High-Energy Impact Rollers

References: Kelly (2000); Landpac (2008a); Pinard (1999, 2001).

The procedure involves selecting a representative test section of the project area and conducting a compaction field trial to obtain critical soil engineering parameters (density, modulus, strength, permeability). The test section is compacted with several roller passes, with field testing performed to obtain soil properties at various intermediate roller passes and compaction improvement depths. The number of IR passes, an optimal rolling pattern, roller travel speed (which relates to number of impacts per minute), and other site-specific construction details (desired moisture content, grading needed for a smooth testing surface, etc.) are determined from the test section and implemented for production operations.

There are no documented case histories where IR technology was used specifically for liquefaction mitigation, but general guidance applies. IR technology may be used at sites with soils susceptible to liquefaction during earthquakes — saturated sands, silty sands, sandy silts, and silts are the likely candidates. Where IR technology supports embankments or structures and reduces settlement, it's also necessary to confirm there's no risk of liquefaction or other ground disturbance that could cause loss of support or lateral spreading.

The initial assessment of whether soil will liquefy in an earthquake compares the in-situ cyclic shear strength (Cyclic Resistance Ratio, CRR) against the cyclic shear stress that would cause liquefaction (Cyclic Stress Ratio, CSR). SPT, CPT, and Becker Penetration Tests (for gravelly/cobbly soils) are used to determine CRR from boundary curves separating liquefaction and no-liquefaction zones on a plot of CSR vs. penetration resistance — the widely used SPT version of this chart is Youd et al.'s (2001) "Liquefaction Resistance of Soils" boundary curve for magnitude 7.5 earthquakes (with permission from ASCE); a similar plot exists for normalized CPT tip resistance. As an example, a saturated clean sand with a corrected blow count (N₁)₆₀ of 10 blows/ft and an anticipated cyclic stress ratio of 0.25 will liquefy unless (N₁)₆₀ is raised above 22 blows/ft by densification, or the cyclic stress ratio is otherwise reduced.

The analysis is complex in application: CSR depends on input motions within the soil layer (earthquake magnitude/intensity, distance from epicenter, geologic setting, rock conditions, soil profile), CRR depends on overburden stress, fines content, and static shear stress, and normalizing the penetration resistance involves several corrections — especially for SPT. Input ground-motion information can come from local experience, recorded ground motions, or the USGS Ground Motion Calculator and Interactive Deaggregation tool, which help identify the magnitude/distance scenario contributing most to the site's seismic hazard so it can be scaled to the magnitude-7.5 chart correlations. Widely used liquefaction correlation diagrams and computation guidance for SPT and CPT appear in Youd et al. (2001) and Idriss and Boulanger (2008).

The usual design goal is to densify soil sufficiently to attain a factor of safety against liquefaction triggering (CRR/CSR) greater than 1.5, with a minimum of 1.3 — though no single value suits every condition. Each case should be judged on its own merit where a few scattered points fall short of the criteria: a few isolated low values are quite different from several low values grouped closely together, and a large shortfall is more significant than a narrow miss.

Method summary: SPT (ASTM D1586/AASHTO T-206), CPT (ASTM D3441), and Dynamic Cone Penetration (DCP, ASTM D6951) tests assess compaction improvement depth by comparing pre- and post-compaction results. SPT and CPT are widely used and can measure to depths greater than 30 m; DCP is generally used for pavement foundation layers as a QA process and measures to about 2 m. Difficulties arise where large particles or boulders are encountered, which may require drilling to continue testing.

Accuracy/precision: Generally repeatable for comparing before/after values at specified depths, though difficult with large boulders or rocks; DCP repeatability standard deviation is under 2 mm/blow (ASTM D6951); CPT shows about ±10% variation between labs; SPT N-value coefficient of variation is about 100% or more across different equipment/drillers, reducible to about 10% with the same equipment, driller, and energy calibration (Duncan 2000 reports 15–45%). Coverage: Sufficient evaluation requires many tests; penetration tests reach greater depths than surface tests. Landpac (2008b) recommends three DCP tests every 10 passes over a 40×4 m trial section. Implementation: DCP needs one or two people; SPT/CPT require specialized equipment and trained personnel. General comments: Empirical correlations relate results to strength, bearing capacity, stiffness/modulus, and liquefaction susceptibility; useful for both QA and QC (tracking improvement with increasing passes).

