Soil Stiffness Gauge for Soil Compaction Control

As with most construction today, the emphasis on cost control and quality control of soil is prompting the implementation of mechanistic designs, performance specifications, and contractor warranties. The Federal Highway Administration’s cooperative development of a soil stiffness gauge will enable the validation of design models, the development of performance specifications, and contractor process control for compacted soil structures.

Benefits of statistical process control.

Background
Compacted soil is an essential element in the construction of highways, airports, buildings, sewers, and bridges. Even though soil density is not the most desired engineering property, it is used almost exclusively by the transportation industry to specify, estimate, measure, and control soil compaction. This practice was adopted many years ago because soil density can be easily determined via weight and volume measurements.

Textbook authors Holtz and Kovacs state, "Since the objective of compaction is to stabilize soils and improve their engineering properties, it is important to keep in mind the desired engineering properties of the fill, not just its dry density and water content. This point is often lost in earthwork construction control.” ¹

When soil is compacted for pavements, pipe bedding and backfill, and foundations, the desired engineering properties are the soil modulus or soil stiffness.

State departments of transportation and contractors suggest that the present methods for measuring density are slow, labor-intensive, dangerous, and/or of uncertain accuracy. Hence, construction sites are often undersampled, causing inadequate compaction to go undetected or feedback to be provided too late for the cost-effective correction of problems. Sometimes, the opposite is true. Designers are encouraged to overspecify to allow for the significant variability of the finished product, and contractors are encouraged to overcompact to ensure acceptance and avoid rework. All of which means added cost to the owner.

To eliminate overspecification and overcompaction, statistical quality control can be implemented on civil works projects. The benefit of better quality control is illustrated in figure 1. The normal distribution curve, labeled “Typical Soil Data,” is for 140 measurements taken in sandy soil on an interceptor sewer project. The mean modulus is 67.7 megapascals (MPa) (9,830 pounds per square inch), and the standard deviation is 12.9 MPa (1,872 psi); therefore, the coefficient of variation is about 19 percent. About 95 percent of the measurements are greater than the hypothetical “Design Modulus” of 46.5 MPa (6,750 psi) for the pipe bedding.

Assume that, by instituting a measurement and quality control program, the standard deviation could be reduced to 8.8 MPa (1,275 psi). Then, it would be possible for the contractor to use less compactive effort (number of compactor passes), reducing the average soil modulus to 61.0 MPa (8,850 psi) while maintaining the passing tests at the 95-percent level and saving cost.

Using the Soil Stiffness Gauge

The typical sequence of operations to make a good measurement with SSG are as follows:

Clean ring foot of soil.
Turn on SSG. (Press "On" button.)
Prepare the surface to be tested.
1. Smooth the surface with the side of your boot.
2. Coarse aggregate or stiff clay may require sand to be sprinkled.
3. Ensure the gauge has clearance on the side.
Seat the foot. (Place the ring foot on the soil, and twist the gauge 90 degrees back and forth two to five times using minimal to about 15 pounds of force, depending upon the granularity and softness of the soil.
Enter Data. (Enter target stiffness, Poisson's ratio, and/or site identifiers from scrolled list via SSG display.)
Take the measurement. (Press "Meas" button. SSG will measure site noise and stiffness as a function of frequency. The gauge will display average stiffness, lb/in (MN/m) or modulus, psi (MPa) or percentage of target. If construction noise is present, try to take the test at a distance greater than 25 meters from the operating equipment.)
Store Data. (Press "Save" button. More than 200 measurements are displayed in the operational mode, and 10 measurements of complex, frequency-dependent components are displayed in the research mode.)
Remove SSG from the soil. (Ensure that 50 to 60 percent of the ring foot is in contact with the soil.)
Transfer data. (Standard infrared link, same as used in many nuclear density gauges.)

In addition to the time and cost advantages, a portable soil stiffness gauge that is quick and easy to use will save lives and reduce exposure to injuries by enabling the technician to conduct each test rapidly. Many technicians, preoccupied with performing a nuclear density test or other quality assurance method, failed to hear or see approaching heavy construction vehicles and were run over.

In one incident in which the technician was killed, the U.S. Nuclear Regulatory Commission sent inspectors to investigate because the gauge was crushed, exposing the gauge’s radioactive elements — cesium and americium. The potential for accidents involving radioactivity add significantly to the already tedious safety precautions and record-keeping. A non-nuclear method is in great demand.

Soil Stiffness Gauge
In response to this need for a faster, cheaper, safer, and more accurate compaction testing device, the Federal Highway Administration (FHWA) joined with the U.S. Department of Defense’s Advanced Research Programs Administration (ARPA) to co-sponsor a study to investigate the possible use of military technology to solve this problem. As part of the defense reinvestment initiatives and using funds from the Technology Reinvestment Project, ARPA authorized FHWA researchers to supervise the redesign of a military device that used acoustic and seismic detectors to locate buried land mines.

