We use crushed stone or gravel as mud-mats in footings in clayey soils similar to your case.  We normally recommend a No. 57 gradation crushed stone or gravel, and a thickness of 8 to 10 inches.  If thicker layers are used, we recommend that each 12-inch thickness of stone backfill be densified by making 3 or 4 passes of a walk-behind vibratory sled or plate type compactor over the stone surface.  Density testing is not performed.  The stone is self-seating and much firmer than sand backfill.  You have to be careful when using stone in sandy profiles as surrounding sand can ravel into voids in the stone unlesss a geotextile fabric separator is used.

Thanks for your recommendations.  We have specified a geotextile under the crushed stone to (1) maintain the stone during placement (thanks for your input!) and (2) to help minimize the migration of fines.

if you still want to be able to measure and document the amount of compaction, you could use the relative density method - see ASTM D4253-93(1996) and ASTM D4254-91(1996) which uses a vibrating table to measure the maximum index density and the minimum density of the material and then determine the relative density of the in-place material.  

This test is specifically designed for testing cohesionless, free-draining soils where Proctor methods cannot be used.  The Bureau of Reclamation (Design of Small Dams) recommends a minimum relative density of 70%.

Rel. density (%) = [Dmax x (Dip-Dmin)]/[Dip x(Dmax-Dmin)]

Dmax = maximum density
Dmin = minimum density
Dip  = in place density


 Chuck
 cgopperton@stantec.com
 http://www.stantec.com/
 

I agree completely with KAM.  If there is water present at time of placement, your only options are using fill without fines, placing using CDF, a type of concrete.  These are the only material types that will not be comprimised with oversaturation.  
Crushed stone without fines, or 'railroad ballast' (the stuff that supports train traffic), will be very much less expensive.  I also agree with the GT fabric.  I think it will help preserve the structural integrity of the ballast, as the migration of silt/clay will decreace the friction between the angular grains, allowing them to settle.    

I'm building a ranch home. I had to fill up the footing form bottoms with 3/4 gravel about 16" deep. I'm planing to compact with a plate compactor. Should I use a plate compactor?. Or perhaps some other method considering the thickness of the gravel.

GravelMan...plate compactor should be fine for your application.

note KAMs recommentation above and do it in two lifts

With the 16" thickness of the stone fill, I would use a "jumping jack" style compactor rather than the plate compactor.  It is too late now, but I would also recomend compacting the subgrade prior to the placement of the stone.