Effect of cobble to bolder sized rock on clay swell

[sc]

Having never really dealt with clays with rock inclusions that vary from cobble sized to boulder sized (but <1.00m) I thought I would ask the experts.

The clay fills all voids and is moist (all year round) and is a basaltic clay.

Does the inclusion of these rocks (basalt) with a volume of approximately 30-60% per m3 have an effect on the shrink swell of the basaltic clay?

Any comments are appreciated.


regards
sc

 

[Focht3]

There are two types of coarse aggregate/clay mixtures: where the aggregate &quot;floats&quot; in the clay, and where the clay infills between aggregate that is in full contact with adjacent particles. Your description suggests that you are dealing with the &quot;floating&quot; arrangement.

In general, the volume of rock will reduce the swell that would occur if the soil mass were all clay. But this is a difficult problem to analyze, and geotechnical engineers may or may not make a &quot;rock correction&quot;. It's really a matter of judgment and local practice -

The &quot;clay infill&quot; problem is different; I suspect that the treatment of this condition varies quite widely.



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora.

 

[sc]

Focht3,

One particular area I am looking at is actually a mix of relatively dense &quot;interlocked&quot; cobbles/clay mix (approx. 600mm to 1200mm deep) overlying a clay with larger floaters (approx. 600mm - 1500mm in size) and cobbles.

The main reason I asked the question is that another local geotechnical engineer appears to have established an estimated ground surface movement based on lab testing of the clay only and disregarded the effect of the basalt that is present. By doing this the design of the house footing is going to be very conservative (in my opinion).

My own &quot;gut feel&quot; is that if the volume of clay is decreased and the volume of rock increased then there is less potential for swell/shrinkage to occur. It therefore follows that the rock has a major influence on the shrink/swell potential of the soil.

My discussion with the geotechnical engineer leads me to believe that he just ignores the issue because he would rather be very conservative than &quot;accurate&quot; (I use the term very loosely). He openly admits that there is essentially no local evidence to back up his approach.

Are there any references that may help me to improve my own understanding of the effect of rock cobbles and boulders on expansive clays.

thanks
sc

 

[cdh61]

sc,

The floaters will have little impact on overall swell and settlement. But the floaters can have an impact on differential swell and settlement depending on how close and how many are located near the base of the foundation. The lab testing can not account for cobbles and boulders and represents the worst case of swell and settlement. It is difficult to account for floaters in analysis on swell or settlement because the floaters are random and in such cases recommendations on the conservative side are common.

Also, if the foundation is placed on a relatively thin zone of dense cobble/clay, then the underlain soil properties can govern the design, which in your case is the clay zone with floaters.

You are correct in saying a decrease in clay with an increase in cobbles/boulder will reduce the potential for total swell and settlements. But, the overall soil condition (matrix) needs to be assessed, and eventhough you may have 30-60% rock by volume, the density of the clay can still govern in the design if the overall swell or settlement of the clay exceed acceptable limits.

regards

 

[Focht3]

Keep in mind that expansive soils pose two different, but related, concerns: soil heave and swell pressure. The presence of the &quot;floaters&quot; or &quot;infill&quot; will not affect the swell pressure, but can affect the soil heave. One soil - two potential problems. And differential heave due to varied amounts of &quot;floaters.&quot; Yes, one soil - three potential problems. (Er, I'm experiencing Monty Python withdrawal - and it's late.)

By the way, what do you consider to be &quot;very conservative&quot;?

As a former Terzaghi Lecturer has said, &quot;It is better to be approximate and correct than precise and wrong.&quot; A soil swell estimate is really an order-of-magnitude value; I seldom talk about heave estimates in increments smaller than one-half inch for fear of implying accuracy that simply doesn't exist. Even though I can calculate the estimate to as many decimal points as I want -

When LRFD finally &quot;addresses&quot; the geotechnical aspects of foundations (in a coherent manner), expect soil swell to get a BIG load factor. Plotting swell test results against PI, Liquid limit, initial dry density, CEC, etc. results in a classical &quot;shotgun&quot; pattern. A &quot;good&quot; soil heave prediction, in my opinion, means getting the swell magnitude within an inch of the actual value - and the differential heave within half an inch...



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora.

 

[sc]

To help understand where I'm coming from let me explain a little about Australian methods of site (soil) classification for residential housing.