FHWA’s partners in this cooperative research and development project were Humboldt Manufacturing Co. of Chicago, Ill.; Bolt, Beranek & Newman (BBN) of Cambridge, Mass.; and CNA Consulting Engineers of Minneapolis, Minn.

The result of this cooperative development is the Soil Stiffness Gauge (SSG), shown in figure 2. SSG measures the in-place stiffness of compacted soil at the rate of about one test per minute. SSG weighs about 11.4 kilograms (kg), is 28 centimeters (cm) in diameter, is 25.4 cm tall, and rests on the soil surface via a ring–shaped foot.

The prototype model was modified to make a soil stiffness gauge that is portable, lightweight, and safe to use. Resting on the soil surface, SSG produces a vibrating force that is measured by sensors that record the force and displacement-time history of the foot.

The device has been “beta-tested” by FHWA and several state highway agencies. Thousands of soil stiffness measurements have been successfully made at highway embankment sites and pipe backfill sites on sand, clay, and sandy loam soils. When converted to density values using correlation charts, these measurements are within 5 percent of measurements made with a nuclear density gauge.

Production devices are being made for further evaluation at sites representing a cross section of U.S. applications and soils. Future models will include on-board moisture measurement instruments and a global positioning system.

Method
The stiffness is the ratio of the force to displacement: K=P/d. SSG produces soil stress and strain levels common for pavement, bedding, and foundation applications. It is a practical, dynamic equivalent to a plate load test. Figure 3 compares SSG to a plate load test. In both cases, a force P is applied to the soil via a plate or ring. The soil deflects an amount d, which is proportional to the foot geometry, Young’s modulus, and Poisson’s ratio of the soil. The soil stiffness, as measured by SSG, also relates to shear modulus, void ratio, and density. Figure 4 presents the basic relationship.²

Technology
Plate load tests are commonly conducted by jacking against a large, loaded truck (to provide a reaction to P), while taking great care to measure the deflection. Large forces are necessary to produce enough deflection to measure. SSG uses technology borrowed from the defense industry to measure very small deflections, allowing much smaller loads. SSG does not measure the deflection resulting from the SSG weight. Rather, SSG vibrates, producing small changes in P that produce small deflections. To filter out the deflections resulting from equipment operating nearby, SSG measures over a frequency range. Figure 5 is a schematic of SSG showing the major internal components. Not shown are the D–cell batteries that power it. The foot bears directly on the soil and supports the weight of the device via several rubber isolators. Also attached to the foot are the shaker that drives the foot and sensors that measure the force and displacement-time history of the foot.

Table 1 — Organizations Interested in SSG Use or Evaluation
Interested Organizations	States Represented
Engineers	Ala., Ark., Calif., Ill., Ind., Ky., La., Md., Mich., Minn., Neb., Ohio, Texas
Construction suppliers	Minn., Ohio, Texas
Universities	Alaska, Ga., Iowa, Minn., Mo., Utah
Departments of Transportation	Ala., Calif., Del., Ga., Ill., La., Mich., N.C., N.J., Pa., Va.
Contractors	Ala., Calif., Colo., Minn., Pa., Wis.

Calibration
SSG is calibrated via the force-to-displacement produced by moving a known mass. The value of the mass is precisely known and is less susceptible to change than a reference elastomeric pad or soil sample. Calibration in the lab or office requires a mass of sufficient size to represent a typical range of soil stiffness — e.g., 10 kg represents approximately 4 meganewtons per meter (MN/m) at 100 hertz (Hz) and 16 MN/m at 200 Hz. The mass is approximately the same shape as SSG’s foot and is rigidly bolted to the foot during calibration. SSG, with the mass attached, is supported in the upright position off of a rigid floor by a very compliant fixture. The fixture is sufficiently compliant so that the mass is effectively unrestrained in the measurement frequency band.

The calibration is performed by pressing the “Cal” button on the SSG display. SSG compares the measured effective stiffness of the mass against what is expected. A simple frequency-dependent correction is recorded. Note: During factory calibration, a software correction is made to account for the effect that the mass of the foot has on stiffness measurements.

Development Approach
SSG was developed from a comprehensive application, market, and technology knowledge base. The development began with an FHWA contract to BBN in cooperation with CNA. Similar to the Strategic Highway Research Program, the purpose of this contract was to infuse new technology into the transportation industry. Specifically, technology proven in the detection of nonmetallic land mines for the U.S. Army was to be transitioned to the application of soil compaction evaluation in the field. Successful proof-of-principle demonstrations were performed by BBN on a significant range of soil types and conditions. This success prompted the consortium led by BBN to recruit Humboldt Manufacturing Co. to commercialize SSG.

Prototype gauges have been manufactured, and they have been or are being evaluated by FHWA; the departments of transportation of Minnesota, New York, and Texas; and the University of Massachusetts. These field evaluations are quantifying how well SSG performs in practice on a broad range of soils, applications, and conditions.

