U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

 
REPORT
This report is an archived publication and may contain dated technical, contact, and link information
Back to Publication List        
Publication Number:  FHWA-HRT-13-046    Date:  October 2013
Publication Number: FHWA-HRT-13-046
Date: October 2013

 

Federal Highway Administration Design Manual: Deep Mixing for Embankment and Foundation Support

CHAPTER 12. QC/QA

12.1 Introduction

A well planned and executed QC/QA program is critical to the success of a DMM project. The program's two parts may be generally defined as follows:

The contractor provides all the personnel and equipment necessary to implement the QC program. The owner's representative is mainly responsible for the QA program, although many of the QA activities are frequently assigned to the contractor, such as coring, sampling, and testing. The owner's representative observes construction on a full-time basis and reviews the submittals from the contractor to verify that the QC and QA programs are being properly implemented.

12.2 QC

The QC program is the responsibility of the contractor and generally includes the following components:

12.2.1 Role of QC Personnel

The DMM contractor provides the QC personnel (a QC technician and a project engineer). The QC technician monitors the operational parameters in real time to ensure the DMM operation follows the predetermined procedures. If an installation parameter deviates from the allowable range of the predetermined value, the QC technician informs the rig operator and the batch plant operator to adjust the installation parameters to address the conditions. The project engineer supervises the work of and provides technical support to the QC technician as well as reviews the installation records of the real-time monitoring system.

12.2.2 Bench-Scale and Field Validation (Full-Scale Field Testing) Programs

Preproduction bench-scale testing and full-scale field testing programs should be planned and executed prior to production, as outlined in chapter 10.

12.2.3 Deep Mixing Work Plan

At least 30 days prior to the start of deep mixing work, the DMM contractor submits a deep mixing work plan to the owner for review and approval, as detailed in section 9.3.4.

12.2.4 Materials and Production Monitoring

The QC program includes monitoring and documenting materials handling and construction procedures, including the following:

For a successful DMM operation, real-time monitoring must be used to control and document the current operational data, summarize the preceding operational data at intervals of 3 ft (1 m) or closer, and detect any non-complying operational parameters or records. A real-time monitoring system is a computer-based QC device that indicates the instantaneous installation parameters of a DMM element while the mixing is in progress. Using a real-time monitoring system, deficiencies can often be corrected before the DMM element is completed.

Monitoring additional items is necessary for TRD, including tool rotational speed, horizontal movement speed, and viscosity of soil-binder mixture during production. These procedures and parameters are controlled through the coordination of the contractor's rig operator, batch plant operator, and QC technician/engineer.

Layout

The contractor should accurately stake the locations of the proposed DMM elements shown on the construction drawings prior to installation. GPS technology attached to the mixing equipment has been successfully used to control and document DMM element locations. The engineer is responsible for reviewing the locations of DMM elements. The contractor should provide an adequate method of allowing the engineer to verify the as-built location of the elements during construction. Misplaced elements will be reviewed by the engineer to determine if they will interfere with the proposed construction. The contractor is responsible for correcting the location or alignment of misplaced elements that will adversely affect the project quality. The contractor should correct misaligned elements that interfere with the project in a manner acceptable to the owner.

After production is complete, the DMM contractor should submit as-built drawings indicating the locations of the DMM elements in terms of project coordinates for review and approval by the owner.

Binder Handling and Slurry Preparation

Mix designs verified during the field validation program should be used in production. Revalidation through laboratory or field testing is necessary for changes that exceed 10 percent of previously approved mix designs.

For the dry method of deep mixing, the binder is stored onsite or is delivered just in time to a container that feeds the mixing machine. The binder must be kept dry throughout the storage and delivery operations.