Method summary: Load vs. deflection curves determine modulus of subgrade reaction, soil elastic modulus, or bearing capacity, by increasing a static load and recording plate deflections. Avalle and Grounds (2004) used a 320-mm diameter plate; loose surface material needs excavating before testing. ASTM D1195-93 covers repetitive tests (airport/highway pavement design); AASHTO T-222 covers non-repetitive tests.

Accuracy/precision: Widely accepted; ASTM D1195-93 notes precision and accuracy couldn't be determined due to soil variability and lack of a reference test. Coverage: Sufficient evaluation requires many tests; Landpac (2008b) recommends one static PLT over a 40×4 m trial section. Implementation: Needs experience and special equipment (a heavy truck, dozer, or other equipment as reaction force), plus a trained field engineer to interpret load-deflection curves. General comments: Direct measure of modulus/bearing capacity; useful as a QA tool for verifying design values.

Method summary: Light Weight Deflectometer (LWD), Falling Weight Deflectometer (FWD), and Clegg hammer tests measure plate deflections under dynamic impulse loading. LWD applies lower contact stresses (≤ 0.2 MPa) than FWD (up to 1.0 MPa); FWD uses an array of deflection sensors to assess subsurface stiffness/modulus to about 2 m depth. Clegg hammer tests measure hammer decelerations under impulse loading (ASTM D5874). FWD equipment is trailer-mounted; LWD and Clegg devices fit in an enclosed box. Different manufacturers use different sensor types (geophones, accelerometers, sensiometers), which affects modulus values.

Accuracy/precision: LWD values vary by device/sensor type; Clegg hammer has a coefficient of variation around 2% or better in good lab conditions (ASTM D587); no standard reference exists to establish accuracy for any of these methods. Coverage: Fast tests (under 5 minutes each); LWD/Clegg reach shallow depths (≤ 0.5 m) vs. FWD (up to about 2 m). Implementation: Needs experience and special equipment; LWD/Clegg require less training and cost less than FWD. General comments: Direct measure of dynamic elastic modulus; usable as a QC/QA tool.

Method summary: Sand cone (ASTM D1556/AASHTO T191), rubber balloon (ASTM D2167/AASHTO T205), test pit sand/water replacement (ASTM D4914/D5030), nuclear gauges (ASTM D6938/AASHTO T310), and drive cores (ASTM D2937) measure moisture and dry density; results require a reference lab test (Proctor or relative density) to determine in-situ relative compaction. Many of these methods don't work well where large rocks are present.

Accuracy/precision: Precision generally can't be determined for sand cone, rubber balloon, and test-pit methods due to soil variability; ASTM D6938 reports a standard deviation of 0.3–1.2 lb/ft³ for wet density and 0.3–0.5% for moisture content. Coverage: Represents material within about 0.3 m of the test surface; typical earthwork specs require one test per 1,000–10,000 yd³. Implementation: Requires some training; nuclear gauges carry regulatory constraints and require specialized training. General comments: Widely used in earthwork QC/QA, but limited on oversized-particle or variable-fill projects, and doesn't relate easily to strength/stiffness properties used in design.

Method summary: Surface wave testing, cross-hole seismic cones, Ground Penetrating Radar (GPR), and resistivity provide quantitative strength/stiffness information via shear wave velocity data with depth. Use of GPR or resistivity on IR projects specifically is not documented in the literature.

Accuracy/precision: Not documented in the literature. Coverage: GPR can offer better spatial and vertical coverage than conventional seismic methods (cross-hole logging, seismic cones, MASW). Implementation: Requires special equipment, setup, and trained personnel. General comments: Usable for both QC and QA.

