Smart questions
Smart answers
Smart people
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Member Login




Remember Me
Forgot Password?
Join Us!

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips now!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

Join Eng-Tips
*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.
Jobs from Indeed

Link To This Forum!

Partner Button
Add Stickiness To Your Site By Linking To This Professionally Managed Technical Forum.
Just copy and paste the
code below into your site.

pelelo (Geotechnical) (OP)
16 Dec 10 19:52
Hello engineers,

I am working on the geotecnical report for the foundation of tower.

I use sigma W in order to estimate inmediate settlements.

As you may know, you need to enter the Modulus of Elasticity values (E) and poisson, for each soil layer.

I have some Modulus of elasticity values according to the SPT-N values. From 0-4, 5-10, 11-30, 31-50, 51-100, >100. It doesn't matter if it is a cohesive or cohesionless soil.

I am not sure if this is ok. I would like to confirm with any of you with some experience what are the best / most useful Modulus of elasticity values geotechnical engineers use nowadays.

Thanks
Ron (Structural)
16 Dec 10 22:41
I'm confused by your question.  What are the actual modulus values you are getting?  Why not run cyclic triaxial tests to get an actual modulus value?

I would expect the following:

silt and clay....5 to 7 ksi
sand...10 to 15 ksi
clayey sand...15 to 20 ksi

 
BigHarvey (Geotechnical)
17 Dec 10 4:11
Obviously the same N value will give you different values for E in granular soils and in cohesive materials.

This being said, the best in-situ method to get deformation moduli is the pressuremeter because it is the only test where you have a pseudo-elastic loading phase.

CPT in sands will allow you to get reasonable values as well but not in clays.

Correlations between E ( pressuremeter modulus ) and N is as follows :
N/E for clays : 0.8 to 1.1
    for silts : 3
    for sands : 2 to 6
You will obtain the deformation modulus from the pressuremeter modulus by multiplyng it by 3 for sands, 2 for silts and 1.5 for clays.
Just be careful with correlations : you will only get orders of magnitude. Do the proper testing to get the right input data.  
pelelo (Geotechnical) (OP)
17 Dec 10 7:31
Thanks Ron and Bigharvey,

Unfortunatly we don't have the appropiate instrumentation to obtain the modulus of elasticity.

I knew there was something wrong about our E values. I will follow your advise.

What would you guys expect for silty sands (sm) and silty gravel
(Gm) as the site I am working on is mainly composed of these soils.

Or is there any reference you recommend I should get?

Thanks again
 
Helpful Member!  BigH (Geotechnical)
17 Dec 10 7:43
I know that many have a problem with Bowles, but you can check out his Foundation Engineering book - 5th edition.  There is also Hunt's Geotechnical Site Investigation Handbook - with lots of correlations in it as well.  Correlations are fine for gaining perspective on the problem but nothing beats site specific data .  .  .
Helpful Member!  FixedEarth (Geotechnical)
18 Dec 10 2:15
Pelelo

I have always used E = 400 to 600 times the Su value for ohesive soils. According to Budhu's "Soil Mechanics and Foundations" book 2nd edition, page 605, cohesionless soils have the following E values (secant values):

Loose sand     = 400,000 psf
med-dense sand = 800,000 psf
dense sand     = 1,600,000 psf

Try to get your hands on a copy of "Manual on Estimating Soil Properties for Foundation Design", 1990 by Mayne and Kulhawy. It maybe online somewhere.

 
BigH (Geotechnical)
18 Dec 10 3:02
Fixed Earth - ranges of Es for cohesive soils I've seen are much greater than you have given . . .  Bowles (3rd Ed) shows for Ip>30 or organic Eu = 100 to 500x Su; for Ip<30 or stiff; Eu = 500 to 1500x Su.  Very stiff clays Eu = 1500 to 2000x Su.

