Soil Stiffness versus Proctor-Any correlation method? 6

[IJR]

Pals

I have designed an important slab on grade(heavy warehouse loading) and as usual have specified a compacted base to 95% proctor.

The geotech has surprised me by submitting to me values of The Elastic Modulus of Soil,Es, as measured after compaction of some layers of the base.

Some "googling" has confirmed that The Modulus method is now favored to the old density(proctor) measurements, being simple to use and producing more meaningful data. But I have ended up with no more practical info.


Any correlation between Es and Proctor?. Am I supposed to use Es to find deflection for the thickness of the base compacted? Any formula for that?Should I see texts on soil mechanics?


Thank you in advance

respects
IJR

 

[Ron]

IJR...not sure what you Googled, but the equivalent parameter for a pavement slab (warehouse slab on grade included) would be a resilient modulus value; however, this is not an easily correlatable parameter to be used for quality and uniformity in the field. It's great if you are using a falling weight deflectometer to determine a load-deflection curve for a pavement, but generally has limited value for run-of-the-mill construction.

The resilient modulus can be correlated to a CBR value (Mr=1500xCBR), but that give you a stability parameter not a compaction parameter. You can achieve compaction without achieving stability; however, you must achieve a certain level of compaction to achieve the DESIRED stability.

Stick with the Proctor and have the geotech give you a correlation curve for the two if he has done any field testing or lab testing. He should be able to give you a modulus vs. compaction curve that you can use for field QA work....Proctor curve still needs to be done.

If you want to use the modulus value, you can use it in an elastic layer analysis of the pavement. You can find a reasonable treatment of this analysis in Pavement Analysis and Design by Yang H. Huang or Principles of Pavement Design by Yoder and Witczak.

 

[WHEngineer]

IJR, did the Geotech provide comparative modulus values with varying compaction levels, or were the modulus values similar and represented typical field conditions? I assume the Geotech provided the elastic modulus values for use in settlement analysis in the building design. For instance, using the strain-influence method (Schmertmann & Hartman, 1978), you can determine the immediate settlements (in sands) and verify that the elastic modulus values meet the required settlement criteria. Typically, as a Geotech, I am used to the structural engineer providing the maximum settlement criteria and I will perform the settlement analysis during exploration and testing. Then, I can provide options to the structural for meeting the objectives during the design phase. Perhaps the Geotech felt that the settlement analysis was outside his scope of work in this project. If the soils compacted to a specified level (95%) meet the performance criteria, use of compaction testing is still likely the easiest testing methodology.

 

[kslee1000]

I am kind of lost here. Which stage you are in - design, or construction? What you asked and expected your geotech to provide? Answers may clear up the confusions.

 

[BigH]

kslee1000 - you hit what I was thinking before but didn't have time to comment. The compaction, to me, referenced by "Proctor" is a construction tool used to ensure that a material has been placed in a competent fashion. The Es or, as Ron indicated, the Mr is a design aid. Proctor has little meaning to the slab stiffness design if the base layer that is compacted is only a few hundred millimetres in thickness - or if he is talking about "improving" the subgrade to that level of compaction - whereas compaction would only extend 0.5 m effective depth (assuming he isn't compacting clay). For the mat stiffness design, Es would be used (or the more, I still believe, more commonly the modulus of subgrade reaction). You would use the Proctor only for compaction control.

 

[IJR]

kslee

I am in the construction stage. What I usually do, is do thickness design for the slab using specs like ACI360, assume a reasonable Soil springs and ensure that I will have that spring by specifying a good base and compaction level.

Then during construction I have a geotech do lab work on the compacted base, and see if any improvement is necessary. Usually none. But then I always have had my proctor value.

respects
ijr

 

[BigH]

But how thick is your compacted base? and what real effect does it have on the foundation system stiffness?

 

[Ron]

IJR...I see nothing wrong with your approach...your Geotech seems to be confusing the issue.

 

[fattdad]

+1 on the lack of correlation between relative compaction and soil modulus (or modulus of subgrade reaction).

Here's the way I do this in practice.

Do a soil exploration
Get representative bulk samples
Classify the soil.
Do a Proctor (i.e., moisture-density relations)
Prepare a sample for CBR testing (targeting 95 percent relative compaction)
Obtain the 4-day soaked CBR value.
Correlate CBR to modulus of subgrade reaction.
Provide a design subgrade reaction value for the slab-on-grade design.
Provide enhancement to the design subgrade reaction value for varying thicknesses of subbase material (per ACI 360).


During construction, for cut areas, we would proofroll the native soil to make sure that the subgrade did not have pockets of disturbance. For fill areas, we'd have proper documentation that the contractor acheived the requisite compaction - but I'd also proofroll to make sure that surface areas of disturbance were removed, if present
Document the placement and compaction of the subbase stone.
Test the concrete being delivered.

