Determining sand density at depth in a borehole

[dBasement]

Does anyone have a methodology for determining the density of non-cohesive sand at a depth up to 50m? There will be groundwater. I am considering either a hollow stem auger or sonic drilling apparatus and drilling to just above the test zone. I would then need to either use a clamshell or sample recovery tool of some type, but I am stumped as to how to get a disturbed or undisturbed sample.

 

[GeoPaveTraffic]

While I don't really like it, a modified California sampler with rings would allow a sample that could probably be in tack enough to run a density test. Also, there is no need to run sonic equipment and you should not use hollow stems. If you are drilling in sand below the water table, use mud rotary drilling methods.

Mike Lambert

 

[dBasement]

Thanks Mike. By California sampler, I am thinking a large-diameter SPT sampler? I agree and and am going to source a mud rotary rig. The o-rings I assume are to seal up the tip of the sampler. Would you just use the weight of soil recovered and the measured volume of the sampler to determine the density? Sounds a little sketchy when you account for slough. If you were to say it can't accurately be done, I might just throw my hands up in despair and say that.

 

[emmgjld]

I have used a Dames & Moore Sampler (2.5"id)with the appropriate lines.
I usually try a 2.5"id Shelby tube, usually surprised at what can be obtained. If you try the Shelby, I recommend a short (16" to 24")tube, pushed to capacity and check density of entire length and 3 or 4 intermediate samples, if you have the capability.

 

[fattdad]

what's the end game? Are you trying to determine whether the density is 112 or 120 pcf? Is this for the benefit of calculating effective stresses at depth? Is this to assess relative density with respect to critical void ratio (i.e., for liquification analysis)?

f-d

¡papá gordo ain’t no madre flaca!

 

[dBasement]

f-d...I'm really not certain. The request came down from significantly higher up the food chain. My only additional information is that it is related to the hydrogeology on the site and that SPT information is insufficient. It wouldn't be appropriate to discuss more of the particulars on an open forum.

 

[GeoPaveTraffic]

A modified CAL sampler, effectively the same as a Dames and Moore sampler, is actually somewhat different than a split spoon sampler. It has brass rings inside that retain the sample. It is common practice in the Western US to drive these samplers and then run densities on the material in the brass rings.

Like I said, I'm not really a fan; but it is the best option that I have seen to collect density samples in sand.

Mike Lambert

 

[Ron]

A shelby tube dipped in muriatic acid and rinsed, then allowed to lightly corrode works well to obtain a sample of sand at any depth. This will allow a reasonable determination of density.
The light corrosion creates enough friction to hold the sample intact until removal.

 

[BigH]

freeze it . . . then sample (an offhand thought)

 

[dgillette]

Don't laugh - it's been done for nuclear power plants. However, I doubt Mr. or Ms. dBasement has a nuclear-power-plant-type budget for installing liquid nitrogen pipes. There are the state of the art, the state of the practice, and the state of the economy.

Stepping tentatively outside of the box: Is the right question being asked by the higher-ups? dBasement mentions hydrogeology, which makes me think this is a permeability question, which makes me think grain-size distribution would be more important than density. I believe the permeability is much more sensitive to small changes in D10 or D15 than it is to density. Dense #10 sand is much more pervious than loose #50 sand.

 

[fattdad]

As related to hydrology, it may be a matter of deterimining the porosity, which via the block diagram is related to density, moisture content and specific gravity. Porosity is required to learn the true ground water velocity, which is greater than the Darcy velocity. Darcy velocity is V=ki and velocity is V=ki/n, where n=porosity.

That's what comes to mind when you say this is related to hydrology.

f-d

¡papá gordo ain’t no madre flaca!

 

[mudman54]

i was thinking osterberg sampler...most east coast drillers have them (if they are worth their salt)