Differences in methods for SPT correlation to relative denstiy

[MRM]

Hi,
I'm interested in getting to the bottom of a question that's been on my mind. I know we have some very knowledgeable participants, and I hope someone has once wondered the same thing and now has the (an) answer!

The most well known correlation for taking SPT blow counts to a relative density estimate is the ol' Holtz and Gibbs correlation. I've reviewed and studied the original paper regarding this correlation. Another well-known correlation is one developed by Bazarra. The first thing I noticed when I was first introduced to these correlations, was that the Bazarra correlation yields a much lower relative density estimate for a given overburden pressure and N count than the Holtz and Gibbs correlation. This is, of course, with all variables being equal (particle size, shape, OCR, aging, etc.), and without worrying about correcting N values, GWL effects, particle dilatancy and compression effects, and so forth.

I've also read a famous paper by Lacroix and Horn that suggests that the Bazarra method probably gives the best results when dealing with a highly oc engineered fill and the Holtz and Gibbs method is more applicable to normally consolidated, natural sand deposits. This seemed to make sense and explain the difference to some extent. However, I can't ignore the horz/vert stress ratio (k) attenuation with depth that occurs in an engineered fill. In a deep fill, the lower lifts revert back to k=0.5 +/-, while the upper layers may have k values of 2 or more. I'm wondering if anyone uses the Bazarra method in the upper 10 -15 feet or so of an engineered fill to estimate relative density, and then switches to the Holtz and Gibbs method when the k values have been reduced to near the original value of approximately 0.5. The Bazarra method would almost always give a conservative estimate of relative density, so perhaps that’s another way of thinking of it. I would be interested in knowing which method is more accurate for certain deposits, though.

Also, I'm already familiar with all the caveats of using estimated values of relative density determined from the SPT, so that may help simplify the responses and keep the focus to the discussion at hand.

Thanks for your comments!

 

[MRM]

I'd like to rephrase the question;

Does anyone use the Gibbs and Holtz charts to estimate in-place relative density for natural sand deposits and/or deep engineered (controlled) fills? If so, do you think the answer you get is good for each?

Thanks for your thoughts!

 

[Longhorn2]

I have seen the correlations by Skepton, Meyerhof, Marcuson & Bieganousky, Gibbs & Holtz, and Skepton & Jamiolkowski, but I dont recall one by Bazarra, can you tell me what it is? Did he create a chart solution or an equation?

 

[Focht3]

Personally, I use both the Marcuson & Bieganousky and Gibbs & Holtz approaches, then apply engineering judgement. And I tend to lean toward the Marcuson & Bieganousky method results. But I'm usually dealing with fine to medium natural sand deposits, not man-made fills.