How to monitor concrete crack movement using ACI guidance

How does ACI guidance turn crack observations into a repair decision?

Monitor concrete crack movement by mapping the crack, fixing a displacement gauge across sound concrete, and recording changes alongside concrete temperature, moisture and loading. ACI RAP-1, reapproved 2009, calls for the damage cause to be assessed and corrected before epoxy repair and identifies later movement as a reason epoxy may be unsuitable. Use a gauge that resolves the movement relevant to the repair, observe the temperature and load cycles that drive it, and document the release decision before injection.

Start by naming the repair objective: structural rebonding, water exclusion, edge support or appearance. The concrete guide hub connects the investigation to the relevant repair system.

ACI 224.1R-07, published March 2007, provides the evaluation and repair framework. ACI RAP-1, Structural Crack Repair by Epoxy Injection, 2003, reapproved 2009, supplies the specific public guidance used here for epoxy selection. These publications guide the investigation; the project identifies separately its specifications, adopted requirements and structural design acceptance.

ACI 364.9T-03(11), Cracks in a Repair, addresses repair-cracking consequences. The 2021 ACI PRC-364.9-21 publisher description addresses cause evaluation and cracking mitigation in repair design.

Width and photographs establish condition; a time series establishes behavior before repair selection. Use the epoxy and polyurethane injection guide for execution after this investigation establishes the repair basis.

Is the crack active, dormant, cyclic or progressively widening?

An active crack exhibits repeatable opening, closing, shear, out-of-plane displacement or continuing change beyond the installed system's uncertainty. A dormant classification belongs to a stated observation period, resolution and set of service cycles. Record “no detectable movement over the stated envelope” rather than predicting permanent dormancy. Reclamation's Guide to Concrete Repair, second edition, 2015, distinguishes daily or seasonal cycling from progressive widening and ties repair selection to that behavior.

Retain both peak-to-peak excursion and net change. A crack that opens and closes back to its initial reading remains active throughout that excursion. Cyclic movement can be expected at a designed joint while still defeating a proposed rigid repair.

Crack width is the actual surface gap at a marked point; gauge displacement is the change between mounting points. Zeroing a gauge across a 1.20 mm crack does not make its width zero. Exposure-based crack-width guidance cannot be used as an allowable movement threshold, and a small gauge reading does not establish member capacity.

New distortion, offset or load-related change requires prompt structural assessment. ACI 546R-14, section 1.2.1.1, calls for immediate safety issues to be reported to the owner, access restricted around loose overhead or vertical concrete, and compromised members stabilized or affected areas removed from service. Activate the project response without waiting for more routine gauge readings.

Which competing causes should the record distinguish?

Compare timing with physical evidence. Reversible opening linked to concrete temperature suggests a thermal component; rainfall or groundwater changes accompanied by survey movement point toward a different investigation. Traffic-linked displacement, corrosion staining with delamination, and progressive map cracking each require their own corroborating observations. FHWA-HRT-04-150, Chapter 4, July 2006, explains why surface patterns must be interpreted with material and deterioration evidence.

Use the crack types and causes guide for the broader diagnosis, including drying shrinkage. Investigate gradients with the slab curling guide, corrosion with the spall repair guide, and continuing foundation displacement with the underpinning guide. Compare restraint locations with the joint layout. Movement records support those diagnoses without independently identifying internal damage.

Which crack monitor resolves the movement you need to measure?

Select the instrument against the smallest movement that changes the repair decision. Resolution is the readable increment; accuracy describes agreement with the reference; repeatability describes consistency on repeated readings. Range, calibration and mounting stability also affect the result.

FS means full scale. A microscope reads surface width at a mark; a transducer reads displacement between anchors. The anchor span includes concrete strain as well as the crack, so preserve span and orientation. Use separate axes for opening, shear or out-of-plane displacement. A sound-concrete control helps investigate mounting drift or local strain, but it also moves thermally: subtracting it automatically can remove real behavior.

Photographs record location, crack-tip extension and deterioration. Reclamation 2015 describes photogrammetric monitoring; quantitative images need a scale in the crack plane, repeatable viewpoint and lighting, and validated processing resolution.

What do the manufacturer documents disagree about?

GEOKON's current Model 4420 web specifications list ±0.1% full-scale laboratory accuracy. Manual X.6, April 13, 2026, Appendix A, conditions that accuracy on polynomial calculation and gives the different linear accuracy shown in Table A. Use the serial-number calibration and correct calculation mode, then verify field repeatability and change uncertainty.

Where should gauges go on the location plan?

Put stations where they test different movement mechanisms: a representative maximum opening, a branch or tip, a relevant support region, opposite faces and contrasting sun exposure. A single opening gauge cannot resolve rotation, shear and differential vertical movement. Select additional axes or survey datums to match those questions; settlement reference points belong outside the suspected moving zone.

