done in that way - still, would 94% be okay?? When constructing, people get bent all out of shape if you don't reach the 95% value. Here in India, MOST requires 95% of the MDD of Heavy Tamping (similar to the Hernia Test - oops, the Modified Proctor) for embankments. Why? I don't know. I've used 95% standard with no problems elsewhere. Some people just like things to be "heavier" - the bigger the better. I believe too that the 95% was put out by someone years ago and 'ell, it's here to stay since engineers just crop specs from one site to another - for the most part. There was a fine paper written in Ground Engineering on the "95% Fixation" - I have a scanned copy I would be happy to send out (if my serer will allow the size - but I can send page by page). If anyone is interested - bohica@indiatimes.com .

I once had an argument with a junior engineer in our company. He insisted on specifying 100% Standard Proctor (light tamping, roughly). The fill was to support some lightly loaded to moderately loaded foundations using good engineered fill. I was leaning on him to accept 95% Modified Proctor. He argued that "most" of the soils labs did standard as a matter of course (and the modified was, like I indicated above - a hernia test). I countered that the use of "modified" - at least to the contractors in our area - tells the contractor that more care is needed in the fill placement and compaction. I won as I was senior and had to sign off on the report - but this shows that there are good arguments on one side or the other. It is a matter of experience, personal prejudice, etc.

A guy named Moynihan - a prof at a school in NJ - wrote a damn good book on compaction. I suggest you get it.

 

[jheidt2543]

Back to the original question, why 95% vs other percentages.
I have seen a number of specifications with various % compaction specificed for the same job, depending on where in or around the building the fill was placed. In the "yard" 85%, exterior side of the footings 90%, under footings and slabs-on-grade 95%, pipe trenches under paving 95%.

We must keep in mind that compaction effort is directly related to cost. So, require the amount of compaction necessary for the area of use, but don't over specify, it just adds costs.

 

[MRM]

Great points gentlemen! When dealing with compaction testing and acceptance criteria, you really need to keep the idea of the statistical density of a compacted fill based on the testing done. That is, if you get some higher than the specified relative compaction (or relative density in some cases) results and some lower, use of engineering judgement is of great importance when accepting the work completed. An observational approach coupled with the testing also helps to give a good overall "feeling" that the engineer can use to accept or reject it as well.

jheidt2543 had another important point too regarding the economy of the compaction work-you're wasting money if you try to get 95% relative compaction in landscaping areas-specify a lower relative compaction if the situation permits!

BigH, I'd like to contact you regarding the paper you're talking about. Even after reading numerous papers on the subject of compaction, there's always something else you can pick up from reading a paper from a new perspective.

As a side note for those who may not know, a very good collection of papers regarding the determination of maximum densities using various methods, field measurement of fill density, and the nuances of estimating strength and settlement properties from the relative density or relative compaction measurements is the STP collection "Evaluation of Relative Density and its Role in Geotechnical Projects Involving Cohesionless Soil" (ASTM STP 523, 1973). As the name implies, it deals mainly with granular soils. This is a very well known collection of papers and I've always thought that it should be required reading for every geotech student in a graduate level program or for any practicing geotech engineer involved with signing off on, or specifying compaction work.

 

[BigH]

Just send your email addresses to the one I gave and I will send out - will start in couple of days. Moynihan's (maybe spelling is wrong) should be required reading, too.

 

[Focht3]

I think the fixation with 95% Hernia, er, Modified Proctor has a lot to do with where you started your engineering practice - and what kinds of materials were being placed. THS (The Hernia Standard) is common among those who cut their teeth in granular soils, while the standard Proctor is more common among those who started out in clay environments - especially where expansive movements are a real problem. (Overcompacting a CH clay is a quick way to damage a building.) BigH started out in Ohio, then Canada - so he has a THS fixation. I started out in Houston - a clay environment - so standard Proctor makes more sense to me.

