Shear Strength Testing for Gravel Materials 6

[ONENGINEER]

To find out the shear strength parameters of a fill (pit run) materials consisting of gravel and cobbles, what sort of laboratory tesing are recommended. Thank you.

 

[fattdad]

Do you need to demonstrate some critical value? I mean if you get a safety factor within reason and you make an informed estimate (e.g., 34-36 degrees, which is likely conservative) then where's the value in using a lab to show it's actually 42 degrees (or greater)?

If it wasn't for the problem of soil moisture, you could also just make a trial excavation and back-calculate a friction angle.

f-d

¡papá gordo ain’t no madre flaca!

 

[ONENGINEER]

Thank you for he answer. The fill is not in place. The intention is to put the actual parameters for the materials to be used at site during the embankment construction.

 

[dgillette]

The first question is "What do you actually need?" Do you need to assure yourself that the friction angle is at least 36 degrees once you roll it, or do you need to have a precise value, or what?

There are very few labs that can actually test material that coarse for strength, and it's likely to cost a pile of money. We once hired U of California at Berkeley to do some 12-inch (30 cm) diameter CIUC triaxial tests for us (long story), but we had to sieve out the larger pieces to be able to test in that size cell (which of course complicates the calculation of relative density and other things). A long time ago, both Berkeley and Raul Marsal at I forget which university in Mexico were doing larger triaxial tests that that, even on rock fill.

To get around the effects of moisture and capillarity, and any cementing agents in a test cut, you could try a test FILL (instead of exc) to see what you get for angle of repose, recognizing that it might be misleading if the stuff is very rounded. A test fill is likely cheaper than a lab program for material that coarse. A test fill can also tell you about compaction procedures, but with material that coarse, it will be tough even to measure density on the fill, which would likely require a 48-inch or larger ring test (water replacement), let alone min and max.

Or, unless this is a very large project with a very large testing budget, you could just do like fattdad says, and assume a conservative strength value (34-36, or more), and see if your design gives a decent factor of safety. Unless you are really tight on borrow quantity or have some unusual material, or there is some other condition that prohibits building flatter slopes or whatever it would take, you probably can't get in too much trouble that way. In 95 cases out of 100, that's probably what I would do (and have done).

 

[ONENGINEER]

many thanks for the both useful comments. I would further appreciate if you could explain about the test fill. From waht I understood, I think I can find the soil angle of repose while the stock pile of soil is dumped at site by a loader, etc. That would then be the friction angle under loose condition. I may wait for 1-2 weeks so that the newly placed stock pile of soil stablises internally. I guess the angle of repose may well be 36 degrees.

I then design my slope based on a factor of safety of 1.5. So when the materials are compacted at 95% Proctor during the actual construction, I will certainly have a higher friction angle and as a result a factor of safety of greater than 3. Do you have any comment on this approach? Thank you again.

 

[cvg]

I would recommend a test fill, that would be incorporated into the embankment (assuming you are satisfied with the quality)

"A properly executed test fill program should determine the
most effective type of compaction equipment, the lift
thickness, and the number of passes; maximum rock sizes;
amount of degradation or segregation occurring during
rolling; and physical properties of the in-place fill, such as
density and grain-size distribution."

 

[dgillette]

What exactly are you building (and how big is it), and how much would your design be affected if you assume φ is 36 degrees instead if 42 degrees? In other words, how much does a precisely determined friction angle matter?

Last things first:

Having a calculated FS of 1.5, it's rather unlikely that the actual FS could be 3, unless your stability analysis was not only conservative, but ridiculously conservative.

You will not be able to use Proctor tests on this material, and it is no small task to measure either the vibrated maximum density or the fill density with cobbles present, or even just large gravel. Think about whether you really need to quantify the density accurately, or if you can simply use a procedure spec, something along the lines of 50 cm lifts, each rolled by 4 passes of a ten-tonne vibratory smooth drum roller, with water applied on the fill as needed. Here is where a test fill is helpful by telling you whether the compaction has reached its limit with 4 passes or it needs 6(for example), whether additional passes would just break up the particles without accomplishing anything, whether a water hose helps, and things like that. Plus, it gives you an excuse to go out to the site and play with soil and machines. The test fill will need to be several lifts high, preferably two lanes wide, and long enough to replicate the actual placement techniques that will be used.

