DMT

[lovethecold]

Thanks to an earlier question, I now have heard of the flat Dilatometer Test. How many engineers use this in their day to day geotechnical work? The company I work for only does consulting work (ie, no design). Would there be sufficient reason that we would want to purchase a setup to run this test? In other words, would it provide significantly better data that others would be willing to pay for.

From what I could find doing a google search, it seems the DMT would be good for slope stability and settlement analysis. We live in an area with deep soft fat clays that are known to have slope stability problems down to as flat as 7:1. Or else we typically deal with medium to stiff glacial till (sandy lean clay).

Is there a maximum depth to which this can go? What is the maximum stiffness of soil which this is good for?

Thanks for your input.

 

[qhtony]

I honestly haven't used DMT before, and not seen any one using one in my city. Hardly heard about it as well....

 

[Mccoy]

I think a similar issue has been already discussed a couple of years ago, maybe you might run a search.

As to the purchasing it really depends, that kind of test is not so popular for various reasons, beginning with the price, twice as the electrical CPT where I live.

 

[Mccoy]

lovethecold,

I just remembered there has been a recent poll on the main forum of the Italian geotechnical engineers, I went back to it.

It turned out that in Italy, the mother country of the dilatometer, most of'em never used it.

Consensus was that such tests are complex,lenghtier and more expensive with respect to other tests with similar results (electrical CPT) and often contractors don't have the equipment to run'em.
The blade will often bend and the lateral membrane breakup.

That's clearly a test for specific jobs with an higher than average budget. A niche market.

In my opinion pros are the limited soil disturbance, good precision and displaying the features of both a penetrometer and a pressiometer...

 

[lovethecold]

Thanks for the reply Mccoy. I will shelve the idea of trying to get my boss to purchase one.

 

[palmahouse]

I like the DMT. I used it to estimate Young's modulus in sand for settlement analyses for large tanks on medium-dense sand (where settlement was the key issue). It allowed me to be less conservative with the estimated modulus.

I cannot, however, see how you would use DMT data for slope stability analyses, where strenth is the issue. DMT's basically give you a modulus. For soft clays and stability analyses, I sould stick with CPT and rotary-wash borings and traixial testing.

I would leave it up to the subcontractors to do the DMTs anyway. The CPT contractors (like Gregg In-Situ) do it well.

Depth depends on soil conditions. In soft clays, you can probably go 100 feet. One issue in any soil, if you hit gravels, it rips the diaphram on the tool.

 

[casimmons]

The DMT is a valuable tool, depending on the soil conditions. It works well in sands, silts, and clays, as long as they aren't too dense/hard to push the blade. It is no good for rocky or gravelly materials. The DMT gives modulus info, and undrained shear strength. I like it better than the pressuremeter because you get much more data in the same amount of time. It is also a very simple test, and the data reduction is very easy. The pressuremeter requires more skill in running and reducing/interpreting the data.

 

[Mccoy]

I like the DMT. I used it to estimate Young's modulus in sand for settlement analyses for large tanks on medium-dense sand (where settlement was the key issue). It allowed me to be less conservative with the estimated modulus.

That's a case where the DMT is very valuable indeed. You have a clue on OCR of sands and can apply better correlation than the usual E=2.5 qc, overconservative in OC sands...

Only geological/sedimentological judgment can be a (qualitative) substitute for it

 

[Mccoy]

Sorry I quoted casimmons instead of palmahouse.

Drawback of these boards is you don't have an edit tool.

Now quoting casimmons, the DTM is not such a simple test, sure so compared to the pressiometer, much less so compared to the static penetrometer (with the possible exception of pore pressure dissipation readings)

 

[palmahouse]

Mccoy,
Thank you for your insights and experience. I was informmed that for a similar tank project with similar soil conditions (also in our region, intollerable differential settlements were estimated and significant ground remediation efforts were implemented. They did not use DMT, PMT, or CPT - just SPT My work, which included SPT, CPT, and DMT, showed that we can support the tanks on conventional footings with little remediation and that settlements should be tollerable (we are dealing with old alluvium, and the data shows high OCR and modulus - much higher that 2.5 qc and 5N. We will be monitoring settlements using surveying during construction and tank filling as a backup - but I am confident we will be o.k. If anything, I was probably conservative too - we will see.

 

[Mccoy]

That's a terrific way to go, palmahouse, diversified in situ testing plus feedback on the settlements analysis.

