Screwy looking test results 2

[stochas]

In the course of a forensic investigation of failed residential foundations in compacted fill, I was provided with the following test data by a geotech firm:

LL = 36%
PI = 17%
USCS = CL

Standard Proctor Tests:
MDD = 127.2 pcf OMC = 12.8%
MDD = 121.0 pcf OMC = 15.8%
MDD = 107.5 pcf OMC = 20.2%
MDD = 97.6 pcf OMC = 25.1%

It appears to me that the Proctor results plot at or above typical ZAV curves for normal soils. Also the wide range of densities and moistures seems extreme for a single soil type.

Are these test results indicative of potentially problem soils? If so, what soil properties or testing procedures could yield such test results? Finally, should these test results have sent up red flags to the geotechs involved?

The litigation in this case is completed, so this inquiry is only meant to satisfy my own concerns.

Thanks,

stochas

 

[BigH]

1. Do you have the specific gravity of the soil - then you'll know if above ZAV. You are correct in that the spread does appear quite wide - sure the soil is of same source? same stratum? It may have come from a couple of sources.
2. What were the other problems? I know of a case history where housing was built on compacted clayey fill. They had settlement problems. When they dug test pits they found an interesting situation where the upper part of a "layer" was significantly stronger than the lower part (by hand vane testing), the moisture content of the upper was lower than the lower zone, etc. Not sure they could do density - but they did notice a "sandwich effect" of the "stronger" upper zone to lower zone. Apparently the layers were placed too thick and/or the compaction was on the "light" side.

 

[PSlem]

I've seen a big range like that where the contractor had just pushed wet clayey silt into place for the backyard. Troxler testing showed a range from 88-95% and a similar large difference in the MC

 

[Focht3]

The results look wrong - clearly. And the first two (three?) aren't from any CL materials I've ever seen; perhaps a GC - but not CL. Or they used the wrong mold constants or weight conversions when they made the calculations.

Do you have any other lab tests or field observations you can share? What conclusion(s) did you reach? Did the results plot in a normal pattern, even if it's above the zero air voids line, or do the high values appear random?




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

 

[bb29510]

Your optium moisture is too high for the test, need to start at about 6% and work up in 2% points. Once you get to the optium. point on the other side sometimes will follow the graph and sometime not, because of the propities of the soil reacting with the large amount of moisture. The specfic gravity graph is really not that accuracy, Its mainly a broad graph. which doesn't mean your results are incorrect.

 

[BigH]

If you have results on the wrong side of the ZAV curve - and this is so presented in reports/field reports, etc., be careful . This is why I suggested you make sure you plot the ZAV to the correct Gs value. I know of one very reputable firm that was in major headaches in one of their offices because there was a problem with the project and the field test results were on the wrong side of the ZAV. Just a bit of caution.

 

[stochas]

Thank you all for your comments. I do not have access to the lab Proctor data or specific gravity information.

Other information which is relevant is that the soils are residual soils on Triassic Basin sedimentary units of NC which are known to be moderately expansive. I suspect that oven drying of the soils in the Proctor tests would have driven off bonded water resulting in lower MDDs and higher OMCs.

Overall for the entire project, there were three broad groups of Proctor results: MDD = 95 to 100 pcf, 105 to 110 pcf, and 120 to 128 pcf. Each group had MDDs plotting above Gs=2.75, but all results were suspicious.

The end result is that the fill constructed of these materials failed several times on exposure to water causing severe foundation distress.

 

[cdh61]

stochas,

Focht3 is correct in the first 2 sets of MDD results. The 3rd set does match data in our files for soil typical of your reported LL and PI results (LL from 35 to 40%). The 4th OMC is to high for a soils with the LL and PI reported. You do not have one soil type.

As for potential problem soils, all sets of data (MMD and OMC)could be protential problem soils, but exactly what do you mean by potential problem? Such as frost action, settlements and swelling? Please post stratigraphy encountered at site and location (frost zone?).

Also, what about doing grain size distribution analysis on the soils you have MDD values for? This would tell you if the soils are the same, and the likelihood of a uniform fill.

My opinion is the MDD values should have sent up red flags since they represent different soil types.

regards

 

[Focht3]

Okay, I think I get the picture - the test results in your original post were the values provided by whoever ran the lab compaction testing, but you don't have the actual test calcs. Is this correct? How do the field density test results compare to the lab optimum values?



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

 

[stochas]

Thanks again for your comments.

I was provided the previous data of work performed by others for review after the compacted fill at the project site was found to be too soft at time of footing construction. I do not have access to the Proctor lab data and further testing is not a option at this time.

The field density tests show that all tests were greater than 95% of the MDD but that the field moisture contents were 0.0 to 2.1 percent DRY of optimum. The site was totally in 4 to 12 feet of compacted fill so the site stratigraphy has little influence.

I suspect moderately expansive soils are present and that the Proctor data show this.

My basic question is: Should any lab technician and/or engineer have been suspicious of these Proctor results, to the point of running retests or other tests such as the specific gravities?

Thanks

 

[Focht3]

My basic question is: Should any lab technician and/or engineer have been suspicious of these Proctor results, to the point of running retests or other tests such as the specific gravities?

In my opinion, YES.



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

 

[dirtdoctor]

One quick comment -

The lean clays you have in the Raleigh area should be installed wet of optimum conditions. Even the slightest bit of 'dry-packing' can cause big problems when exposed to water. And, depending on the test method used to determine in-place density values, the material could actually be drier than test reports indicate. Nuclear gauges are all too often stuck in the ground and results are never verified by performing oven moisture checks. My advise to all technicians that have worked for me is to always run comparison moisture contents versus a nuke gauge, or run the risk of looking a fool.

As a general rule, if you can count on the lab tests to be done right, use field wet density values (versus Proctor maximum wet values) as a quick field comparison. Then, get the final numbers with overnight oven moistures. Stove moistures done in the field can also be a big issue, as most techs don't 'cook dirt' like they should

 

[BigH]

Dirtdoctor - . I had "problems with crushed stone compaction and found out the actual lab moisture was 70% of the nuke moisture content - this made a 2% difference in the compaction level (to our benefit!).

 

[DRC1]

We used to do a lot of sand cones for FAA work. We would get field moistures by using a 6" loaf pan. We would put in a half in thick layer of dirt and weigh it, (we also had the tare wt. on the pan)Then we would put in a bottle of "Dry Gas" (2 if it was really wet or clayey) and light it. It would burn out in a few minutes and we would get fairly accurate dry weights. It was also kind of fun.