For the wet method of deep mixing, the binder production equipment must be capable of providing slurry with consistent and verifiable quality. Dry binder, mainly cement or slag-cement, is stored in silos and fed to mixers for shearing and agitation. To accurately control the proportions of the slurry components, the amount of water and binder must be determined by weight using automatic batch scales in the mixing plant. Admixtures, if used, can be delivered to the mixing plant by calibrated auger. However, the DMM contractor must prove that the calibrated auger can deliver the quantity of dry admixture with accuracy equivalent to that measured and delivered by weight. Equipment for proportioning used during binder production should be calibrated prior to initial use and repeated every 3 months or every time the batch plant is relocated. Simple checks of material quantities should be made routinely, such as counting the number of bags or truckloads of binder materials that have been used. These quantities should be reported in the daily production report.

Dry binder and water must be mixed in the slurry plant for uniform suspension of binder in the water. The uniformity of binder slurry should be verified by specific gravity tests of the slurry in the agitation tank. A maximum holding time of 4 h in the agitation tank is recommended. Holding time is calculated from the beginning of the initial mixing.

The contractor should measure the specific gravity of the binder slurry at least twice per shift per slurry plant using the methods outlined in ASTM D4380.(129) This simple and quick test provides an indicator that the binder slurry meets the mix design criteria established in the bench-scale testing. Early indication of slurry density allows the contractor the opportunity to adjust mix proportions prior to injection. Other verification methods, such as coring and wet grab sampling, are completed after mixing when changes can no longer be made. The specific gravity of the binder slurry measured during production may not deviate by more than 3 percent from the established specific gravity. If the specific gravity is lower than that required by the mix design, the contractor should add additional cement and remix and retest the slurry. Alternatively, the engineer may request that the DMM contractor recalibrate the batch scales and perform additional testing. The specific gravity measurements should be indicated in the daily production report (see section 9.3.6).

Binder Injection Rate

The binder injection rate per vertical foot (meter) of column is determined in accordance with the design mix, which is based on the bench-scale testing program and the contractor's experience (not applicable for TRD). The mix design and required binder injection rate are verified during the field validation program by assessing the uniformity of the core and the strength of core samples. During production, the binder rate must be monitored constantly and controlled on a real-time basis. The contractor must record in the daily production report the weight of dry binder or the volume of binder slurry injected for each 3 ft (1 m) (measured vertically) during penetration and withdrawal for each element. These records can and should be used to calculate the binder factor as a function of depth much more reliably than chemical testing methods such as performing cement content tests of hardened soil-cement mixtures.

If the weight of dry binder or the volume of binder slurry injected per vertical foot (meter) is less than the amount required to meet the binder factor or volume ratio established during the field validation program, the element must be remixed, and additional binder must be injected at the design binder injection rate to a depth at least 3 ft (1 m) below the deficient zone.

The contractor may request that the established binder factor or water-to-binder ratio of the slurry be modified during the production installation. To verify acceptable results for the proposed modification, the engineer may require additional testing or a new test section at no additional cost to the owner.

Mixing Tool Rotation Speed and Penetration/Withdrawal Rates

Each DMM column must be installed without interruption. If installation is interrupted for more than 1 h, the element must be remixed while injecting binder at the design rate for the entire length of the element. The DMM equipment should be able to remix the element with additional binder within 24 h. Setting of the binder mixture generally will not prevent remixing by properly sized equipment.

The mixing tool rotational speed and penetration/withdrawal rates are adjusted so that resultant mixing of the soil and binder slurry will produce the required uniformity and strength. The required rotational speed and penetration and withdrawal rates for the various soil layers encountered are selected by the contractor and verified during the field validation program.

The rotational speeds and penetration/withdrawal rates must be monitored on a real-time basis during production. If BRN is more than 15 percent below the value determined to be reliably acceptable from the field validation program, the column/element section must be remixed while injecting grout at the design binder injection rate. The rotational speeds and penetration/withdrawal rates must be recorded in the daily production report.

The contractor may request that the established mixing parameters be modified during the production column installation. To verify acceptable results for the modified parameters, the engineer may require additional testing or a new test section at no additional cost to the owner. Alternatively, if sufficient data are available from production columns in support of modified mixing parameters, the owner may accept such data in lieu of an additional test section.

Element Verticality

The equipment operator should monitor and control the vertical alignment of the mixing tool stroke in two directions (longitudinal and transverse to the element alignment). Vertical alignment should be maintained within 1 percent of plumb during the element installation. Manual or automated verticality readings should be recorded in the daily production report at the frequencies outlined in the specification.

Element Top and Bottom Elevations

The termination depth of DMM elements is designed to meet the foundation requirements of the structure, as discussed in chapter 6. For designs that specify the top and bottom elevations of the DMM elements, the constructed elements must extend from the specified bottom elevation or lower to the specified top elevation or higher. The specified top and bottom elevations may vary across the site depending on in situ ground conditions and facility requirements.

For sites that have a well-defined competent bearing stratum, the necessary bottom elevation can be based on refusal criteria determined from penetration speed, vertical load from the mixing tool, mixing energy, and power consumption needed for mixing tool penetration. The refusal criteria can be developed during the field validation program by installing test elements within 5 ft (1.5 m) of an existing boring and recording the operational parameters encountered when the intended competent stratum is reached as indicated in the adjacent boring. If mix designs or operational procedures are modified during production, refusal criteria must be reestablished.

The total depth of penetration can be measured either by observing the length of the mixing shaft inserted below a reference point on the mast or by subtracting the exposed length of shaft above the reference point from the total shaft length. The contractor is responsible for achieving the specified top and bottom elevation requirements and for recording the actual elevations. However, remedial measures for elements of insufficient depth could significantly and adversely impact project costs and schedule, and it is helpful for the engineer to observe and confirm the element termination depth during construction. The mixing equipment must be adequately marked to allow QA personnel to confirm the penetration depth. The depth may also be determined by instruments and displayed in real-time. The contractor should measure and record top and bottom elevations in the daily production report.

If the depth to the competent soil layer at the bottom of the DMM element is found to be different from that indicated on the plans, the engineer may direct the contractor to shorten or deepen the element. The contractor should be compensated based on the decreased or increased amount of deep mixing as the engineer varies the termination depths. However, the contractor should not be compensated for any portions of the elements that are above the top elevations or below the bottom elevations shown on the plans that are not approved by the engineer.

Bottom Mixing

When the mixing tool reaches the design depth, bottom mixing is generally required to provide an adequate level of mixing in the lower portion of the DMM column. Bottom mixing is conducted by lifting the mixing tool approximately 5 to 10 ft (1.5 to 3 m) above the design depth while maintaining mixing action and repenetrating to the design depth. The zone and procedure of bottom mixing should be established during the field validation program. (Note that bottom mixing is not applicable for CSM or TRD.)

Control of Spoils

The contractor is responsible for controlling and disposing all waste materials produced as a result of the mixing operation in accordance with the project requirements. Areas for containing and processing the spoils should be designated on the project plans.

The contractor's selection of means and methods can be heavily influenced by requirements and procedures for handling spoils. Spoils may be handled in several different ways. Often, the spoils are contained at the ground surface until they are sufficiently cured to be stockpiled and used for engineered fill. If unacceptably high pH levels preclude the use of spoils as fill, removal and offsite disposal may be necessary.

12.2.5 QC Documentation

The contractor should report the QC activities and results in the daily production report and submit the report to the engineer by the end of the next business day. The engineer should review daily production reports in a timely manner. The data submitted in these reports are indicators that the contractor is adhering to the procedures established during the field validation program and properly implementing the QC program. Strength and uniformity of the treated soil are used for acceptance, as described in section 12.3.6.

12.3 QA

QA is generally performed by the engineer (or owner's representative) and includes the following tasks:

12.3.1 Role of QA Personnel

QA personnel observe the DMM installation and QC operation performed by the contractor, communicate with the contractor's QC personnel, and review the QC submittals. QA personnel may also perform independent sampling and testing. The QA personnel should inform the QC personnel and the owner/engineer immediately if deficiencies are identified. Early correction of a deficiency often reduces costs and schedule delays.

12.3.2 Engineering Properties to be Verified for QA Purposes

For ground improvement applications, the most commonly used engineering property for QC/QA is unconfined compressive strength. Permeability and strength are used for earth retention or groundwater control applications. Other engineering properties required for the design of most DMM projects (shear strength, tensile strength, and modulus) can be obtained by correlation with unconfined compressive strength.

12.3.3 Coring

Core samples provide the best representation of the hardened in situ DMM column. Assessing full-depth continuous cores of DMM elements is most frequently used as the basis for approval of the uniformity and strength of treated soil. Core testing data reflect the inherent variation of subsurface soil conditions and consequently exhibit greater variation in engineering properties in comparison with data obtained from testing wet samples.

Coring Methods

Full-depth continuous core samples may be obtained using coring methods available in the United States. Triple tube sampling techniques or equivalent provide the highest core recovery and lowest core disturbance. Double tube sampling techniques can also be used to retrieve the core samples.

Core recovery is calculated as the total length of recovered core divided by the total core run length (expressed as a percentage). Total length of recovered core includes the lengths of both treated and untreated soils. Percent treatment is calculated as the total length of recovered core minus the sum of the lengths of unmixed or poorly mixed soil regions or lumps that extend across the entire diameter of the core divided by the total core run length (expressed as a percentage). Unrecovered core is considered untreated soil for the purpose of determining percent treatment unless convincing documentation can be provided by optical logging of the core hole walls that the lack of recovery was due to the coring process.

Cores should be taken continuously from the top to the bottom of the column. Each core run should be approximately 3 to 5 ft (1 to 1.5 m) in length, and core diameter should be at least 2.5 inches (64 mm). To calculate the core recovery for each run, the elevation of the bottom of the core holes should be measured after each core run. Cores should be retrieved at a distance of one-fourth the column diameter from the column center. This location has been shown to yield the most representative samples of the treated material. Material at the center of the column may tend to be higher in binder slurry content, especially if a binder slurry with relatively high water-to-binder ratio is used. Obtaining cores located at the periphery of the column can be difficult because the core barrel tends to exit the column and penetrate into the native soil. Inclined cores are occasionally obtained to locate the interface between adjacent columns. If drilling tends to exit the column at this coring location, the contractor may drill one-fourth of a column diameter along the centerline of an element or shear wall so the core enters the adjacent column in the same element.

The contractor should determine the time interval between column installation and coring that will allow the treated soil to gain adequate strength and avoid low core recovery and sample disturbance. For 28-day strength testing, the core samples can be retrieved at 20 to 26 days after installation. Core samples retrieved at earlier curing ages tend to have lower recovery and higher sample disturbance.

Core samples with diameters smaller than 2.5 inches (64 mm) tend to exhibit increased sample disturbance and reduced core recovery. Core samples with diameters greater than 3 inches (76 mm) have improved recoveries and less disturbance but can be more costly to retrieve and difficult to handle and transport. Reducing the coring rate (core distance drilled per hour) will usually improve core recovery and reduce sample disturbance.

Core operators with experience in coring soft rocks can retrieve core samples from treated soil with satisfactory recovery and quality. Core samples generally reflect some sample disturbance due to the core process even with a good coring tool and a skilled core operator. The presence of gravel in the mixed soil can cause cracks and other damage during coring and thereby reduce core recovery. During the coring process, gravels inside the soil-binder mixture tend to break or grind the core samples. In cases of poor core recovery, an optical televiewer may be used to supplement assessment of uniformity. Worn or inadequately maintained cutting heads, core rods, and other coring devices tend to reduce the recovery and quality.

All core holes must be filled with slurry with 28-day strength equal to or greater than the specified strength of the treated soil.

Coring Locations and Frequency

For each field validation test section, at least one element for each mix design should be cored for the full depth from the top to the bottom of the element.

For production elements on typical DMM projects, one full-depth continuous core should be made for every 3 percent of elements. An element is defined as the treated soil produced by one setup of either a single- or multiple-axis machine. For smaller, more critical, or more complex projects or for projects at more critical locations within otherwise typical projects (i.e., structure foundations), the engineer could specify that more elements must be cored, up to 4 percent of the total production elements. For a larger, less critical, and less complex project (i.e., a large DMM embankment foundation project in similar subsurface soils along the entire alignment), the engineer could specify that 2 percent of the production elements should be cored. At a minimum, five production elements should be cored at full depth so that a reasonable amount of data are collected, even for small projects.

Some deep mixing equipment produces a relatively large treated area in each element, whereas other equipment produces a relatively small treated area in each element. For example, if the same project were done using two mixing machines that both produce 3-ft (1-m)-diameter columns, and the same column overlap is used, but one machine is single-axis and the other is a six-axis machine, then up to six times as many cores would be necessary for the single-axis machine as for the six-axis machine when the number of cored elements is specified on a percentage basis. A justification for requiring a smaller number of cores for equipment that produces larger treatment areas per element is that the same binder factor, mixing parameters, and blending action apply to the entire area treated. Nevertheless, an engineer may want to consider adding a treatment area criterion to the percentage criterion for determining the number of elements to be cored so that a sufficient amount of data can be collected even if the contractor uses equipment that produces a large treated area per element. For example, an engineer may want to specify that full-depth coring be done on 3 percent of elements or for every 860 ft2 (80 m2) of treated ground, whichever produces the greater number of cores. In this example, the 3 percent criterion would control for all types of elements that produce a treated area smaller than 25 ft2 (2 m2) after accounting for overlaps between elements, and the 860 ft2 (80 m2) criterion would control for all types of elements that produce a treated area larger than 25 ft2 (2 m2) after accounting for overlaps between elements.

The coring frequency should be selected by the engineer during the design stage based on consideration of project size, criticality, and complexity. The selected coring frequency can be stated in the specifications either as a percentage of elements or as a combination of the percentage of elements and the treatment area, depending on the project needs.

For TRD or cutoff walls, every 1,000 yd3 (750 m3) of treated soil or every 300 ft (90 m) of wall in the horizontal direction should be cored. For small-sized projects, at least five elements should be cored to provide a reasonable amount of information for evaluation of deep mixing work.

Core Sample Handling and Testing

Upon retrieval, the full-depth samples should be provided to the engineer for logging, selecting test specimens, and assessing whether uniformity and recovery criteria have been satisfied. Following logging, the engineer selects specimens for strength testing. At least five test specimens should be collected from each full-depth continuous core for unconfined compressive strength testing. Test specimens should have a length-to-diameter ratio of 2 or greater.

Engineering judgment must be used to select test specimens to minimize the potential for biasing the data. Samples should be selected carefully to represent the deep mixed element rather than focusing on samples that appear to be unusually weak or that contain inclusions of unmixed soil that are not proportionately representative of the entire column. For example, testing a core sample containing a gravel-sized piece of unmixed soil would simulate testing a column containing a boulder-sized piece of unmixed soil. This situation is unrealistic unless there is highly unlikely evidence that boulder-sized pieces of unmixed soil exist in the column.

Immediately following logging and test specimen selection by the engineer, the entire full-depth core sample, including the designated test specimens, must be sealed in plastic wrap to prevent drying and transported to the laboratory by the contractor. The samples must be stored in a moist room in accordance with ASTM C192 until the test date.(130) Treated soil samples must not be submerged in water during curing unless they are sealed in a water-tight, zip-sealed plastic bag. It is important to remove as much air as possible prior to sealing to avoid sample swelling.

The portions of the samples that are not tested must be retained by the DMM contractor for possible future inspection and confirmation testing by the engineer until completion and acceptance of all DMM work. If a large volume of samples cannot be reasonably stored on the job site, cores from elements deemed satisfactory may be disposed of prior to project completion if approved by the engineer.

The unconfined compressive strength testing should be conducted by an independent laboratory retained by the DMM contractor and approved by the engineer. Testing for 28-day unconfined compressive strength should be conducted in accordance with ASTM D2166, except that loading should continue on all specimens until the cylinders break sufficiently to examine the interior of the specimen.(131) The broken specimen should be photographed so that the engineer may document any apparent segregation, lenses, and pockets in the specimen.