Method summary: Compacted soil sampling from borings ("undisturbed" samples) supports laboratory strength and modulus testing; samples from on-site stockpiles support classification, compaction, strength, and modulus testing. Lab-compacted samples can behave differently than field samples due to stress and structure differences, and different lab compaction methods (impact, static, vibratory, gyratory, kneading) yield different properties.

"Undisturbed" samples (Shelby tube) cannot be obtained for granular materials; disturbed samples from stockpiles must be reconstituted in the lab for testing.

Accuracy/precision: Depends on the specific test method; most are widely used and standardized by ASTM/AASHTO. Coverage: Represents only the sample obtained; experience is needed to interpret layering from multiple borings. Implementation: A drill rig is needed for "undisturbed" samples at depth; shallow disturbed samples can use hammer-driven Shelby tubes or test pits. General comments: Enables visual inspection and material for QA testing; helpful when performed before and after compaction.

Method summary: CIR mapping continuously records the deceleration (g's) of the IR drum using accelerometer(s) instrumented in the drum, mapping the results on an on-board computer via GPS. This technology is relatively new and proprietary; limited literature is available (Jumo and Geldenhuys 2004; Landpac 2008a).

Initial passes

After 40 IR passes

Accuracy/precision: Not yet quantified in the literature; detailed field testing is needed for that assessment. Coverage: Provides nearly 100% spatial coverage. Implementation: Requires specialized equipment and proprietary software. General comments: Effective for QC process control; has potential for QA if site-specific correlations are developed through field trial test sections.

Method summary: Determine coefficient of permeability and infiltration rate, respectively. An in-situ Guelph permeameter (Auzins and Southcott 1999) has monitored permeability change with increasing IR passes; an in-situ double ring infiltrometer (ASTM D5126, Avalle 2004a,b) has monitored infiltration change after IR compaction.

Accuracy/precision: Difficult to quantify, but can provide a good relative measure of compaction. Coverage: Most tests represent only shallow depths (< 0.5 m). Implementation: Water-based tests are time consuming (hours to days to reach steady state); air/gas-based tests are rapid (e.g., about 30 seconds per test for the White et al. 2007b air permeameter, on granular materials only). General comments: Faster tests suit both QC/QA; slower tests suit QA only.

Method summary: IR can be used as a "proof compactor" or "proof roller" to visually identify soft spots needing additional compaction or rework, typically by monitoring surface depressions or ruts under the roller.

Accuracy/precision: Cannot be quantified. Coverage: Depends on the number of inspectors on site. Implementation: Easy to implement, but needs continuous on-site monitoring. General comments: Widely used for QA on earthwork projects; provides only a qualitative measure — quantitative surface-depression measurement is covered under Elevation Monitoring.

Method summary: Ground elevation change provides a simple and effective means to assess improvement under IR compaction. Selecting test locations is complicated by the rough surface left behind the roller — addressed by lightly trimming the surface with a grader. Elevations can be tracked with string lines and tape, or more advanced options: GPS, robotic total station, laser levels, and LIDAR.

Accuracy/precision: Depends on the method used; laser levels and robotic total stations are accurate and precise, string lines/tape less so. Coverage: Depends on the number of measurement points; roller-mounted GPS gives near-continuous data. Implementation: GPS, robotic total station, and laser levels may require specialized equipment and training. General comments: Can be an effective QC or QA tool.

Method summary: Ground vibrations from IR passes are monitored to assess their influence on nearby building structures, by measuring surface wave velocities at various distances from the impact source and comparing them against local specification limits.

Accuracy/precision: Not discussed in the references reviewed. Coverage: Depends on the number of sensors used. Implementation: Requires specialized equipment and trained personnel. General comments: Used only as a QC measure to mitigate disturbance to nearby structures — it does not measure the compacted soil itself.