The plan for the next construction season calls for sending 12 to 24 gauges to sites representing a cross section of U.S. applications and soils for the following purposes:

To begin characterizing local soil stiffness to facilitate the current move towards more cost-effective design and specification of highways and buried structures.
To demonstrate the cost and quality benefits of controlling soil stiffness/modulus in the field.

Numerous organizations in various states have expressed an interest in evaluating or using SSG for these purposes. Discussions with many of them have been initiated to coordinate SSG use during the upcoming construction season. Table 1 illustrates the diversity of these organizations.

The consortium has initiated the process for standardizing the method embodied in SSG with both the American Association of State Highway and Transportation Officials (AASHTO) and the American Society for Testing and Materials (ASTM).

In the coming year, work will be initiated to bring the following capabilities to SSG:

Progress
A good example of SSG performance to date is the Minnesota Department of Transportation’s (MnDOT) trunk highway 610 project in Brooklyn Park, Minn. This work measures the soil properties (modulus, stiffness, water content, density, etc.) of reinforced concrete pipe (RCP) bedding and backfill and of the response of RCP to soil overfill. About 1006 meters of arch and round RCP have been placed to carry water from a wetlands mitigation site east to the interchange of trunk highway 610 and state Route 252.

At one MnDOT site (figure 6), more than 1,300 soil stiffness measurements have been made successfully using SSG. Testing was conducted on sand, clay, sandy loam, and mixtures of these soils in an in situ and compacted state. Soil stiffness and moduli at this site range from 0.5 to 22.1 MN/m (3,000 to 126,000 lb/in) and from 4.1 to 192.9 MPa (600 to 28,000 psi). The variability in soil stiffness (figure 7) along the bedding was surprising even to MnDOT.

This site illustrated well the need for specified performance as opposed to specified methods of compaction. It also illustrated how SSG could enable the statistical quality control of specified performance. The relationship between stiffness and density (figure 8) also proved to be well-behaved.

Figure 9 illustrates how soil stiffness measurements can be used in controlling road construction. In this example, an actual compacted roadbed has been excavated; a drainage pipe installed; and soil reinstated, but not fully compacted. A medium-size vibrating roller was used to complete the compaction as stiffness measurements were made with SSG. The figure shows that the reinstated soil required additional compaction and that the undisturbed soil, adjacent to the trench, was overcompacted. Hence, the pavement added later would experience stress concentration and, possibly, premature failure because of the two stiff trench sides with softer soil between. Companion density measurements were not so revealing.

The SSG evaluations to date suggest the measurement accuracy illustrated in table 2.

SSG Performance Specifications
The version of SSG that is currently under development will be available for use early this coming construction season and is designed to perform as specified below:
Soil Stiffness
Stiffness	3 MN/m (17k lbf) to 22.1 MN (126k lbf)
Modulus	26.2 MPa (3.8k lbf/in²) to 193 MPa (28k lbf/in²)
Measurement Accuracy (typ., % of absolute)	<±5%
Depth of Measurement	10.2 to 15.2 cm (4 to 6 in) (a function of foot outside diameter and SSG mass)
Laboratory Calibration
Accuracy (% of actual mass)	<±2%
Range (effective)	4 to 16 MN/m (22.8 to 91.4 lb/in)
Electrical
Power Source	6 D-size disposable cells
Battery Life	Sufficient for 6,700 measurements
Mechanical
External Materials	Aluminum case & foot
Rubber isolators	Aluminum case & foot
Vibration	0.1 in @ 125 Hz
Level re Vertical	±5°
Operating Temp.	0°C to 38°C (ambient)
Storage Temperature	-20°C to 50°C
Humidity	98%, without condensation
Gauge Size (without handle)	28 cm (11 in) OD
Gauge Height (without handle)	25.4 cm (10 in) H
Weight	11.4 kg (~ 25 lb)
Shipping Weight (with case)	16 kg (~ 35 lb)
Standard Accessories
Transit Case
Infrared Data Link	(PC side & rudimentary software)
Optional Accessories
Laboratory Calibration	10 kg (~22 lb) mass
Mass & Fixture
The performance specified above Is preliminary and subject to change.

Conclusion
Soil stiffness is the desired engineering property when soil is compacted for construction projects. However, until now, engineers have used soil density as a measure of soil compaction because there was no easy method for measuring soil stiffness. Measuring soil density is slow, labor-intensive, and potentially dangerous. The new lightweight, portable soil stiffness gauge not only provides a means to measure the desired engineering property, but it is faster, cheaper, safer, and more accurate than the current standard methods.

Scott Fiedler is a product manager for Humboldt Manufacturing Co. in Norridge, Ill.

Charles Nelson is president of CNA Consulting Engineers in Minneapolis, Minn.

E. Frank Berkman is vice president of BBN Systems and Technologies in Cambridge, Mass.

Al DiMillio is the federal program manager for geotechnical engineering research for the Federal Highway Administration at the Turner-Fairbank Highway Research Center in McLean, Va.