You might have problems with getting good representative measurements from the nuke gauge for density testing with oversize rock in the material being tested. One large rock that gets directly under the moisture reader on the bottom of the gauge will send the density to the sky, and be useless. Sounds like this could happen.

I would suggest a criteria based on SPT borings with a minimum N value instead of compaction.

Problem with criteria based on SPT borings is that the size of the stone is too large.  The spoon's opening is, what?, 1-3/8".  With stone sizes of 3 inches and 'abundance' greater than 3/4", I think that the SPT is out to lunch.  We have talked elsewhere about doing compaction on oversize.  Consensus seems that a 12" mould should be used - there apparently is a procedure out there (search for the thread - it is quite recent).  On larger stone aggregate (say fine rock) you basically compact with a heavy roller for a specified number of passes.  Probably, the best way to go about this is to do a trial roll and see the "density" vs number of passes.  You can do large-scale pits, lined with a thin film liner and water for volume - to get a density.

To Big H & dmoler...those are good points but that is not my focus right now.  I am trying to determined maximum density of 3" minus material in the lab and am leaning towards ASTM D 4253 which does use a 11" mold.  Once I get this max density, what percentage compaction should I recommend in the field for structural fill?  I wonder if there are any references out there.

I can't put my finger on any good reference.   There is something in http://www.utexas.edu/research/ctr/pdf_reports/1874_S.pdf
showing 4253 gives you lower maximum desities than other methods.  This is what I recall also.  So you may want to increase your field percentage density requirement.

Charlieroc,
The thread below discusses a little about what you're talking about.  See if this shows up:

thread261-37564

To make a long story short, in my testing I've found that the maximum density attained with ASTM 1557 is approximately equivalent to ASTM 4253 vibratory table test.  Therefore, if you want 95% of the 1557, I would go with 95% of the 4253.  Caveat:  My testing generally included SA-SR fine to medium sandy soil with little silt content (<3% by mass) a small amount of coarse sand, and no substantial gravel content.

I've seen similar testing results that show the same thing as CarlB describes, hence my caveat noted above.  Gradation and particle shape make very big differences in what type of compaction test will yield higher values-ASTM 4253 vs. ASTM 1557, for example.  Also refer to ASTM STP 529(?).  If I find that number is incorrect, I'll let you know the correct STP #.

In terms of testing, I agree with dmoler that you'll have trouble with a nuke gauge.  I agree with BigH that SPT is not the way to go too.  BigH also alluded to a "means and methods" specification for compaction, which could be valid for part of the testing program.  If you need some additional quantitative results of the compaction, a larger scale "lined pit" test as BigH offered, is probably a good way to go if you need to use this particular gradation as engineered fill.  
  

ASTM STP 523 rather.  There are some great papers in that collection about the topics of which test results in the greater unit weight and testing oversize material.

I was assuming (yeah, I know, dont do that) that there was not really large rock, but the max size was not given.

Since sampling of the geomaterial is not needed, only to determine if the placed and compacted fill will support the structure, why not use the SPT for that purpose, as long as the rock is not too big. I have done verification with similar projects with max aggregate of 1.5". I also had the fill sampled and it must pass a classification requirement(SP, ect.) with max aggregate size.

I really perfer the method that BigH mentioned above (# of passes correlated to density) but the only problem with that (maybe not a problem if someone pays for this) is a person has to stay at the site to monitor and verify construction in accordance with this. If they call after everything is done (unfortunately a usual practice) then I have used the SPT approach.

Nothing is perfect in construction, nor goes according to plan.

charlieroc,

As an update to the previous thread dealing with how to laboratory test your oversize material, I did some looking around for field compaction verification references. USACOE seems to like doing test fills (or at least they did at one time) for rockfill placement. Unfortunately it seems that few, if any of the USACOE reference papers listed in the EM bibliographies on this subject are available electronically.

You may need to contact the Corps to obtain a paper or other facsimile copy of these studies.

Hope this was helpful.

Jeff

Jeffrey T. Donville, PE
TTL Associates, Inc.
www.ttlassoc.com