We basically make a visual appraisal (backed up by limited testing in some cases) of the soil and classify the site into 1 of 6 groups. These groups are related to the estimated ground surface movement (divided into 0mm, 0-20mm, 20-40mm and 40+). The groups are A, S, M, H, E & P. A for solid rock and pure sands grading through to extremely expansive clays and P for problem sites. These classifications then relate to a group of standard footing designs (that are considered to be conservative).

The site has been classified as H based on testing of the clay between the cobbles and boulders. The difference in beam depth is 100mm and an overall increase in reinforcement size. This doesn't sound much but keep in mind that we generally don't have to deal with deep seated movements greater than 3.00m or frost heave. Overall it generally means about a 25-30% increase in construction cost for a slab.

So by very conservative I guess I mean that the geotech is ignoring certain benefits of the cobbles present and treating the soil as being much worst than it may be. This takes a conservative standard design and makes it even more conservative by taking it a classification lower.

I hadn't stated this previously but we also have a layer of good quality limestone fill (compacted to 95%std) overlying the natural soil (cobble/clay layer). The Australian standard allows for a reduction in classification if the fill is controlled and less than 080m in depth. That is it recognises the more stable base created by quality fill.

Not sure if that was clear enough but it will have to do as I think I'm heading down the Monty Python track and saying alot of nothing.

So unfortunately I think that given there is generally not sufficient knowledge in the general domain to determine the level of difference that the cobbles and boulders make, I am better of to leave the classification as is. Really what I'm saying is that I don't have the experience yet (in this particular area) to make a good argument to the geotechnical engineer (who really has no more experience in this particular area than I do).

Thanks for the discussion and I hope others can add to it.

regards
sc

 

[Focht3]

Let me leave you with this thought to ponder: the cost of the slab is a relatively small part of the cost of a house. What will that 30% increase in slab cost do to the final cost of the house?



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora.

 

[ChrisMcLean]

No-one expects the Spanish Inquisition.....

 

[BigH]

I think Focht3 hit the head on the nail - cost. For one house, I think I would go conservative - what is a little extra concrete vs headaches (to owner, to builder, to engineer). Now if this is a subdivision with many houses, the little costs add up and then a more detailed picture needs to be addressed, adjudged to pros and cons and handled appropriately.

 

[sc]

I agree that the increase in cost should not really be a factor, but poor joe bloggs that has been saving to build the home thinks it is a factor. In this case the added cost is only 5% which in engineering terms is only half of the general contingency amount that should be allowed for.

Yes there are about another 15-60 houses to go in on this subdisivion with little chance of the soil varying. Overall the cost increase would be significant if it was all being build by the one developer, however it is not.

As I have said my gut feel is that the classification and subsequent design will be too conservative, but I just don't have the experience yet to make that decision for the owner. So I will go with the original recommendation.

In the mean time any relevant references would be appreciated.

thanks

sc

 

[cdh61]

sc,

Lets see, 5% cost increase, not much local evidence to back up geo statement, 15-60 more house to build... There appears to be more work for geo engineer (company). I would phone geo back the engineer and ask what it would cost for a little more investigation work to see if cost savings are possible. With all of the above, savings & potential for more work, it may be in the geo persons best interest to conduct more work now for future work to provide savings to others and your client. No guarantes of more work for geo, but it is worth a phone call.

Regards

 

[Focht3]

I continue to be concerned that &quot;refinement&quot; of the recommendations amounts to nothing more than trying to &quot;fine tune with a coarse knob.&quot; (My apologies to those of you with no life experiences using manually tuned television and radio sets -)



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora.

 

[sc]

My mistake should be 15-20 houses.

I have spoken to the geotech and the local building surveyor and both agree that the classification is conservative given that there are 10 existing homes (all <2years old) sitting on the same soils all classified the same class but with less cobbles/boulders. None of the existing show signs of movement and we have had an up & down 2 years with respect to rainfall.

So despite this I won't be &quot;fine tuning with a course knob&quot; (good way of putting it). As there is still insufficient evidence or gain in knowledge for me to make the change. (I don't have the resources to do the field & lab work to confirm my &quot;gut&quot; feel).

Again any relevant references would be appreciated.

regards
sc

 

[Focht3]

Unfortunately, you won't know how good (or bad) the foundation design is for at least another 20 years or so. Remember that the houses should be built with a design life of 50 to 100 years.

And we seldom know if a design is too conservative because it meets the performance criteria. We know when it isn't conservative enough - that's when the foundations experience failures. Unfortunately, we learn more from failures than from successes. Too bad more failures aren't documented and published in the literature -



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora.