Surface Sampling

[ebenozen]

I'm interested in collecting core samples from the surface, without using a drill rig, for triax testing. Details on collecting the sample and extruding it to ultimately get an undisturbed and testworthy sample would be appreciated.

Thanks

 

[Ron]

Why? Is this an academic exercise? Check Sowers' book on Geotechnical....he gives an example of surface sampling, but it is by excavation, not by coring or tube sampling....though after exposure, you could push a Shelby tube in from the surface without much difficulty, assuming reasonable soil conditions.

What kind of soil are you sampling and what do you expect to gain from the data?

 

[ebenozen]

This is not academic. In this case, the material at the surface is expected to be continuous for roughly a 20 meter vertical height. The material is a weathered saprolite. In terms of classification, it borders on being hard soild/very soft rock. I'm ultimately hoping to determine a shear strength that I plan to use for slope analysis.

How could I go about driving a Shelby tube from the surface? I'm trying to think through the details.

I will check Sowers' book. Thanks for your comments Ron, any others would be greatly appreciated.

 

[BigH]

In a "soil" that hard (or a "rock" that soft) it would be hard to drive in a shelby tube. I would suggest that you dig a test pit big enough to stand and work in (think of safety in the pit!) and then obtain a block sample. It is well explained in Fang's Foundation Engineering Handbook 2nd Edition, page 50.

 

[ebenozen]

What kind of testing did you have in mind with the block sample and how are the samples prepared from the block?

I also plan to collect a few large-scale direct shear samples (10" square sample) using a custom made metal box carefully driven into the material (have done this before).

My ultimate goal with this thread is to figure out a way to get a core sample for triax testing. If anyone has ideas, I'd appreciate it.

Cheers

 

[Ron]

From a block sample as described by BigH you can trim samples for triaxial testing, direct shear, consolidation or unconfined compression. In addition to the safety, be sure that you maintain the moisture conditions as drying can cause problems with sample integrity.

 

[BigH]

OP repeated says "drive" - this doesn't necessary give one an undisturbed sample unless the "special" ring is very large and you only want the small middle. With the block sample, you would trim the sides to match, roughly the sides of the box; trimming the sides deeper than the box. Then you can cut across the bottom below the box; if the box isn't snug, fill it with grout; lid up and carefully forward to lab. As Ron says, you can get all your trimmings of test specimens from it.

 

[motorbiketocrank]

Off hand, I'd question the approach your taking. Saprolite weathering will vary with depth. And typically, slope stability would be controlled by "discontinuities" in the material and not the cemented strength. Not sure why you don't want to do borings but I'd strongly consider trying to do a couple with a core barrel to get a good feel for the material. Perhaps even use a remolded sample in direct shear to model behavior along the discontinuities as a soil.

 

[ebenozen]

Response to your comments geobdg (very good ones)...

I am trying to get an undisturbed sample and diamond coring destroys the soft saprolite, so I'm considering other options. I haven't tried borings and driven Shelby tube samples and wanted to start with considering options for obtaining samples from the surface without using a drill rig. Budget is limited.

It is our experience with this material that discontinuities do not have a strong impact on slope stability. The joints are present as you suggested, but the weathering process to form saprolite seems to have limited the continuity of the joints in slopes in this material elsewhere which is why I want to do the triax testing.

I also considered doing remolded direct shear tests as you suggested, but this type of testing seems to homogenize the material and yields predictable results. These test results seem to be consistently 0-2 psi choesion with a phi ranging from 30-35 deg.

 

[aeoliantexan]

I'm not familiar with saprolite, but if it behaves like a soft rock, you may be able to cut rectangular samples with a gas-powered abrasive saw used for concrete. You might also consider a concrete coring machine such as those used for sampling pavements and use compressed air rather than water to remove the cuttings.

 

[fattdad]

I guess my biggest concern is whether you can telegraph the strength data from the upper couple of feet to the strength characteristics of the residuum at a greater depth. You see, the near surface wetting and drying cycles will affect the strength to a greater extent than seasonal wetting and drying will affect the conditions at 5 or 10 ft. Then again, I don't know where your job is located.

I like the idea of a block sample obtained (safely) from a test pit.
I agree that you can carve (sculpt) any shaped sample at a later date to fit into the triaxial cell or direct shear box.
I'd bet you can't push a shelby tube and if you did, it'd affect the strength results.
I wouldn't rule out in-situ testing.
I'd seriously consider rotosonic drilling, but the mobilization and footage rate for a few shallow borings may be too great.

f-d

¡papá gordo ain’t no madre flaca!

 

[ebenozen]

I'm sure we can cut/trim a block out of the saprolite, but what could we do with it?? Sculpting a sample to fit into a triax cell sounds really iffy. Have you ever done this? If there is a good way to do this, then this would be option #1.

As fatdad suggested, I planned to excavate a trench and collect the sample from the bottom. The project site is remote, but we have access to heavy equipment, so cutting a good wide safe trench won't be a problem.

My first thought was to push a Shelby tube (minus a rig) somehow, which led me to fish for ideas.

Sonic drilling is another good idea, but logistics/budget are a problem.

The only problem with in-situ testing (plate load test, etc.) is with the limited range of normal stresses that can be applied. At the last Gecongress, I saw a very interesting presentation where an in-situ test was done at a quarry in Irwindale, CA. by loading the walls of a trench with concrete until collapse occurred.

 

[fattdad]

there is a product specifically made for soil trimming. Here's one example.

http://www.shop.certifiedmtp.com/Ad...-BCL-382.htm?gclid=CIr18cL0_7gCFQ2Z4Aod1DsAKw

f-d

¡papá gordo ain’t no madre flaca!

 

[oldestguy]

I've run several large area in-place direct shear tests on different materials and, if you think about it, you easily can manufacture the equipment. You need a way of applying a normal pressure (I used a proving ring to measure load)and a resistance to that load (I used a large back-hoe bucket) and a screw jack to apply the load. To keep the bucket from gradually applying more load as its hydraulics leak, you set a strut under it to keep it there.

For applying a shearing load to the shear box, I used a porta-power hydraulic hand pump and hydraulic cylinder, fitted with a pressure gage (for knowing the shearing load). Push against the sides of the test pit for a reaction.

The shear box was 12" square with a serrated plate inside for applying the normal load. I used heavy walled square tubing for the box construction, welded. The larger you make the box, the less chance for errors, assuming you can apply suitable loadings.

The shear box is set on the material to be tested and the material below is gradually trimmed to allow the box to gradually settle down with undisturbed material inside. Add filler where any side voids occur (can use mortar or other suitable material). Trimming of sample is a pretty standard test procedure for any similar activity.

If you want to measure strain, use a standard lab dial gage, suitably mounted.

For near rock-like material you would need separate set-ups for each data point, as I think you would need with any shear test..

Look at the many direct shear testing devices out there. Someone had to design each and build them. Why not you for this special job?

 

[Edwardcollins]

Please refer any guide or books to know more about surface sampling. Geotechnical Engineers can help you for more reference in Surface sampling. I have done soil sampling with the help of DST Consulting Engineers (

http://www.dstgroup.com)

at Ottawa in Canada.

 

[Mccoy]

You can drive shelby tubes by careful manouvering a backhoe on the bottom of an excavation, or by using some special appliances rigged to the helical anchors used in CPT small rigs, but it works reasonably well in soft soils, don't know if it would be possible in the saprolite you mention.
Also, I share fattdad's concern about the extrapolation of the parameters to the near surface layers to those more in depth, unless of course the shallower thickness of the slope is the one at risk.

 

[mikeofBelAir]

I have a couple of questions..... How representative of the actual soil/rock will your sample be? If the soil/rock weathering varies considerable, then do you want to test the most competent of the material, or the least competent? Would testing a sample that is remolded from the original material give more consistent results, and provide a more realistic set of values for your stability analysis?

 

[coneboy]

In the past we had to sample a near surface, very hard clay/silty clay/sandy clay (Su = +400 kPa) at a semi remote location.
On our first attempt we crushed and folded all the shelby tubes.
Then we lost the use of the drill rig so we ended up at the position you are at.

Our second attempt using home made, heavy wall, robust shelby tubes was successfull.
We pounded the tubes in with a vertical hammer device. (not the best idea)
Pulling the tubes out was also equally challenging.
I never actually got to see the lab results as I was rushed on to another project.

In agreement with Fatdad and McCoy:
Years later, I saw some CPT data for this area and observed that the soils slowly decreased in strenght with depth. The surface soils that we sampled were merely desicated and lightly glaciated. The CPT data showed that by 15 meters in depth the Su values were in the 100 kPa range.


Coneboy