1. General — Section Included: requirements for improving soils with low density and collapse potential, as an alternative to (or combined with) over-excavation and replacement. Since the maximum effective improvement depth under IR is about 3 ft, for a 5-ft treatment zone the upper 2 ft is undercut, the remaining 3 ft is improved by impact roller dynamic compaction (IRDC), and the upper 2 ft is replaced with on-site or imported soils meeting the embankment criteria (Section 02056M).

1.2 Related sections: Section 02056M — Embankment, Borrow, and Backfill; Section 02316M — Roadway Excavation. 1.3 Definitions: IRDC — a process where a large square weight with rounded edges is rolled along the ground surface, imparting compactive energy each time an edge strikes the ground; improvement depth depends on tonnage, number of passes, prewetting, and material characteristics.

1.4 Submittals: a detailed safety program submitted before work starts, covering protection of personnel, adjacent structures, and equipment maintenance; an IRDC construction schedule plan 30 days before mobilization; surveyed Record Drawings (stamped by a licensed Utah surveyor) with each payment request; and, if the Contractor elects to use calibration test pads, a test pad plan submitted 15 days before mobilization (UDOT approves or denies the calibrated pass count within three business days).

1.5 Measurement & payment: by the square yard of IRDC coverage, regardless of the number of passes; payment includes tools, equipment, labor, and materials to complete IRDC within the lateral limits shown on the drawings. 1.6 Qualifications: the IRDC Contractor must demonstrate prior experience with an impact roller for dynamic compaction and be pre-qualified, documenting at least one prior project over 50,000 SY successfully densified by dynamic compaction using impact rollers.

1.7 Quality: the Contractor performs quality control during IRDC, including density measurements between 1 and 3 ft using nuclear density test methods — one test per depth (1, 2, and 3 ft) every 5,000 SF, from dug test pits or trenches. Quality assurance: the Geotechnical Engineer documents subgrade wetting (if required), number of passes per area, and speed during dynamic compaction. A Contractor-elected calibration test pad option allows no fewer than three test pads (≥ 1,000 SF each) in different treatment-area locations, with nuclear density testing before and after compaction between 1 and 3 ft.

1.8 Protection of persons & adjacent properties: a detailed safety program verifies protection of on- and off-site personnel and structures, including a Structure Monitoring Program; a maintenance program with daily/weekly/monthly tasks. A Utilities Monitoring Program requires a qualified specialist to detail vibration monitoring before, during, and after compaction for all structures within 100 ft, coordinate with utility owners on allowable vibration limits, and ensure IRDC vibrations don't exceed those limits.

Part 2 — Products: the equipment must provide at least 33,000 pound-feet of energy. Import soil, if needed, follows Sections 02056M and 02316M. Part 3 — Execution: coordinate IRDC scheduling with other contract work; protect survey control points, benchmarks, monitoring wells, and existing structures; identify and protect utilities; finish clearing/grubbing before starting; review drawings for IRDC limits; determine existing grade elevation with the Project Surveyor and perform a post-excavation survey. IRDC provides effective compaction to depths up to 3 ft; over-excavation/re-compaction can supplement it for deeper treatment. A schedule table specifies station ranges, minimum treatment depth (5 ft in all three ranges on this project), and minimum treatment elevations. Treatment extends from the outside edge of existing pavement to 5 ft beyond the toe of the proposed slope (not exceeding 1 ft within the right-of-way). A minimum of 22 passes at the manufacturer's recommended speed densifies the underlying soils; the Contractor keeps daily records (location, pass count, operation rate) and provides dust control.

3.3 Test pad: the Contractor may construct test pads to calibrate the number of passes needed to improve in-situ soils to a dry unit weight of 100 pcf; the calibrated pass count may replace the 30-pass default (Article 3.2) with UDOT and the geotechnical engineer's acceptance. 3.4 Water handling: means must be provided to remove surface water accumulating in excavations. 3.5 Tolerance & survey control: the Contractor provides survey services to maintain both the lateral limits for IRDC and the roadway alignment, recording IRDC lateral limits as Record Drawings updated with each payment request.