In addition to tests performed by the contractor, additional confirmation tests may be performed by the engineer on samples collected by the DMM contractor. Both the DMM contractor's testing and the engineer's testing, if performed, must demonstrate that the required strength criteria are met prior to acceptance of the work.

Coring Considerations and Potential Drawbacks

Generally, 2 to 3 weeks of in situ curing time must elapse to allow sufficient strength gain before cores can be retrieved with an acceptable level recovery and without excessive disturbance. If early strength is needed for modification of mix designs, wet samples (see section 12.3.4) can be collected and tested in combination with coring. Costs for core sampling are generally higher than for wet sampling.

Obtaining good core recovery in treated ground with gravel or cobbles can be difficult or impossible. When coarse-grained soils prevent core recovery even with high-quality triple-tube coring methods, acceptance should be permitted based on the strengths from wet grab samples combined with optical logging to verify thoroughness of mixing.

Core samples are not generally suitable for permeability testing. Erratic testing results may result because of fissure cracks induced during the coring process. When permeability testing is critical to the design intent of the structures (e.g., a DMM wall for excavation support and seepage control), wet sampling must be used to produce test specimens for permeability testing.

12.3.4 Wet Grab Sampling

Wet grab samples are produced from bulk samples obtained from discrete locations within DMM elements. Test specimens from the bulk samples are cast and cured under consistent conditions. A single bulk sample can produce nearly identical duplicate test specimens for parameter studies on the effects of binder type, quantity, age, and curing conditions.

During a field validation program, wet sampling can provide information on the reaction of the binder slurry with the in situ soil that can be used to modify mix designs. Wet grab samples tested at 3, 7, 28, and 56 days or more can be used to develop the relationship between the strength and curing age to provide the DMM contractor with information on the rate of strength gain for predicting production strengths.

In production, wet sampling and testing can be used to identify potential weak zones in the treated soil, thereby providing an early indicator before coring operations are performed at 28 days. If wet samples produce results that are consistently acceptable, the frequency of wet sampling can be reduced as the project progresses.

Testing data from wet sampling should be used as an indicator rather than as acceptance criteria. The sampling device tends to retrieve a greater proportion of binder slurry volume than mixed soil because clumps of relatively unmixed soil do not tend to flow as easily into the wet sampling device, possibly making the sample less representative of the overall treated material. In addition, the curing conditions (i.e., temperature, drainage, and pressure) differ from in situ conditions. Since wet samples are obtained from discrete locations within elements, the samples may not represent natural variation in subsurface conditions.

For earth retention or groundwater control applications, permeability tests are generally performed using wet samples for more consistent and reliable testing results. Core samples are generally not suitable for providing samples for permeability testing due the potential for side wall leakage due to the roughness of the core surface and the potential for fissure cracks induced during coring.

Wet Grab Sampling Methods

Wet grab samples should be retrieved from DMM columns immediately after installation and before hardening of treated soil. Various bailer-type sampling tools, including tubes or boxes of different configurations, are used to collect samples. Recommended procedures for sample handling and specimen preparation are described in appendix A. It is important to use standard procedures for preparing and testing specimens to allow test results to be used and compared consistently.

Wet Grab Sampling Locations and Frequency

The contractor should perform all wet grab sampling in the presence of the engineer. The contractor should notify the engineer at least 1 business day in advance of beginning sampling operations. For each test section, a minimum of three wet samples should be retrieved for each mix design used.

In production, for embankment and foundation support applications, one wet sample (i.e., one selected depth at one location) should be retrieved every two production days or for every 2,000 yd3 (1,500 m3) of treated soil, whichever produces the higher sampling frequency.

The contractor proposes locations for wet sampling as outlined in the QC program, considering input from the owner/engineer based on subsurface conditions, DMM layout, review of the QC results, and observation of the soil mixing operation. The sample locations should be distributed uniformly both laterally and vertically within the deep mixed zone. Sampling depths should be selected to ensure that wet samples are retrieved from every main soil stratum underlying the site.