A little more $$ into the geotechnical side, a lot less $$ into remediation and RC...

Wish more contractors would reason like that

 

[BigH]

Question to you lads (and lasses). When you do get the "modulus" from the DMT, etc. do you use that for your settlement analyses or do you modify for any increase in modulus as the load is applied - i.e., modulus valuse increase "with depth" hence, after, say 1/4 tank (on sands) is applied, then the modulus is not the same as the initial modulus of your field investigation . . . just wondering.

 

[lovethecold]

In regards to using the DMT for stability analysis, I was thinking of the article at the following website:

http://www.marchetti-dmt.it/pdffiles/totani_bali2001.pdf

Go to section 4.4 on page 4. The problem with most of the soils where I am at is that most slopes are in the process of failing. It seems to me, knowing where and the extent of the failure plane would be of great advantage in our slope stability analyses.

As far as using it for settlement analyses, most footings that require more than 2000 psf and or more than 6 or so feet wide generally go to deep foundations. Shallow foundations capable of more loading are almost always on sandy lean clay. It seams sandy lean clay (which, in this region includes cobbles and boulders) is a poor choice of soils to use the DMT in.

palmahouse, I agree with Mccoy, it would be nice if more folks would do proper soil investigations. Sometimes it seems the larger the project the more they want to go with a "cut corners engineering" approach. Too bad, it is these projects that would benefit the most from a proper soil investigation.

Thanks for all the responses, this turned into an interestng discussion.

 

[Mccoy]

In regards to using the DMT for stability analysis, I was thinking of the article at the following website:

The authors make up the whole 'Marchetti' research group at the local university, engineers who've been working with the dilatometer over 20 years now.

I feel that any proposal needs a validation from third parties though, since, as you well know, every ugly creature tends to be beautiful to the eyes of its creator.

Specifically, I have no feedback whatsoever on the use of the proposed method here in Italy, and again, the dilatometer itself is used very little.

 

[Mccoy]

When you do get the "modulus" from the DMT, etc. do you use that for your settlement analyses or do you modify for any increase in modulus as the load is applied - i.e., modulus valuse increase "with depth" hence, after, say 1/4 tank (on sands) is applied, then the modulus is not the same as the initial modulus of your field investigation . . .

If I get the hang of it, you mean correct the modulus so we have something in the fashion of the result of an oedometric lab test where the modulus itself varies in function of the applied stress?

I'm not aware of any correction done or proposed (beside mere overburden corrections) and, you are right, if delta pressure is significant the modulus might change significantly; but then what kind of estimate would you propose?

 

[Mccoy]

BigH,

I found a relatively recent relationship which yields the edometric modulus in function of qc, s'vo, delta_s'v, where the latter is as usually increase in soil stress due to loading. That's for CPT, not DMT

It's in Eurocode 7-2 and was originally proposed in DIN 4094-1:2002 (the German standards).

 

[BigH]

When I worked in NJ, the firm liked to divide the load into quarter loads, apply the first quarter and determine the settlement. Increase the modulus value due to the increase due to the new load, then apply the second quarter. Repeat until done. I couldn't convince them that all they had to do was to use the "half-load" modulus throughout the whole of the range - but in any event, the settlement, actually, would be smaller than just using the initial modulus value. (read this for "sand" - of course, in clays, this doesn't really apply to the consolidation) I don't have access to the Eurocode or German standard. Would be interested in a synopsis, though.

 

[Mccoy]

Did they measure the modulus by lab tests?

Here they usually calculate it from resistance vaues of static or dynamic field testing.

 

[BigH]

They ran triaxials on reconstituted lab samples to "nearly" the same density and then took appropriate E values from the resulting curves. Not high tech, but this was back in the very early 1980s.

 

[Mccoy]

They ran triaxials on reconstituted lab samples to "nearly" the same density and then took appropriate E values from the resulting curves. Not high tech, but this was back in the very early 1980s.

Ah Ah, I believe hi tech sampling methods like frozen sand samples are so costly that reconstituted samples keep being run of the mill for such problems.

I heard about one research they started here in Italy with frozen samples then they had to stop off because lack of funds.

To wander a little Off topic, I'm still a big fan of the Japanese school findings on frozen samples and SPT (Hatanaka & Uchida 1996, Hatanaka et al. 1998).

Don't know if there are most recent, exhaustive studies like those...