mjohan:

Some data is needed.  For starters, pleae provide the following:

1.  Soil stratigraphy.  Thickness from pile cap bottom or from ground and giving the depth to the pile cap, engineering properties such as "N" value, undrained shear strength, effective friction/cohesion.
2.  Types of piles.  pile type, pile size (diameter)

25 ton piles is quite low - are you using timber piles?  You've indicated a 4-story building with full basement.  Why are you on piles - for the columns only or wall support too?  

The basement, assuming 3m deep would "off-load" some 60kPa soil pressure.  What is the total gross loading to go on?  How does it appear to the 60kPa?  Did you run any analysis of doing the building on a mat foundation and using the unloading as countering the building loads?  Just a thought.

Give us some more info and keep an open mind to other solutions.  

Hi BigH,

Let me give you some more info.

Top 10 FT - N-values ranged from 1 to 50.  Less than 10 was predominent.

10 FT - 40 FT - N-values in the range of 30 bpf.

This site is on a landfill.

The site is on both cut and fill.  
Yes, the piles are timber.
The piles would support all loads, including garage slab.  Do you feel the settlement is excessive and requires a different analysis?  Should the geo-tech consider a different pile?

I actually had a discussion with a co-worker regarding the possibility of mats.  I have never designed a mat before.  I am assuming from a cost perspective, it would be benificial if the mat eliminates the piles.  Is this true?  What is the approximate required bearing pressure of the soil for a mat to be utilized, without piles( Column loads 225 kips, exterior wall loading 6 kips/ft)?  The building is 160'X100'.  Just a guess, I am thinking the mat might be 6' deep.

Thanks    

  

I re-read the geo-tech report and a 2.3 inch settlement was based on a column load of 180 tons over 30 years.

The piles are also 8" diameter timber piles.

I think BigH forgot to mark this thread, or he'd have responded by now.

I, for one, never provide settlement estimates to a tenth of an inch.  It implies a level of accuracy in prediction that doesn't exist - particularly with piles in a landfill.  With timber piles driven through a landfill, with materials having blow counts of 10 or less...I'd say that 2 to 3 inches of settlement isn't unreasonable.  If the piles are entirely within the landfill material, the estimate may be entirely too low.  And I would expect above average levels of differential settlement are possible, on the order of 2 inches - or more.

You have a really poor site -

I believe this property is some of the last land available in an east coast shore town where property values are through the roof.  I suspect it was never bought until now, because the site is so poor.

A big concern for these piles is a clay layer which is 30 feet down.  The 8" piles would stop short of this layer, @ 20 feet, yielding a net capacity of 25 tons.  Reaching bedrock is not available since it is about 1000 feet down.  

Has anyone designed a building with similar conditions?  If so, was consideration given to differential settlement?  The main concerns are not necessarily structural, but maintenance.  ie, cracking stucco, gypsum, etc.
One idea we have discussed is to design a post tensioned slab tied into the grade beams.  I would be curious to know if other structural engineers have designed a multi level building with a post tensioned slab to minimize the differential settlement.  If so, what programs did you use to assist in this design?

I am also curious if a different type of pile might be available which would not settle as much.  Maybe concrete caisson piles.  

Focht3, thanks for the reply.  To answer your question, the piles are not in the landfill at 20'.

Thanks in advance,
Mark

Is it possible to use expanded base piles in your area, either spread boreds or Franki type?  With spread boreds you should be able to get 4 or 5 ksf bearing; best check with a geotekkie!

As far as surfaces, if differential movement is expected, I'd build some flexibility into the wall system and avoid using brittle materials like stucco...

I'll tell of a job I did in Sault Ste Marie Ontario - and I saw our company's files and they did a plant site somewhere else.  Might help you in your quest - hopefully.

Soils - 20 to 25 feet of loose silt to soft clayey silt
        10 feet of compact (medium to non-Canadians) sand
        100 feet plus of soft normally consolidated clay.

This was a school building extension - two stories.  Designer had suggested steel piles of 16 inch dia for 25 tons.  Architect asked me and I though wood piles deriving purchase in the middle of the sand might be used for 15 to 20 tons.  Did investigation - confirmed in my mind; recommended wood piles.  Did pile load tests and got 0.2 inch settlement at 40 tons.  Also did an uplift test to "check" out possibility of downdrag.  Recommended 15 tons for design - with caveat that the piles be no closer than 5 to 6 feet apart.  Why? - so that each pile acts independently with no overlap and hence no loads to underlying soft clay.

