interpreting blow count to find friction angle

[cepe01]

Hi. I have a very simple and maybe dumb question, but how do you interpret blow count N-value when there is a wide range in the blow count?

For example I am reading a boring report with N-values range from 18 to 50 blows which ranges from medium to dense. Can I just take the average and use N=34 to find my angle of internal friction?

 

[racookpe1978]

No, because the soil and rock may vary across the site.

Even "IF" you knew expected that the soil was perfectly uniform across the site, an average still would not make sense, since it varies so much from hole to hole.

 

[Mad Mike]

You have layers of soil beneath your site, some being more dense than others. You haven't mentioned the soil composition but based on your density descriptions I assume it is mostly sands.

You've given no indication of what you will be using the friction angles for, but under most circumstances it would not be sensible to just assume an average- for example if you are looking at embankment stability, the low strength layers might be the controlling factor.

Even estimating friction angles from SPT N-values is risky unless you know the soils very well...I would only be confident in this correlation with relatively sandy materials. Perhaps ask your geotech. for his opinion.

The question is not at all dumb, but requires further information if you're wanting an intelligent answer.

Cheers,
Mike

 

[killswitchengage]

There is what we call probabilistic calculation in geotechnical engineering. If you have the same soil type having different parameters you should compute the average value, the standard deviation and the average deviations. After that you need to pick a value that verify a certain condition but sadly i have never done such a job since its not required in my country so i want to point you toward the Eurocode 7 or any book in fact .

 

[oldestguy]

Assuming the blow count gives an approximate indication of relative density for granular and consistency of cohesive soil, the conservative approach would be to err on the low side with how you use the different numbers. One reason for making several test borings and other soil tests is to give you some indication of the variations from place to place and therefore gives you some basis for making judgements as to how to use the data. In other words one boring on building site may or may not be suitable, so we generally scatter them across the site and maybe do more if needed to clarify where to go with making judgements.

 

[Okiryu]

Based on your N-values and your description of "medium to dense", I guess you are talking about granular soils. So, I assume that you are trying to get the phi value to get the bearing capacity factors for calculating the bearing capacity of shallow foundations. There are several correlations between SPT N-value vs. phi.

If this is the case, I would try to have an idea of the preliminary size of the footings then just use the N-values within the zone of influence of the footings (2B of independent and 4B to 6B for continuous footings). Then, I would take the average N-value within the zone of influence of the footing minus one standard deviation (to account for soil variability). You can use two standard deviations but I think that it is too conservative (at the end you may be applying a FOS of 3 in your results)....

You can see here for the average minus the standard deviation approach (just statistics stuff):

https://en.wikipedia.org/wiki/68–95–99.7_rule

 

[oldestguy]

Here is a chart from Peck and others.

 

[fattdad]

If you have some funny yellow dirt and the blow counts range from 10 to 20, sure you can take an average N-value for the funny yellow dirt and derive some character of that layer. You may have multiple layers and you can do that for those layers too!

As you look from boring to boring you may have some average for the yellow layer that ranges from 12 to 17. Some may advise you to take the lowest average N-value and use that site wide.

Hope that helps a bit?

f-d

ípapß gordo ainÆt no madre flaca!

 

[Okiryu]

You did not mention the depths for your N-values. So, perhaps you can also consider for checking the phi values based on the effective stresses and N60 values. Then use the proper phi for the depth of influence of your foundations. See attached correlation chart from Coduto's book.