1.0 Scope: a Particular Specification covering in-situ treatment of soil by impact compaction. 2.0 Impact compaction plant: a 25 kJ, dual-drum, three-sided impact compactor is used, with compactor energy rating calculated as mass of the rotating drums × lift × the gravitational constant; the compactor is towed by a 4-wheel-drive tractor of at least 200 kW, capable of maintaining 10–12 km/hr towing speed on up to a 5% uphill grade. 3.0 Definition of an impact compactor pass: for the dual-drum 25kJ, 3-sided compactor, a pass is two runs over a 4 m wide lane, where one drum's second run travels the middle of the inter-drum space created by the first run; overlapping drum paths within one pass is not permitted. 4.0 Setting out: the treatment area is defined on the drawings, with setting-out markers controlling operations to ensure every area receives the specified number of passes.

5.0 Preparation & maintenance of compacting surface: compaction proceeds as soon as possible after clearing/grubbing to exploit available in-situ moisture; water is applied as needed to prevent surface crumbling, aid compaction, and control dust; the surface is lightly graded (a grader unit fitted to the roller, or a separate grader pass) to maintain an even operating surface and prevent deep footprints.

6.0 Impact compaction trials: before full-scale work, trials are run on a representative ≥ 40 m long, 4 m wide section (with turning circles outside the trial area) to determine the number of passes needed and confirm the Contractor's plant/methodology. A trial testing table specifies test type, count, and which pass number(s) each test runs at: settlement (20 m length at 1 m intervals, every 5 passes); DCP at three chainages (2 m DCP length, every 10 passes); density testing (0–900 mm at three chainages, at passes 0, 20, and 60); one plate load test (at pass 40); Continuous Impact Response (every 10 passes); maximum dry density/optimum moisture (4 tests, at pass 60); one foundation indicator and one soaked CBR (at pass 60). The target compaction state is based on the number of passes at which soil substantially attains a near-elastic condition, with target coverage determined from settlement results and submitted to the Engineer for approval within one working day; a trial report is compiled after all lab results are received.

7.0 Compaction: impact compaction must be completed before services, drainage pipes, or culverts are installed, under adequate Contractor supervision. Thirty passes are used for tender purposes; if more or fewer passes are ordered, the Contractor's compensation is adjusted accordingly. Vibration monitoring is required where structures lie within a 20 m radius, keeping Peak Particle Velocity below 25 mm/s (a cut-off trench, if needed to mitigate vibration, is paid separately). A daily Compaction Record Book documents the date, area/section, lane, pass-count range for that day, area dimensions, and area×passes compacted, submitted to the Engineer daily for verification.

8.0 Quality control procedure: (1) CIR measurements monitor and record rotating-drum deceleration correlated to GPS position, recorded during the first three passes and the last three passes, presented to the Engineer as raw data (date, time, Y, X, and G/deceleration) and as a triangulated AutoCAD/Modelmaker model with calculation/interpretation of relatively weak areas. (2) DCP tests (2 m depth, 25 m grid, method TMH 6:1984 ST6) run before and after compaction, with additional testing in relatively weak areas flagged by CIR. (3) Relative density testing (TMH1:A10, moisture correction TMH1:A17) on a 50 m grid at 0–300 mm and 300–600 mm depths, with sufficient modified AASHTO dry density samples to determine relative density. (4) Soaked CBR samples (TMH 1) taken in relatively weak CIR-flagged areas. (5) Plate load tests: a 450 mm diameter rigid circular plate loaded in stress increments (0, 45.4, 0, 45.4, 90.9, 136.3, 181.8, 227.2, 272.7, 318.1, 363.5, 409.0, 454.4, 0 kPa), with settlement recorded at 30 s, 1, 2, 4, 8, and 16 min; done 400 mm below natural ground under soaked conditions (12-hour soak minimum), one per 750 m² compacted. (6) A project report details all test results from compaction operations.

9.0 Measurement & payment: Item 9.1, establishment and de-establishment of impact compaction equipment, personnel, and ancillary equipment — measured as a lump sum, payable on completion of impact compaction operations.