The contractor should report all attempts, successful and unsuccessful, to obtain wet samples. Some deep mixed material may not be able to be sampled readily because either the mixture is too stiff or the material may not flow back into the void left after the sampler is extracted, possibly leaving a damaged element.

Wet Grab Sample Handling and Testing

The sampling tool is inserted into the DMM column to a designated depth, filled with treated soil, and lifted to the ground surface. The treated soil material is then poured into a container, screened for oversized lumps (gravel versus unmixed soil), and placed in 3-inch (76-mm)-diameter, 6-inch (152 mm)-long molds for use as test specimens using procedures similar to those described in appendix A. Normally, eight test specimens are prepared from each wet sample. The engineer may request additional test specimens for QA testing. The volume and composition of oversized lumps should be measured and described. Care should be taken to avoid additional mixing or kneading action on the sample during screening so that the sample is as representative as possible of in-place mixing conditions. The wet treated material should be placed into the mold in three to five layers. After the placement of each layer, the specimens must be tapped or vibrated to remove trapped air bubbles. The specimens should be sealed to prevent moisture from entering or leaving the specimens, and the sealed specimens should be stored in a humid environment.

Immediately after test specimens are prepared, they should be stored until testing in an environment with 100 percent relative humidity and temperature between 68 and 77 °F (20 and 25 °C). If approved by the engineer, the specimens may also be cured at a higher temperature to simulate the in situ curing temperature. It has been reported that temperature in treated soil columns with a high binder factor can exceed 100 °F (38 °C) for more than 3 months in the ground.(132) Once prepared, the specimens should not be moved until they have cured sufficiently to prevent disturbance during transportation.

Laboratory unconfined compressive strength tests on cured wet specimens should be conducted in accordance with ASTM D2166, except that loading should continue on all specimens until the cylinders break sufficiently to examine the interior of the specimen.(131) The broken specimen should be photographed, and any apparent segregation, lenses, and pockets in the specimen should be documented. For field validation testing, unconfined compressive strength testing may be performed on specimens at 3, 7, 28, and 56 or more days. For full production work, unconfined compressive strength testing may be performed at 7 and 28 days.

Laboratory permeability testing should be performed on cylinders at 7 and 28 days for the test section and usually only at 28 days for the production elements. Laboratory permeability testing should be conducted in accordance with ASTM D5084.(133)

Wet Grab Sampling Considerations and Potential Drawbacks

Wet sampling has the following drawbacks when it is used as a tool for QC/QA:

12.3.5 Other Verification Methods

Exposure and Inspection

DMM columns can be excavated and exposed for observation, sampling, and testing. For zones of mass stabilization, a large diameter inspection shaft can be constructed within the DMM block. An alternative to personnel entry is a down-the-hole camera, which can be used to inspect the inside surface of the borehole, especially in zones where core samples cannot be retrieved, such as in a gravelly soil stratum.

Penetration and Pull-Out Tests

Common penetration and pull-out tests include the following:(134)

Other In Situ Direct Testing

Numerous in situ direct testing methods have been investigated or applied for the evaluation of the in situ strength of DMM material. However, these methods have not been adopted for routine use like the coring and wet sampling methods. The following are in situ direct test methods:(134)

Geophysical Testing

The use of the following geophysical testing methods has been investigated or applied for the evaluation of the in situ strength of DMM column, but the methods have not been adopted for routine use like coring and wet sampling:(134)

12.3.6 Acceptance Criteria

The engineer should determine the acceptability of the test results. The treated material must meet acceptance criteria relative to geometric layout, strength, and uniformity. The following subsections include examples of generalized acceptance criteria. However, the engineer should develop project-specific acceptance criteria based on the requirements of each project.

