Proctor Testing for Course Soils 1

[martin888888]

We have a site design for a proctor to be taken per the ASTM D-1557 method. The gradation report came back showing soils that are too course to run the proctor.

Is there a method to perform a proctor with course soils? Is there a way to write a specification to control the soils so a sensible / cost effective proctor can be done?

 

[fattdad]

Read the method, there is a procedure to remove the oversized material, run the Proctor and then adjust the curve according to the weight percentage of oversize material. When you go into the field, you have to ascertain whether the fill at the location of the test also has a similar amount of oversize as the test sample and if not make a field adjustment.

It's all weight and volume relationship stuff (and simple math).

f-d

¡papá gordo ain’t no madre flaca!

 

[martin888888]

The procedure actually says that over 30% the proctor can not be done with the ASTM D-1557 method and that very few labs are set up to do one. It goes on to say that there are other more expensive methods that can be used.

 

[HSIII]

What did your sieve specify as the actual percentage retained on the 3/4" size? Did you run this sieve washed or dry?

If you ran it dry, I would try running it again iver a washed, sometimes fines can clump up with other particles and cause a 1/2" to be retained on a 3/4". Not often, but it can happen, especially if it is a silty material.

Unfortunately, according to D1557, there is no other ASTM which can accomodate for more than 30% retained on the 3/4" sieve.

Another way to go about this could be to call for processed 1/2" or 3/4" material in accordance with certain DOT specs. You would get a more granualted material, it's easy to have a proctor run on the material, and the overall cost really wouldn't be that much greater, at least in my experience.

 

[martin888888]

Thanks, we were right at 30%. I am not a geotech and have never run the tests. Is it normal to do a dry test?

WSDOT has a test method #606 that is used to determine density and mositure content for materials over 30% that we are going to use.

You dont think that would be expensive to spec that material can not be greater then 1/2" or 3/4"?

 

[HSIII]

Like I said, in my experience, calling the spec to be 1/2" processed instead of 3/4" natural, it was actually cheaper to get the 1/2" processed then to run continuous sieves until a suitable natural material was discovered (One time it took in excess of a dozen sieves to be performed before the material was accepted).

Dry sieve vs. washed sieve is really an opinion stance. Certain engineers like the washed sieve so they can determine the actual amount of fines in the material, but an added bonus is that it provides a "cleaner" sieve of the material (and just may knock your 3/4" retained to under 30%). I would recommend reading up on ASTM D422 for washed sieves.

As an added note, you may just be able to argue your case to use D1557 yet. If you read section 5.3.1 of D1557 closely, it says "soils containing more than 30% oversize fraction...are a problem." So if you're right at 30.00%, you could try to argue that you can use D1557 to determine your proctor.

Finally, you could try D4914 or D5030, which could determine "field" dry unit weights. However, the downside to these two methods are they're not widely used and they're rather costly.

 

[martin888888]

maybe 1/2" or 3/4" for proccessed roadway base material or something similar to that. But for fill material, that small of material is a bit excessive.

Thanks for the info, very helpful.

 

[Ron]

Martin...be careful which in-place density method you choose to determine density. The drive sleeve is not appropriate and nuke can be questionable, both because of material displacement. Consider a few sand cone tests to correlate if you use nuke gage....otherwise, use sand cone.

For your question regarding test method, yes...the process is called scalping where the Proctor sample is run dry, usually by hand, through a 3/4" scalping screen. The retained material is weighed to get the proportion.

For a grain size distribution, it is usually an extra procedure to run a large enough sample through a short nest of sieves for a wash 200 test, then do full distribution on the sample after drying.

 

[BigH]

The OP is not very specific on what his gradation actually is. The normal sand cone or rubber balloon might be "too small" - it might be better to dig a controlled hole of larger size - maybe 2 ft by 2 ft and depth of layer - then line it with a thin malleable plastic sheet and use water to determine the volume.

 

[BigH]

Sorry this thought came just after I hit the SP icon - - if this is more like small rockfill, then the best way might just to do a few test strips and measure the "settlement" after each pass to determine the suitable number of passes to use. If you google compaction of rock fill - there is a good series of papers that explain compaction of rock fill - I'm not at my work computer now so can't get the references in short order.

 

[Ron]

BigH...good point. I was thinking that with 30 percent retained on the 3/4, then 70 percent passes...likely enough to fill voids...although if most of the 70% is still retained on the #4, he could have the problems you noted.

 

[cvg]

seems like you hired a geotech to run tests on your material, why isn't he advising you on the appropriate methods for your specification? Since he already has all the data and local experience, he is in a much better position to do this than anyone on this forum...

 

[martin888888]

Your right, in a normal case thats how I handle it. In this case I got into the back end of this and just trying to get some feed back as I am a site civil and not a geotech.

I think we are diving too far into this with the balloon methods and whatnot. Its a simple 4' max. pit run fill for an asphalt parking structure. Specs call for maximum particle size of 6" with max. lift of 1'. Would not make sense to ask them to make the particle size 3/4", nor is it good engineering practice.

I have not run into the situation much where the fill material comes back too coarse to run the D1557 method so I was just looking for feedback on how this is normally handled. I know one way I have heard of it being handled before is, as the lifts are completed to do a proof roll without testing. This can be subjective and takes a very experience inspector and contractor. Personally, I would feel uncomfortable with this method if buildings were going to be place on top of this fill.

 

[bb29510]

If the material is too coarse too run as a proctor then all you can really do is a unit weight, but the problem also arise that it cant be tested as compaction due to the voids in the material. Some fines are needed for compaction. If fines are there then a proctor is possible.

ICC special inspector, Structural masonry, Reinforced concrete.

 

[BigH]

@OP - if you think that it is going "too far" talking about balloon methods, etc for a simple 4' fill, then why are you talking about proctors anyway? Let's say that you could have actually done a proctor that is representative . . . how would you measure the field density to compare it to? Either you test the fill for compaction via standard methods or, as was indicated, you develop a compaction criteria by test method. See

http://www.geoengineer.org/rockfill4.html

and then go down to "rockfill3", "rockfill2" etc. This gives good information on compaction of larger sized materials.