You can find Paul's downloadable papers at:
http://geosystems.ce.gatech.edu/Faculty/Mayne/papers/index.html
 
FixedEarth (Geotechnical)
18 Dec 10 10:53
Thanks BigH.  
BigH (Geotechnical)
18 Dec 10 18:26
The correlations got me thinking - we all look towards the computer programs to give us exactness - yet, one sees, if one must use correlations - that settlement can vary by a factor of 3 or 4 just depending on which E one uses. . .  even when you can have some lab data, how accurate is the E?  even a difference of 20% on the lab value means a difference to the resulting settlement - or wall movement or . . . So, even with FEM, how "accurate" are we?
hokie66 (Structural)
18 Dec 10 20:43
Not very.  Must be frustrating for you guys, BigH.  Settlement estimations are more black art than science, I think.
Ron (Structural)
19 Dec 10 0:27
Ah yes...we measure with a micrometer, mark it with a crayon and cut it with an axe.
pelelo (Geotechnical) (OP)
19 Dec 10 7:45
I agree with all you,

Due to the lack of accuracy using correlations, FEM analysis could lead to different results.

This problem gets even more serious when based on the settlemnt values you have to provide recommendations. For example using my E values due to the high settlement results  I get, I could recommend any type of deep foundations, on the other hand if someone gets lower settlement  values he could recommend anything else or maybe nothing as the settlments might be less than the allowables.

At the end, all this means money to the client.

 
Helpful Member!  Mccoy (Geotechnical)
21 Dec 10 18:44
Don't you guys find that literature Nspt- Es correlations are usually over conservative (with the possible exception of the D'appollonia correlation for sands and gravels)?

Pelelo's soils would be suited to the AASHTO, 1996 correlation: E=0.4÷0.7n [MPa], but I find it way too conservative.

Lately I've been doing quite a few investigations on existing buildings, damaged by an earthquake, but in good conditions prior to the quake.

The above correlations woudl indicate likely settlements problems, whereas these were not observed.

I tried to compare results from geophysical surveys (Vs profiles), going from the dynamic E degraded to the secant E and sometimes it 'feels' better, even though apaprently under-conservative.

Right now I'm using Nspt-Gmax correlations, which afterwards I degrade to Gsecant and transform into Esecant knowing the poisson module .

It sounds convoluted, I know, but this way I get results which would appear to be more realistic.

That is, the soil under houses which had no settlement problems comes out pretty good and not pretty 'crappy'.

Your opinion about that?
Mccoy (Geotechnical)
21 Dec 10 18:48
Pelelo, one possible way out of your predicament might be a good software on Nspt analysis.

Some of'em have filters which allow you to apply soil-specific correlations, plus you have output histograms where you have the whole 'scenery', have a feel for the possible range and might eventually decide to take the median value of the parameter.
Helpful Member!  vulcanhammer (Geotechnical)
23 Dec 10 14:30
A list of elastic moduli for various soils can be found in Table 5-16 of this:

http://www.vulcanhammer.net/geotechnical/FHWA-NHI-06-088.pdf

There's also other information on elastic moduli for rock, etc., scattered in the text.

That having been said, they're still correlations, with the caveats that BigH mentioned.

http://www.pz27.net

Helpful Member!  dicksewerrat (Civil/Environmental)
6 Jan 11 13:49
You might want to look at ASTM D3839.

Richard A. Cornelius, P.E.
WWW.amlinereast.com

Helpful Member!  ntschwanz (Geotechnical)
6 Jan 11 15:37
Virginia Tech has a Center for Geotechnical Practice and Research (CGPR).  CGPR #44, May 2007, is a publication by Duncan and Bursey on Soil and Rock Modulus Correlations.  This is a nice resource.  I would caution you not to throw away common tools in lieu of FEM.  Instead use FEM in conjunction with other tools to help understand behavior and only use FEM to estimate displacement if you are quite comfortable with your soil properties or you assure yourself you are being conservative.  I have a couple of projects now where I have created models that experienced significant full scale loading of the prototypes.  From a predictive standpoint I got pretty close on maximum displacement on one project and overestimated displacement by about a factor of 2 on another.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!

Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close