Percent Compaction

[moe333]

Hi Everyone:
This is not a homework problem! Sorry but this is a little long-winded:
I have Modified Compaction test results (6" mold) and gradation for a SM/SC material that has about 10% retained on 3/4" sieve and 30% retained on the 3/8".
I also have field density tests (some rock corrected some not) so I know what the average compacted dry density and percent compaction is in the field.
I am doing some non-standard settlement/collapse studies where I have collected bag samples of the compacted material.
I want to remold the -#4 portion of the sample in a standard consol ring and do some loading and wetting increments and want to place it at a density and moisture so that it models the compacted field conditions (material with 10% retained on 3/4" sieve).

My assumption (right or wrong) is that the coarse material is "flotaing" in the finer matrix and does not significantly affect the compression characteristics. Given this assumption, I want to compact my -#4 sample to the same percent compaction that the -#4 portion of the field compacted material.
I am thinking of running a compaction test on my -#4 sample and remold it to the average percent field compaction(material with 20% retained on 3/4" sieve). I don't want to base it on dry density since I believe small differences in gradation can affect percent compaction/compressibility.
I'm wondering if there is a way to estimate what the percent compaction of the -#4 portion of the field material is analytically, using oversize corrections.
For example; if the field compaction is say 95% of modified, does this mean that the -#4 portion is compacted to 95% as well?
I may be overthinking this, but just wanted to see if anyone had any thoughts. (Note that I can't do any field testing at this point)
Thanks!

 

[Ron]

Are you sure you have SM/SC material? Based on your gradation listed, it is likely you have a GM/GC material. Considering that, if you use a consol ring with a height less than 3 or 4 times your maximum particle size you'll have results adversely affected. Maybe you can double to consol rings to compensate for that.

I don't see anything wrong with your premise, though I'm not sure it will be either representative or accurate.

 

[fattdad]

If I understand the OP correctly, here's how I'd approach the problem.

Let's say you have a field density test that shows 120 pcf at 12 percent moisture.
Let's say the bulk sample from that location has 30 percent material coarser than the No. 4 sieve and the specific gravity of that plus 4 material is 2.68.
Let's say you run a proctor on a bulk sample of that minus -4 material and you get 115 pcf at 18 percent moisture.

What's the relative compaction of the field density test?

So, you have 30 percent of the sample that shows a dry unit weight of 167.23 (the unit weight of the plus 4 material) and 70 percent of the sample that has a different unit weight. . .

120 pcf= 0.3(2.68*62.4)+(0.7*Z), where Z is the dry unit weight of the -4 materials.

Z=99.76 pcf

So, the relative compaction of the -4 material is 99.76/115=86.67%

Prepare your minus-4 sample to 86.7 RC and run your consolidation test (bearing in mind the limitations on size in the odometer and such as Ron mentioned).

f-d

¡papá gordo ain’t no madre flaca!

 

[GeoPaveTraffic]

fattdad's approach seems as resonable as any, though I'm not sure what the results will really tell you about the settlement potenital of the original fill. Doesn't appear that collapse should be a problem.

If you follow through with testing the -#4 material, I expect that you will greately over estimate the amount of settlement. I say this due to the amount of gravel that will not consolidate at all. Even if the gravel does not have point to point contact, there is less "soil" there to consolidate.

Maybe with more information, someone can give you a better/different approach to solve the original problem.

Good Luck.

Mike Lambert

 

[moe333]

Thanks guys,
I am also skeptical that this program will provide anything very meaningful, but I am charged with implementing it. The results will need to be tempered with simplified estimates. I think the lab testing will overestimate the settlement and response to wetting (I shouldn't have said collapse initially). Fattdad, I follow your logic and that may be the way to go although there are also some oversize correction (lab and field) issues I need to work out. It seems I would want to have a reltively higher percent compaction of the -#4 to account for loss of oversize material. The material is SM/SC (~60% passing #4).

 

[moe333]

Fattdad, looking at your example a little closer, wouldn't the relative compaction of the -#4 material be 99.76/120=83.1%?
Also, there doesn't seem to be a direct connection between the percent compaction from the field density and the percent compaction of the remolded sample. It is connected more to the density.

 

[fattdad]

Moe,

I don't think so. . .

The 120 measured in the field had some amount of plus 4 material. If you assume that the plus 4 stuff is all at 100 percent compaction (i.e., relative compaction of rock fragments is meaningless), then you'd compare the compaction of the calculated density of the -4 material and compare that to the lab density of the -4 material.

This is all based on first principals. ASTM's method is very similar, but you take the lab curve and "correct" the curve for the percent oversize material (i.e. redraw the curve). I like "my" way better.

f-d

¡papá gordo ain’t no madre flaca!