Swell Pressures and Overexcavation 1

[JohnnyB_]

Hello -

I am a residential structural engineer trying to correlate the amount of overexcavation required in order to approximate post construction heave and settlement of less than 1 inch. For instance, I just received a geotechnical report that estimates the swell potential of the soil to be any where between 3500 psf and 7600 psf. The report is recommending a 2.5 foot overex for spread footings. How are they coming up with the amount of overex needed in order to properly mitigate the structural movement due to soil settlement/swell potential? Is there a good rule of thumb or some type of white paper on this correlation?

The reason I am curious is, due to my location in rural Colorado, it is very hard to get a full geotechnical engineering report within a reasonable timeline. Too much work not enough geotechnical engineers. The way I am trying to bypass this issue is to classify the USCS soil type using ASTM C136 and if the fines are greater than 15% I will have the geotechnical engineer run a swell/consolidation test on the soil giving me a potential psf swell value. Using this potential psf swell value I will estimate the overex needed to support the structure.

Please let me know your thoughts. Thanks in advance.

 

[MTNClimber]

First let me provide this disclaimer, I do not recommend taking responsibility for providing overexcavation recommendations unless you are a geotechnical engineer experienced in the area that you are working. In the 90's and 00's a lot of Colorado geotechnical engineering firms had to close their doors from law suits due to swelling soils. I would not recommend opening yourself to that liability.

The whole idea is to over excavate and moisture condition the expansive soil which will be exposed to change in moisture. So first you have to identify how expansive the soil is within the anticipated depth of wetting. The depth of wetting is still contested and depends on how much the site immediately around and under the building is going to be exposed to change in moisture. I've seen some papers that say its only 10 to 16 feet and others that say it's closer to 24 to 30 feet below the ground surface. I've seen some places justify less depth of wetting around commercial buildings versus the residential lots where there is typically more landscaping (which translates to more water being dumped into the ground).

Here's a method that made the most sense to me. Say you use a 24' depth of wetting, you should then take intact samples of the soils of the upper 24' and run swell/consol tests on them. Then you run a quick calculation of the upper 24'. If 0'-5' is non-expansive but 5'-10' is 5% expansive and 10'-24' is 2% expansive then the calculation will look like this:
(5ft x 0.00 + 5ft x 0.05 + 14ft x 0.02) x 12in/ft x ReductionFactor = 6.36 inches x ReductionFactor.

This reduction factor is typically put in because the lab tests are run on fully saturated soils, which doesn't typically happen on residential sites. I've seen factors between 0.7-1.0 used, but that is based on engineering judgement.

Now that you have the total estimated swell, you can figure out how deep you would have to moisture condition to theoretically reduce your anticipated swell to a tolerable level that is acceptable to the owner.

That said, I've also seen geotechnical engineers just say, 3 feet always seems to work for me, so I stick to that...

Long story short, get a local geotech involved.

 

[JohnnyB_]

Thanks MTNClimber. I see what you are saying. If you used 3' of overexcavation that would take away 3' of potential swelling soil.

One thing that I have noticed is that when the geotechnical engineer comes to the site he only digs down approximately 5'. Would you say he is assuming that the soil strata is equivalent down to the 'wetting depth' you described?

 

[MTNClimber]

I'm not sure what they are assuming but maybe you're correct. Everyone approaches this differently. I've seen different offices in the same engineering firm approach swelling soils in different ways. Most of it is experience based.

 

[EireChch]

In the UK there are heavily overconsolidated CLAYs which are subject to volume change. This issues is mainly managed through the use of void formers. These are essentially polystyrene type blocks that are placed beneath slabs. The compress as the soil heaves.

I dont think swelling calculations are well proven to be honest. I have seen many ways of approaching it, which isnt really a good sign.

 

[MTNClimber]

I agree that they are not well proven, which is why local experience trumps a random guideline provided by some random person on an engineering forum.

We have void form too but its made from wax coated cardboard.

 

[TigerGuy]

In some soil profiles, there is the surficial zone where moisture changes frequently. Below that, the soil moisture is fairly constant and will not change in volume.

Swell tests will give you the amount of swell potential and the swelling pressure. Those values help to determine the amount of material to replace beneath a structure. If you have soil that can swell at 3500 psf, and your foundation exerts 2000 psf of load on the bearing strata, you probably want to replace some of that underlying soil.

 

[JohnnyB_]

Thank you all for your discussion. It has helped me have a more rounded understanding of the type of engineering required when mitigating the swell potential of soils.