SPT number and soil density relation ?? 2

[elvercho]

I would like to know the relation between SPT number and soil density.

Many thanks for your help.

 

[LHA]

SPT indicates Dr and angle of repose in sand. Generally, the higher the SPT the more dense and higher the angle.

I don't think there is a 1:1 correlation (with a formula to calc it), though. I've only seen ranges: SPT = 10 to 30, is typical sand, angle = 30 to 40.

 

[BigHarvey]

There is an article by H.J. GIBBS and W.G HOLTZ which title is "Research on determining the density of sands by Spoon Penetration Testing" in the proceedings of the 4th ICSMFE, vol 1, pp35-39 which is a refernce text for the subject. It will give you the Relative Density of a sand as a function of N ( SPT ) and the overburden pressure

 

[PEinc]

Most geotech or soil mechanics books have a table that relates blow counts to density. See Jeseph E. Bowles' Foundation Analysis and Design, 4th Edition, Page 141, Table 3-4, and Page 143, Table 3-5. His Section 3-8 is titled SPT Correlations.

 

[BigH]

We've had more than a thread or two in the geotechnical forums that have shown this. I suggest that you check out Terzaghi and Peck (now Terzaghi, Peck and Mesri) - also Fang's Foundation Engineering Handbook has information in it. Remember that SPT relates by approximate correlation to relative density in cohesionless soils - the density, which is what you asked, there is, to my knowledge, no relationship. Just a point - In most of the world - 10 to 30 blows is designated as "Medium" or "Medium Dense". In Canada, it is known as "Compact".

 

[Chewi00]

Ive only seen general correlation relating blow counts to density in granular material. The Burgmeister Classification system relates
0-10 Loose
10-30 Medium Dense
30-50 Dense
>50 Very Dense

I dont know if you could make any closer corellation than that without seeing a grain size curve or doing a relative density test.

 

[moe333]

You should also be aware that before you correlate the field N value to friction angle or description of relative density, the field blow count should be corrected for energy, rod length, boring diameter, as well as overburden pressure. Most geotech. text's have the various correction factors.

 

[BigH]

moe333 - unless the correlations were developed without all the fancy correction factors - when did the correction factors start to be brought in to the practice? When did Terzaghi first propose a correlation (that is still typically used) for the SPT N value and friction angle? Likely before. In recent years, much more emphasis is being put on corrections, etc. but unless you revise the correlation charts that were developed prior to such practice - what is the relevance? And of course, there is questions on whether you make a correction for groundwater levels, etc. I find in practice that most just don't correct all that much except for seismic. Better use of the N60 concept and new correlations (see Terzaghi, Peck and Mesri, Paul Mayne and others) is gaining a strong foothold - but read SPT N values (at least prior to the 1990s) as without. Just a philosophical thought . . .

 

[Ron]

BigH...I agree that most do not apply corrections to the SPT from the field...it is what it is! Schmertman did a lot of research on this in the 70's, but corrections as a result have never caught on in the commercial realm.

It's interesting that with all the research done, all the fancy equipment we have today, and all the brainpower that has gone into geotechnical engineering, we still use the concepts of the pioneers (Terzaghi, Peck, Casagrande, et al).

While the SPT is related to relative density, there has been much correlation done with the small dynamic cone penetrometer (Sowers Method) and in-place density. This should be done as a site-specific correlation due to the variable nature of soils, and really only works well in the upper 5 to 7 feet because of manual equipment handling.

 

[moe333]

Big H, Ron, I agree that the more minor corrections such as for borehole diameter and rod length may not be significant, but corrections for energy and overburden can be significant. The old correlations were based on cathead energy (about 60%) whereas many drill rigs now have auto trip hammers with energy at about 90%. I've looked at friction angle correlations between CPT and SPT at the same boring locations, and they are generally much closer only after you apply the energy and overburden calculations to the SPT N values. Of course there are many other things that can lead to erroneous SPT N values.

 

[Ron]

moe333...that is exactly the research that Schmertman was doing in the 70's. See if you can get a copy of his paper, co-authored by Smith, probably published in ASCE in 1978 or 1979.

 

[BigH]

moe333 - what you say is true and I alluded to it with the "new" correlations. Sure, you can "modify" the previous correlations for energy assuming that they didn't make any corrections themselves in either energy or others and that the cathead energy of 60% was/is "standard" but I've seen drillers use 2.5 wraps rather than the specified 1.5 wrap, etc. and different donut hammers. If, in practice, the industry is now correcting, then new correlations need to be developed based on a "set" Nenergy, say N60 or N70 or N55 (all of which are used extensively in the literature). SPT is a old tool (that I like) but the CPT is a better way to go - with some way to obtain samples at a site for correlation purposes.

 

[moe333]

Ron, the Schmertman correlation is less conservative than the one by Peck-Hanson-Thornburn for peak friction in sands. I have found that by averaging the two correlations, after correcting for energy and overburden, the results are relatively similar (but still slightly conservative) as compared to the CPT correlation.