Acceptance Criteria for Geometric Layout

The DMM element should be installed within the following general geometric tolerances:

Acceptance Criteria for Treated Soils

The strength acceptance criteria have a major influence on the distribution of strength data obtained during full production. An acceptance criterion requiring that all test data exceed a specified value could require that the contractor produce treated soils with strength significantly higher than the design value, which already incorporates a factor of safety, as discussed in chapter 6. Specifications that allow a certain percentage of test results to be lower than the specified value reduce over-conservatism, but such specifications can still be written to fully satisfy the design intent. The recommended strength criteria include the following:

Uniformity acceptance criteria encourage the contractor to provide a level of soil-binder mixing energy sufficient to reduce the occurrence of untreated lumps and variation of treated soils. The recommended uniformity criteria for transportation projects include the following:

Although the uniformity criteria are recommended for typical transportation projects, other uniformity criteria can be considered. Core recovery, maximum size of untreated soil, sum of the length of unmixed or poorly mixed soils greater than the core diameter, and rock quality designation (RQD) have also been used as indices for uniformity. The ranges frequently used for core recovery are at least 80 to 85 percent for every 3- to 10-ft (1- to 3-m) core run and an average of at least 85 to 90 percent core recovery for the full-depth core from top to bottom of the element. If the percent of core recovery cannot be obtained in gravelly soil, optical televiewer logs can be used to confirm uniformity. In some projects, a minimum RQD varying from 50 to 70 percent has been required in conjunction with core recovery. The higher bounds of these ranges are specified for projects that demand high strength or lower variation of the treated soils. A maximum size of untreated soil varying from 6 to 12 inches (150 to 300 mm) has been specified. The sum of the lengths of unmixed or poorly mixed soil regions or lumps that extend across the entire diameter of the core are required to be less than 10 to 20 percent of the core run length. The lower bounds of these ranges of requirements on the maximum individual lump size and sum of the lengths of untreated soil are specified for projects that demand high strength or lower variation of the treated soils.

In some special cases, such as the use of a deep mixed block for uplift resistance, the unit weight of treated soils may be specified. For wet deep mixing in soft ground, the change in unit weight after soil mixing is negligible for typical binder factors and area replacement ratios used for embankment and structure support applications in transportation infrastructure projects. However, in cases that require uplift resistance and when DMM is used at sites underlain by soils with a unit weight greater than the unit weight of the binder slurry, the unit weight of the treated soils will be lower than the untreated soils. The unit weight of binder slurry generally ranges from 91 to 101 lbf/ft3 (14 to 16 kN/m3) for a water-to-binder ratio of 0.8 to 1.2. Slurries with water-to-binder ratios greater than 2.0 have been used for cutoff wall installation or treatment of clay soils with high plasticity. The unit weight of treated soil can be calculated as outlined in chapter 5. Caution should be given when specifying unit weight because the DMM contractor has limited control over the unit weight of the treated soils and the use of a water-to-binder ratio less than 1.0 might be difficult in stiff soils or sandy soils. If needed, the unit weight criteria could include provisions such as the following:

12.3.7 Remedial Measures for Noncompliance

Geometric Layout

Although the rejection of completed DMM work based on geometry noncompliance is unusual, control and monitoring of alignment, verticality, top elevation, and bottom elevation are very important. The QC and QA personnel are responsible for observing geometric layout of the production work on a daily basis. If the element does not fall within specified tolerances, the contractor must correct the construction procedure before production work is allowed to continue. Minor repairs could be made by redrilling before the hardening of treated soil or by installing additional elements to replace the misaligned elements, as approved by the engineer.

Treated Soil

If DMM elements fail either strength or uniformity criteria, the contractor and the engineer should work together to evaluate the operational data, and the contractor may collect an additional core sample in the same element for the engineer to assess the extent of the deficient zone. If the additional core meets the criteria, then the element should be accepted. Alternatively, the contractor should be allowed to core the elements on both sides of the failed element. If those two cores meet the criteria, then the element should be accepted. If the additional cores fail, then the contractor can propose remedial measures, which the engineer will review and accept or reject, depending on whether the proposed remedial measures meet the design intent. Examples of such remedial measures include the following:

 

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101