I think that some engineers get too hung up on the numbers - sometimes I will accept 92% standard Proctor because I'm not concerned about settlement. But I guess my relaxed attitude - on some projects - has a lot to do with my experience. Lacking that experience, I would probably do it "by the numbers."

BigH - please send me a copy of the paper also, at your convenience.

 

[cvg]

I think MRM hit it on the head. The specs are written for "tests" which are done on a limited basis to "control" and "assure" quality. Statistically, it could be shown that by requiring 95%, that even though you will have areas which are below 95%, they should be within 1 standard deviation of the mean and that should be good enough to prevent failure of the pavement or whatever is built on the compacted soil. By looking at the cost of the improvement, the design life and the risk associated with failure (including lawsuits for the poor geotech), the Engineer must determine what risk he can tolerate. Less risk for landscaped areas - 85%, more risk for roads and building foundations - 95%, extreme risk for dams and nuclear power plants - 98% or 100% compaction.

 

[Focht3]

I've "run the numbers" on some big jobs - and don't think too much of the statistical argument. A lot of the variability has to do with natural variability of the soil - not compaction quality per se.

Specifying 95% of whatever standard is usually done "cuz" - that is, "- 'cuz that's the way we've always done it."

 

[BigH]

Here in India - you specify 95% compaction - as Praydot undoubtedly has ingrained in his head. Then, for acceptance of a layer, you use the following:


gamma(average) > gamma(spec) + (1.65 - 1.65/sqrtN)*stdev

I tried to get them to change to %Compaction using same formula but they won't. So you say 95% compaction - then you go and take 95% of it to get the gamma(spec).

So - if your stdev is large (say you get a range of values from 94% to 101% - you can see that the right side gets big - also as more tests are done, it gets big. Go figure.

Just some info for you all not working here - things are different, for sure.

 

[Focht3]

I agree that percent compaction should be used - guess that's what happen when the decision makers don't understand what they are specifying -

The statistical approach makes sense provided the value of gamma(spec) is chosen as a minimum based on some realistic criteria - like a settlement requirement. But it seems to be chosen mostly due to the "cuz" factor. So much for real engineering -

 

[BabbageGeo]

There is another issue here that does not appear to have been considered; that is the moisture content of the soil at the time of compaction and its relationship to the M/C at max dry density for either std or modified Proctor Compaction. For low moisture content soils, particularly those dry of the optimum m/c or below the m/c at max dry density, dropping below 90% of the max dry density may result in unacceptable settlement and or "collapse" on wetting up. I suggest you check out the ASCE Geotech Journal Vol 118, No.9 1992, pp1376-1395, "Review of Wetting-Induced Collapse in Compacted Soil" by Evert, Lawton, Richard, Fragaszy and Hetherington. The paper includes some good references from experience in South Africa as well as the US.

In my experience in typically clay soils in NZ where the soils are generally wet of OMC, using modified proctor as a means of compaction control means burning a lot of diesel drying soil out then compacting it to achieve very little additional benefit. So std compaction is generally used as the reference, except where low plasticity silts or granular materials are compacted. Then modified or "Heavy Compaction" is the std. Given the energy req'd to achieve modified Proctor is more than 4 times that used to achieve std compaction there is usually a significant cost penalty involved in using Heavy Compaction + the risk of setting up unacceptable construction induced pore pressures if the soils are wet of OMC. The relationship between air voids and density also needs to be considered. If water needs to be added to reduce teh energy req'd to ensure air voids at std compaction are say always less than 10%, and water is not freely available or is expensive, or the material will rapidly loose strength with the addition of water then reducing the air vodis through the application of "Heavy Compaction" may be the best option.

It is rare that there is not an area somewhere within a fill that does not entirely meet the specification but has passed undetected. The whole point of testing isthat it is a random statistically based approach to get to 95% or 99% compliance or whatever you chose to try and enforce. Testing is genearlly carried out to confirm a std higher than required if all fill is unconditionally guaranted to meet a specification to allow for "the one that got away", for variations in soil properties, and post compaction performance under cyclic and or slow increases in moisture content. The spec should match the post compaction long term load deformation criteria necessary for its intended end use.