I don't think letting the stock pile age for a week is going to affect things much, unless there is a heavy rain that makes the slopes slough.

Depending on the shape of the particles, the angle of repose in stockpile may or may not give you a very good estimate of the angle of internal friction. If you put marbles in a drained triaxial test, they will give a pretty good φ, maybe 30 degrees or so. If you pour them out on the table, they roll away, angle of repose ~0. Rounded cobbles may segregate and roll when they are dumped onto a stock pile, and they can roll down fairly flat slopes if dislodged. On the other hand, if there is not a problem with rolling, and there is no capillarity or other "binder" present, the stock pile could give you a pretty good estimate of friction angle.

Bon chance!

 

[BigH]

google: "Friction Angles for sand, gravel and rockfill" by Michael Duncan (I have the article but not the URL)

There are some articles by Breitenbach on placement of rockfill, materials, etc. that would prove useful. The articles I have seen seem to point towards using smaller rockfill than I am used to placing. Typically, I would not use more than 40 degrees in design and perhaps less depending on type of rock, potential for breakdown, etc.

 

[ONENGINEER]

Sorry a typo in my previous comment about FoS = 3 instead of 1.5:

I then design my slope based on a factor of safety of 1.5. So when the materials are compacted at during the actual construction, I will certainly have a higher friction angle and as a result a factor of safety of greater than 1.5. Do you have any comment on this approach? Thank you again.

Thank you for your attention.

 

[fattdad]

For the infinite slope safety factor, it's F=tan(phi)/tan(beta), where phi is the friction angle and beta is the slope angle. If you have a friction angle of 36 degrees and a 2H:1V slope (i.e., 26.5 degrees), your safety factor would be, 1.5(ish). The friction angle is likely greater than 36 degrees.

It would really help this discussion if we knew what you were building, the slope heights and such.

Read the Duncan paper too! I'm sure you can get it through CGPR (Center for Geotechnical Practice and Research) at Virginia Tech.

f-d

¡papá gordo ain’t no madre flaca!

 

[GeoPaveTraffic]

dgillette

Interesting that you Berkley run 12-inch-diameter triax tests. My thiess, also a very long time ago, included triax testing on course coal refuse. I compared test results on 1.5-inch, 3-inch, and 12-inch-diameter remolded samples.

Prevailing thought, at least at that time, was that when you tested smaller material you got a lower friction angle due to removal of larger (presumably stronger) particles. My testing indicated that was not always the case; however, the differences were generally small +/- 1.5 degrees.

ONENGINEER

As has been pointed out/asked; we really need to know more about what you are building to be able to provide useful information.

Mike Lambert

 

[ONENGINEER]

The study is for a 10-15 m tailing retaining embankent for which I am prospecting. Meanwhile could someone refer me to any report on best practices to do a test fill at site.

Thanks everyone for the comments.

PS: Have some questions about GeoPaveTraffic seemeingly complicated research but that distances me from my main question here.

 

[GeoPaveTraffic]

For a relatively short dam such as this, I would not expect a lot of effort in determining the properties of a rock fill. As others have suggested, assume a friction angle in the 32 to 36 range and go from there. Note that this range assumes you have a fairly strong rock material, if you are dealing with a soft shale or other soft rock; a lower friction angle might be appropriate.

Mike Lambert

 

[molerat2210]

A large scale direct shear test of a cobble fill would be reasonable if it can demonstrate a higher friction angle for design. Hopefully, any test cost would be offset by savings for embankment borrow. Good luck finding a lab that can do it. It would probably need to be custom built.

 

[dgillette]

Agree with GPT. What would you do differently if you knew it was 34 or 36 or 38 degrees? Probably not enough to justify shear testing.

