The Proctor Test

A little water helps soil pack tighter; too much pushes the grains apart.

Before this Lesson

Water takes up room. Stir more of it into a can of soil, and you'd expect less room for the grains. Now run the experiment, with the same can, the same hammer, and the same number of blows, changing only the water. At 9% water, one soil packs about 102 lb of grains into each cubic foot. At 15%, about 112 lb. The added water somehow made room. Keep pouring and the count turns around: by 21%, it's back to about 104 lb. That hump tells a contractor how wet to make a fill and how tightly it can be packed.

A hammer squeezes out air

Hit damp soil with a hammer and it gets denser, so something must have left. Not the grains. Not the water either, at least not in the instant of a blow, because it can't get out that fast. That leaves the air. In the phase box from Unit Weights of Soil, each blow pushes the top of the box down: the solids and the water stay as they were, and only the air layer gets thinner.

That tells you how to keep score. What you care about is how many grains you got into the space, and that's the dry unit weight, γD=Ws/V\gamma_D = W_s / V in the Handbook's Phase Relationships. The moist unit weight won't do, since it also counts the water. Pour in more water and it can climb even while the grains spread apart.

The Proctor test

The Proctor test runs that opening experiment in a lab. Mix the soil to a target water content and compact it into a steel mold of known volume, in layers, with a set number of blows from a set hammer. The standard recipe is a 5.5 lb hammer dropped 12 in, with 25 blows on each of three layers, in a 1/30 ft³ mold. Take off the collar, strike the soil off flush with the top of the mold, weigh the mold and soil together, and oven-dry a scoop of the soil to measure its water content. That's one point. Then add water and repeat until the weights have clearly started to fall.

Drop 12 in Hammer, 5.5 lb 3 layers, 25 blows each Collar Mold, V = 1/30 ft3 Base plate
The standard Proctor mold and hammer

The Handbook has no Proctor test, compaction curve, or zero-air-voids line, so everything you need for them is here. Reducing a point takes three moves:

  1. Subtract the mold, to get the weight of the moist soil WW.
  2. Divide by the mold's volume, for the moist unit weight: γ=W/V\gamma = W / V.
  3. Take out the water with the conversion from Unit Weights of Soil: γD=γ/(1+ω)\gamma_D = \gamma / (1 + \omega), with ω\omega as a decimal.

When a question gives the moist unit weights, start at the third move. In SI, the scale reads a mass, so get the weight from W=mgW = mg, in the Handbook's Concept of Weight, with g=9.81 m/s2g = 9.81\ \text{m/s}^2 (its Fundamental Constants table prints 9.807, which won't change your answer). The same 1/30 ft³ mold is 944 cm³, or 0.000944 m³.

Plot dry unit weight against water content and the points trace a hump, the compaction curve. Its top gives the two numbers the test exists to find. The height of the peak is the maximum dry unit weight, γD max\gamma_{D\,max}. The water content straight below the peak is the optimum moisture content, ωopt\omega_{opt}, where "moisture" means the same as "water." With a table of points, take both from the highest dry value. Field specifications are written against these two numbers, which is the subject of Field Compaction Control.

Worked Example Five points, one peak
1 of 3

A standard Proctor test uses a 1/30 ft³ mold that weighs 9.40 lb empty. Find the maximum dry unit weight and the optimum moisture content.

Proctor test data
Water content (%)Mold + soil (lb)
913.10
1213.47
1513.69
1813.71
2113.61

Why the curve rises and falls

Here are those five points, plotted for a soil with Gs=2.70G_s = 2.70.

6 9 12 15 18 21 24 100 110 120 130 140 Water content, ω (%) Dry unit weight, γD (lb/ft3) Zero-air-voids line, Gs = 2.70 γD max = 111.9 lb/ft3 ωopt = 15% Compaction curve
The compaction curve from the Worked Example

Why a hump? The mold's volume never changes, so grains, water, and air are all competing for the same space. Scale the mold up to a cubic foot and work out each one's share: the solids take γD/(GsγW)\gamma_D / (G_s\gamma_W), the water takes ωγD/γW\omega\gamma_D / \gamma_W, and the air gets what's left. Here are three of the points.