 

[cvg]

collapsible soils are indeed a problem here (Arizona). not so much for compacted fills, but for foundations. We have many dams, levees, dikes and other structures founded on collapsible holocene deposits. It is theorized that upon wetting, the holocene soils may collapse, causing subsidence of the dam and potentially a dam failure. The solution is to remove the soils and replace them with proper compaction at OMC.

 

[Focht3]

[blue]BabbageGeo[/blue]:

Welcome to [blue]Eng-Tips[/blue]!

Collapsible soils (natural and man-made) are a potential problem; they have been discussed in other threads -

[blue]BabbageGeo[/blue] wrote, The spec should match the post compaction long term load deformation criteria necessary for its intended end use. I agree. I think that sentence summarized what many of us have been saying in this thread - don't blindly follow a standard without understanding what you are trying to accomplish.



Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[baobab]

Is there an empirical relationship between the various density specifications for soils - Proctor (Standard and Modified) and Mod AASHTO ?

 

[Focht3]

A general empirical relationship? No. But you can develop one for a particular soil from a given source. Of course, if the material changes, the relationship also changes...



Please see FAQ731-376 by [blue]VPL[/blue] for tips on how to make the best use of Eng-Tips Fora.

 

[BigH]

I agree that relationships could/should be developed for your own area soils - then you can be more in the comfort zone - but as Focht3 points out, I grew up on imported granular (crushed limestone) in Ontario and, well - we used to use 97% Modified = 100% Standard as a decent approximation. (93% = 95%).

 

[solutioninc]

This string started with a question on percent compaction for STRENGTH. I think that is why the aspect of compacting clayey soils to achieve a desired low permeability (hydraulic conductivity to be more precise) was not raised.

Whether it be for the required strength or the required hydraulic conductivity, the proof is in the eating of the pudding. We have to test samples from the compacted layers to verify that the required property is achieved. Since this is more expensive than monitoring the density a percent compaction for achieving a uniform layer has been used for a long time.

As we have different optimum moisture content and maximum density for the two Proctor compaction methods, there is a different OMC and max density FOR THE COMPACTION EQUIPMENT USED IN THE FIELD. Generally these values lie in between the two proctor values. But strength requirements dictate compaction dry of optimum and hydraulic conductivity requirements dictate compacting wet of optimum, use of modified for the former and standard for the latter are safe recommendations. This is because the OMC for the standard test is most likely to be larger than the OMC for the compactor used in the field and specifying moisture content during compaction to be more than the OMC from the test we will be okay. The OMC from modified test is most likely to be smaller than for the field equipment and so using the modified standard will almost guarantee compaction dry of optimum in the field.

Regarding the percent compaction one can use either test provided the resulting density is what is required to give the required engineering property. 100 percent standard could be equal to 90 percent modified and so the resulting density in the field will be the same irrespective of the method used. A series of laboratory tests before writing the specification is the best answer.

 

[Focht3]

Hmmm,

While everything that [blue]solutioninc[/blue] is correct, I wonder about the wisdom of focusing on strength as the only "necessary" criteria. Frankly, I have not encountered a design problem where the strength of a compacted fill was the only design/construction consideration, and the question of post-construction settlement (or swell) was not considered at all. In my experience, "adequate strength" is pretty easy to achieve in clay soils (LL > 30, PI > 20), but I recognize that others may have had other experiences -

I, for one, was not focused at all on permeability. In fact, I did a word search on this forum and it appears [blue]solutioninc[/blue] was the first to raise that issue. In my view, the majority of posters have been clearly focused on the issues of modified vs standard compactive effort, and long term deformation of the fill body. I think we have pretty much stayed on message during this thread. (Sometimes the discussion wanders a good bit!)



Please see FAQ731-376 by [blue]VPL[/blue] for tips on how to make the best use of Eng-Tips Fora.