Just my cookbook explanation of thoughts. . .

f-d

¡papá gordo ain’t no madre flaca!

 

[kslee1000]

IJR:

Please correct me if I am wrong.
From your response, I sensed that the slab was designed utilizing a specific/preliminary Ks. As in the construction, you somehow asked your geotech to verify the onsite modulus after compacted to 95% proctor. Thus your geotech submitted the freshly measured Ks for comparasion. Is this the case?

 

[iandig]

Please also correct me if I am wrong, ks is not the same as Es. As mentioned by me in a number of previous posts, in the UK we are starting to use Dynamic Plate Tests in place of more traditional CBR testing for pavements, and this does return a Es value. Based on the relationships in the appropriate DMRB, you can formulate the relationship between Es and ks but this is so tentative I am note sure how accurate it would be and have not done it [yet]. The relevant DMRB is under IAN 73/06, [download from standardsforhighways dot co dot uk under IAN] and this does relate CBR to Es, and also includes the determination of CBR from plate tests using ks.
As for the relationship between 95% proctor and Es, there may well be one but I would expect it to be different for different soils, in the same way as MC vs CBR and Density vs CBR will be different. In general for clay soils I tend to find 95% of the 2.5 kg rammer will give between 2 and 3% CBR on the wet side, and around 5% for 100% of the 2.5kg MDD. This is a rule of thumb which is useful for assessing results from the lab but needs to be proved for each material, its sometimes more but when assumed it always seems to turn out less [the 'butter side down rule']. Where there is an increase in the granular content, this changes significantly.

 

[IJR]

kslee

Not that way

I dont really assume spring stiffness. I base spring stiffness on the subbase, ie the layers underlying the compacted layers.

When a the compacted layers are brought up to densities that match 95% Proctor, that ensures that the compacted layers are very strong and you do not need to worry about. It is only when they are poorly compacted that you worry on that.

To share with you my little experience, I once placed a huge stadium stand on a thick backfill(10ft or 3metres), and specified both the material and the proctor number, and they did it exactly like that. No problems have so far been reported.

This is the practical procedure I usually use.

respects

ijr

 

[kslee1000]

IJR:

Thanks for the clarification and the "share"
I am still puzzled why the geotech provided you the Ks rather than the proctor you would require, and was looking for. (I kind of playing doctor here

Are you sure the Ks was for your site, it colud belongs to others, submitted by mistake.

 

[fattdad]

Oh my! This is getting confusing!

A silt with find sand (i.e., ML) may be properly compacted to 95 percent compaction, but have a CBR of 3, and corresponding k-value of 100 psi/in. A fine to medium sandy lean clay (i.e., CL) at 95 percent compaction may have a CBR value of 10 and a corresponding k-value of 200 psi/in. If placed on a 4-in thick layer of dense-graded aggregate you could use a design k-value of 110 for the former and 220 for the latter.

The use of soil modulus in industrial slab design is a separate topic. If you have an industrial slab that's 40,000 sf and has a big racking system and designed for 500 psf, you need to use the k-values to design the structural interaction between the rack, slab and subbase. Soil modulus will foretell whether you will get settlement from the deep-seated layers. Let's say you have a soil layer of soft clay that's slightly overconsolidated and not saturated (like maybe some residual soil). Your new areal load may well extend into this soft clay layer and as a result you may realize several inches of settlement at the center of the building versus the corners or edges. You need to understand soil modulus for this case.

For my practice, I'd look at the soil logs to identify whether this is a relavent concern, before I'd even worry about the k-value. After all who cares if the interaction between the rack legs, the concrete slab and the subbase are good if the entire building settles 4 inches in the middle and 1 inch at the corner?

ACI 360 has some folks that are advocating some "long-term" k-value to address what is really properly described by soil modulus and it is infuriating! We already have a method to look at soil compression - it's called soil modulus and load attenuation. Some hybrid "k" value is misdirection!

End of rant.

f-d

¡papá gordo ain’t no madre flaca!

 

[kslee1000]

Now I have to admit that I was the guilty party, who inadently changed Es (as in the original post) to Ks. Sorry to cause this confusion.

 

[JungleBungalow]

So, lets clarify:
Es = Soil elastic modulus
Mr = Soil resilient modulus
= 1500*CBR
These are not the same, soil resilient modulus is a measure of the soil stiffness under rapidly applied loads (like traffic loading). Es is the value used in settlement analyses, not Mr.

ks = Modulus of subgrade reaction, for which I have seen relationships based on CBR values.

In any case, the influenced soil for a mat foundation (roughly 2*width of mat) will extend much further beneath the upper lifts of compacted fill.

Hope that helps.

 

[JungleBungalow]

oops, I spoke too soon...fattdad spells it out quite nicely.