Air 0.25 Water 0.15 Solids 0.60 ω = 9% γD = 101.8 lb/ft3 Air 0.07 Water 0.27 Solids 0.66 ω = 15% γD = 111.9 lb/ft3 Air 0.03 Water 0.35 Solids 0.62 ω = 21% γD = 104.4 lb/ft3 Volumes in ft3 per 1 ft3 of mold
How grains, water, and air share the mold at three water contents

Go from 9% to 15%. The water's share grows by 0.12 ft³, which looks like bad news for the grains. But the air's share falls by 0.18 ft³, more than the water took, and the grains get the difference: 0.06 ft³ more. Why does the air give way so easily? At 9%, the grains grip each other, and the hammer's energy goes into fighting that friction. Wet them a little and they slide past one another, so the same blows pack more of them in. The water is a lubricant.

Now go from 15% to 21%. There's only 0.07 ft³ of air left, and some of it is trapped in pockets the hammer can't reach. Another 0.08 ft³ of water comes in, the air gives up just 0.04 ft³, and the grains have to surrender the other 0.04 ft³. The hammer can't squeeze the water out of the way, so past the peak, each added bit of water sits in space the grains could have taken.

The optimum is the crossover between those two stories. It isn't where the soil saturates, though. At its 15% peak, this soil still has air in its voids, with S=80%S = 80\%.

The ceiling above the curve

The thin line above the curve in the first graph is the zero-air-voids ceiling from Unit Weights of Soil. At each water content, it's the dry unit weight with every bit of air gone, which means S=1S = 1:

γD=GsγW1+ωGs\gamma_D = \frac{G_s\gamma_W}{1 + \omega G_s}

It slopes down because more water leaves less room for grains. At 15%, this soil's ceiling is 119.9 lb/ft³, and the test reached 111.9. Compaction always leaves some air trapped, so the curve stays below the line, running close alongside it on the wet side, where little air remains. It can't cross, either. A point above the line would have SS above 100%, more water than its voids can hold. So when a computed point lands above the line, suspect the arithmetic.

Checkpoint
1 of 3
Water content, ω Dry unit weight, γD P Q Zero-air-voids line
A compaction curve with two marked points

Point P is the top of the curve, and Q sits straight above it on the zero-air-voids line. Which statements are correct?

A.
The soil at P has no air left in its voids.
B.
P's height is the maximum dry unit weight, and the water content straight below P is the optimum moisture content.
C.
Q is the dry unit weight the soil would have at the same water content with all the air gone.
D.
With the same effort, compacting at a higher water content would give a dry unit weight above P.

Hit it harder

Heavy rollers can compact soil harder than the standard recipe does, so there's a second recipe, the modified Proctor test. Where the standard test drops 5.5 lb from 12 in onto three layers, the modified test drops 10 lb from 18 in onto five, with the same 25 blows per layer. In the same size mold, that's about 4.5 times the energy.

What does that do to the curve? You can reason it out. The ceiling can't move, because the soil hasn't changed and there's no hammer in that formula. Harder blows do push the grains closer at a given water content, so the curve lifts toward the ceiling. And the peak, as you saw, comes when the voids are mostly full. Se=ωGsSe = \omega G_s says how much water that takes for voids of a given size. Now predict.

Checkpoint

A standard Proctor test on a soil with Gs=2.70G_s = 2.70 peaks at 112 lb/ft³ and 15%. Which result could the modified test give for the same soil?

A.
104 lb/ft³ at 18%
B.
122 lb/ft³ at 12%
C.
122 lb/ft³ at 18%
D.
104 lb/ft³ at 12%

Recap

  • Compaction expels air, not water, so it's scored by the dry unit weight, which counts only the grains.
  • For each Proctor point, subtract the mold, divide by the volume, then take out the water: γD=γ/(1+ω)\gamma_D = \gamma / (1 + \omega). The peak of the dry values gives γD max\gamma_{D\,max} and ωopt\omega_{opt}. The highest moist value doesn't.
  • On the way up, water lubricates the grains and the air gives way. Past the optimum, little air is left, so added water takes space from the grains.
  • The zero-air-voids line, GsγW/(1+ωGs)G_s\gamma_W / (1 + \omega G_s), is a ceiling. The curve stays below it, and a point above it is an error.
  • Modified effort raises the maximum dry unit weight and lowers the optimum moisture content.