Built it - while initial school building had problems galore, we had none (after 3 years I visited the site with the architect).  Was fine.

Oh yes, saved the school board $86,000 against their original quote of $276k -- even after paying for pile load tests and full time inspection.  My dad - a school board business administrator - would have been proud!!!

Thanks BigH,

I never thought to ask the builder if they wanted to have an actual pile loading test performed.

Mark

I'm a little confused.  (My wife says it's not unusual...never mind. )  Please expand on the details of your project.

A side issue: If the landfill is only 10 feet thick, why not remove & replace?  The owner will be better off in the long run...

Back to your original question:  If the material from 10 feet down is undisturbed native sands, I think the settlement estimate is too much.  After all, the tips of the piles will likely be about 7 to 10 inches in diameter; 2+ inches of settlement for that footing width represents a bearing failure.  Any settlement greater than about 1/4 inch represents skin friction failure.  If you look at the pile group as an equivalent spread footing, the bearing level is at about 2/3 of the pile length - say 13 to 14 feet.  I'm going to guess that the pile group has a diameter of about 6 or 7 feet.  In this case, all of the settlement will occur within the sand (14' + 2*7' = 28').  And the "equivalent" bearing pressure at 14 feet is about 8 to 9 ksf.  The pressure will be in the neighborhood of 800 psf by the time it reaches the clay, and spread over a circle with a diameter of roughly 20 feet.  I haven't run the settlement calcs, but I have a hard time seeing an estimate of 2.3 inches.  Sounds like an effort at CYA gone bad.


I'd run at this project - assuming the landfill remains in place - as a suspended slab.  If you get a good piling system, differential movement is a non-issue.  To reduce the loads, look at timber framing and wood floors.  A suspended slab will facilitate ventilation of methane gas as refuse decomposes.

If you look at other foundation systems, expect to look at much higher construction costs and a longer schedule.  Driven PCC piles, caissons/drilled piers, Franki piles - all can be made to work on this site.  It'll only take time and money!

The geotech has stated that conservative numbers were used when predicting the amount of settlement.  How many safety factors were used??

Focht3, I am interested in your statement regarding any settlement above 1/4" represents skin friction failure.  We had a conference call meeting and the geotech insisted that 2" is an allowable amount settlement.  At this meeting, the geotech said that many books back this type of settlement for timber piles.  I am totally confused?!?

Thank you for all your insight.

I guess it depends on how he/she did their calculations.  Their numbers do seem pretty conservative.  How many safety factors were used?  I don't know; but these types of analyses aren't like LRFD.  It's more likely the "conservative" settlement calculations resulted - in part - from the selection of conservative design parameters...

Think of a plot of skin friction vs deflection as a bilinear curve; at the origin there is no movement - and no mobilized skin friction.  At a deflection of about 0.2 to 0.25 inch, the skin friction is fully mobilized.  For deflections greater than 0.2 to 0.25 inch, the skin friction remains constant, or decreases slightly (5% to 10%.)

I guess a part of the "settlement" calculation could be due to an estimate of the shortening of the timber pile due to the loading.  But that seems to be a stretch -

Are you reasonably close (physically) to the geotech?  If so, try to arrange a face-to-face meeting.  Ask him/her to give you a detailed explanation of the settlement estimate.  Why?  Because it is driving the structural design toward a significantly more expensive solution, and you need to understand what's going on so that you can explain it to the owner.  I've had a few of those meetings.  They can be uncomfortable; but a good design professional doesn't mind explaining "why" at a technical level.

If he isn't close, call and tell him that you need for him to "talk shop" with one of the geotechs in your immediate area.  Be sure to hire your local geotech directly; this may be important in the future.  If your local geotech thinks the geotechnical engineer of record (GER) is blowing smoke, you'll need to go to the owner and ask for another study.  But the odds are that the two geotechs can hash things out, and the GER has a reasonable basis for what he/she is recommending.

Regardless of the answer, you need to have enough confidence to go to the owner with some sort of recommendation.  I don't think you're there yet.