Embankment and Base Compaction Control

The tests in this chapter only mean something when they are part of a control plan: build in thin, uniform lifts at the right moisture, compact them with suitable equipment, test them at a known frequency, and act on the results. This article describes that cycle for embankment, subgrade and unbound base. Why the subgrade matters to the pavement, and how it is prepared and stabilized, is covered in Subgrade Preparation for New Pavements and Subgrade.

Know the Material

Before work starts, each soil or base source on the project should have:

• a classification and Atterberg limits, so the inspector knows how it will behave when wet;
• a moisture–density (Proctor) curve, or a family of curves for variable sources, giving the reference density and optimum moisture; and
• for base, a gradation and any specified quality tests.

When the material in the cut or the pit changes, the reference changes too. Most field disputes over compaction come down to comparing a field density with the wrong Proctor curve (Christopher et al., 2006¹).

Lifts and Moisture

Lift thickness. Compaction energy from a roller falls off with depth, so material is placed and compacted in layers. Specifications commonly limit loose lifts to about 200 mm (8 inches) for embankment, and less for fine-grained soils with light equipment. Thicker lifts are sometimes allowed for granular material with heavy vibratory rollers, usually after a test section shows the full depth reaches density.

Moisture conditioning. Soil is easiest to compact near its optimum moisture content. Dry soil is wetted with a water truck and mixed in by discing or blading; wet soil is dried by discing and aeration, or by mixing in drier material. Many agencies set a moisture window, for example optimum minus 2 to plus 2 percent, in addition to the density requirement. Clays compacted slightly wet of optimum tend to swell less; compacted dry of optimum they are stiffer but can swell more when they later take on water.

Choosing Compaction Equipment

Different rollers suit different soils (see Compaction Equipment):

• Sheepsfoot and padfoot rollers knead fine-grained, cohesive soils and break up clods. The feet "walk out" of the lift as it densifies, which is a visible sign of progress. See sheepsfoot roller.
• Vibratory smooth-drum rollers suit granular soils and base, where vibration lets particles rearrange.
• Pneumatic-tired rollers are used to seal and knead the surface of a lift and for proof rolling.

Testing Frequency and Location

Specifications set a minimum frequency, typically by volume of embankment, by lane-length of each lift, or both. Locations should be chosen randomly within each lot of material so that the tests represent the whole lift, not just the spots that look worst or best. Additional tests are taken wherever the inspector sees soft areas, pumping, a change in material, or areas the rollers can't reach well, such as against structures and in utility trenches. Those added tests are for control; whether they count toward acceptance depends on the specification (Christopher et al., 2006¹).

Each test record should show the location (station, offset, lift or elevation), the material and the Proctor reference used, the wet density, the water content, the dry density, the percent compaction, and any oversize correction.

Proof Rolling

Density tests sample a few small spots. Proof rolling checks the whole surface. A heavily loaded truck or a pneumatic-tired proof roller is driven slowly over the completed subgrade or base while the inspector watches for rutting, pumping or deflection under the tires. Areas that move are marked, investigated (often by digging), and reworked. Proof rolling finds soft spots below the depth of a density test and is especially useful just before paving, after rain, or when the subgrade has been exposed to traffic.

Other Field Checks

Density by itself doesn't measure stiffness, which is what the pavement design assumed. Some agencies supplement or replace density testing with stiffness-based checks such as the dynamic cone penetrometer, the lightweight deflectometer, or roller-integrated measurements from intelligent compaction. These tests need their own targets, usually set on a control strip of the same material, because there is no single conversion between stiffness and percent compaction.

When a Test Fails

A failing test means the lift around it is suspect, not just the spot. A typical response is:

1. Check the test itself: the right Proctor reference, the oversize correction, the gauge standard count, and the moisture correction.
2. If the test stands, rework the area: adjust moisture, recompact, and retest at the same location and nearby.
3. Do not place the next lift over a failed area until it passes. Once covered, the only way back to it is to dig.

Standard Test Methods

• AASHTO T 99 and T 180: Moisture–density relations of soils (standard and modified Proctor)
• AASHTO T 310: In-place density and moisture content of soil by nuclear methods
• AASHTO T 191: Density of soil in place by the sand-cone method

@misc{pavint-embankment-and-base-compaction-control,
  author = {{Pavement Interactive}},
  title = {Embankment and Base Compaction Control},
  year = {2026},
  publisher = {HeadLight},
  howpublished = {\url{https://pavementinteractive.org/reference-desk/construction/site-preparation/embankment-and-base-compaction-control/}},
  note = {Accessed [date]}
}