 

[cvg]

typically a test fill or test strip is done by the contractor just prior to completing the main embankment. It is generally done to make sure the correct size (large enough) equipment is used to achieve compaction. It would not normally be done to qualify the borrow source or to provide design data. The size of the strip needs to be large enough to operate equipment at full speed and should be at least 3 or 4 lifts. About all you will be able to monitor is loose lift thickness vs finished thickness and a visual indication of the amount of breakdown. You should also be able to observe shearing of the fill, laminations, or inadequate compaction if the equipment is too light.

 

[ONENGINEER]

[highlight #3465A4]Thank you again for the comments. I am convinced that lab testing is not vital for the design. I will run the design based on phi= 34 and 36 and 38 degrees, allowing a sensitivity analysis, and most likely choose 35 degree at the final. Now my question is the test FILL. What is the objective of the test pit? If the purpose is to optimise construction/compaction techniques/procedures and verify materials suitability for the intended structure, I understand it and have experienced it during road and embankment construction.

However, how can one estimate the friction angle during a test fill, as indicated by dgillette. The side slopes of the fill may not stand under the angle of repose for a few lifts because the side angles are affected by the method of compaction, the machinery used, labour interference etc. Or does dgillette mean that I should excavate within the compacted fill and measure the angle of repose from the excavation side slopes[/highlight]

 

[BigH]

ONENGINEER, I think that you are focusing too much on the "friction angle" of a rock fill. As most have stated, they would just presume (is a better word that assume) that it would be 36 or 38 deg and leave it at that. This is more based on experience and judgment as to what is reasonable (my number I gave you was for crushed rockfill - like from blasting). You "can't" go wrong with 36 deg assuming the rock fill is relatively clean - say no more than 5% passing the #200 sieve. A 10 to 15 m high embankment is not very high. See the link below on how to look at how to develop a rock fill compaction working procedure.

http://www.ausenco.com/uploads/pape...Compaction_Guidelines_for_Mine_Structures.pdf

Google " Breitenbach rockfill ".

I would make the rockfill test pad, say 8 m wide at the bottom and use a 1H:1V slope as you come up - you won't be putting on any more than 3 layers anyway (don't fret too much about "edges" - your compactor operator will not go to the edge anyway. You will want to try lifts of, say, 400 mm, 600 mm and perhaps even 800 mm thick and see what kind of compactive response you get - assuming you will be using a large 15 tonne or more vibratory roller. As noted by the paper above, paint some marks on the test pad layer and then take elevations of the mark after 2, 4, 6 and 8 single passes of the compactor; develop a pass vs elevation drop curve. Typically, we haven't gone more than 8 passes - so particles don't break down.


I am not sure what materials your embankment will retain (copper tailings?, uranium tailings?, etc) but you need to be cognizant, too, of other and likely more important aspects than friction angle of rock fill. You will need to address filters to the rockfill - presuming this might be a rockfill toe of a, say, till embankment.

 

[dgillette]

You get angle of repose from a stock pile, or the loose side slopes of a test fill, and friction angle is usually similar to that. That's all, unless rounded cobbles want to roll.

 

[oldestguy]

I have solved these questions by making an in-place direct shear test. This is easily done with a square open box made from steel square tubing. Insert in the top a thick plywood load section, roughened on the bottom and a steel plate on top. The downward normal load is from a back-hoe bucket resting on a wood post, just to keep it steady and a proving ring with dial gage and a house screw jack to apply the load. The lateral force is applied with a horizontally positioned hydraulic jack, preferably a cylinder fed by a separate pump with pressure gage(available at auto body shop supplies). A horizontal dial gage also is handy to observe the movement. It uses against a suitable resistance that can move if needed.

I've used this method many times to find on-site properties of waste piles, natural earth as well as compacted fill. Once you run a few, it is a simple operation. My shear box was one foot square, but you can make it any size to suite your material, as long as your loading gear can do the job.

For weak stuff, such as fly ash or waste bark, all you need is a 5 gallon bucket for the load and a spring loaded scale for the lateral loading. Weigh the bucket with the scale.