The following Anvilfield station-layout estimate uses one 3 m wall crack C1 and an adjacent control; crack length and effective thermal length are independent inputs. Plan origin is wall grid A/1; chainage begins at the lower visible tip. Record member, grid, face, gauge ID, photo direction, actual anchor coordinates and spacing, and the separate width-measurement mark.

This wall has six observation stations, including a tip record and a control. M1, M3 and M4 screen movement; upgrade any station used to decide a 0.10 mm criterion to adequate measurement capability. M2 alone cannot release the whole crack.

In a garage, place the plan within the larger parking structure condition survey and phased restoration scope. Access restrictions and closures determine whether a proposed station can actually be read at the required times.

How do you install, zero and check the monitor?

Mount to sound concrete, not loose paint, plaster or delaminated cover. Follow the selected manufacturer's fixing and adhesive arrangement. Avongard's current FAQ requires adhesive to cure before preset pegs are removed. Before drilling, locate reinforcement, tendons and services, and have the repair designer approve fixing locations and depths. FHWA-HRT-13-028, October 2013, Chapter 6, documents GPR location of internal bridge tendons and warns against relying on as-built drawings alone. Mark drilling restrictions on the station plan.

Record actual crack width separately from initial gauge coordinates. Avongard's Plus Tell-Tale record sheet provides signed movement and station-reference fields. Keep raw X and Y readings even when they are not zero, and mark which direction means opening. Photograph the complete station and a readable close-up, including the gauge ID and sign convention.

Anvilfield's manual field check uses three remove-and-reseat readings per visit, retaining all three. Compare their spread with the repeatability allowance established during setup. Published 0.1 mm caliper resolution does not support a 0.05 mm repeatability claim or reliably clear a 0.10 mm release decision by itself; select a finer verified system where that decision matters.

Define the uncertainty of a change between visits, including both readings, calibration, seating, thermal correction and mounting stability. Recheck a sound reference and a known length where the equipment permits; investigate a changed reference instead of averaging it away.

A bumped or debonded gauge invalidates its affected readings. Preserve the last valid datum, document the disturbance and establish a dated replacement baseline. Link old and new series only with an independent transfer measurement; otherwise show the break.

How often and how long should crack movement be monitored?

Anvilfield estimates a 5 to 9 week initial program for a stable, accessible, non-emergency investigation: morning and afternoon for 7 days, including the baseline, then weekly for 4 to 8 weeks. This gives 18 to 22 reading rounds.

These are Anvilfield planning intervals, not ACI minimum durations. Reclamation 2015 supports extended observation to distinguish daily and seasonal cycles from progression; Avongard recommends relating movement to external factors. Include occupied and unoccupied loading, relevant wet weather or groundwater change, and the hot and cold conditions that govern an exposed member.

The seasonal allowance in Table C extends until relevant extremes are covered. Earlier intervention requires a documented design decision and monitoring around stabilization or a movement-accommodating repair. For a brief vehicle passage, validate the sensor and acquisition response against the event duration; the thermal logging interval in Table C can miss it entirely.

What temperature, moisture and loading should you log?

Record concrete temperature at the gauge with every displacement reading. Avongard's current FAQ recommends simultaneous construction-material temperature measurements, especially at high resolution. Use the same marked surface and thermometer technique; shaded air temperature cannot represent a sunlit concrete face reliably.

This log supplies Anvilfield estimate inputs for Table E, using fixed UTC-05:00 times, unchanged loading and dry surfaces. Field logs retain raw instrument readings and actual time zones through daylight-saving changes.

Add operator initials or role ID, units, sensor temperature when different from concrete temperature, reference checks and correction version. Retain duplicate readings. Mark missing and invalid values explicitly rather than interpolating through an event that controls the decision.

GEOKON manual X.6, section 6.3, identifies rainfall, traffic, excavation, seasonal changes and nearby construction among relevant influences. For a leaking wall, use the below-grade waterproofing guide to connect groundwater and drainage observations with water-control scope.

How do you distinguish thermal cycling from a continuing trend?

Plot opening and concrete temperature against time on aligned axes, then plot displacement against temperature. Compare readings at similar concrete temperatures and loading; heating and cooling paths can differ because of thermal lag, restraint and through-thickness gradients. The mass concrete thermal-control guide addresses early thermal cracking; the slab curling guide addresses gradient-driven slab behavior.

Using FHWA's approximate concrete thermal-expansion range of 8 to 12 microstrain/°C, Anvilfield estimates 0.48 to 0.72 mm of free movement for a 3 m effective length over a 20°C change. Microstrain here means micrometres of length change per metre per degree. Establish effective length and restraint before assigning that movement to a crack.

Anvilfield estimates Table E using a 3 m effective length normal to the crack, 10 microstrain/°C within FHWA's range, Table D temperatures and an imposed opening trend of 0.02 mm/day. All thermal expansion closes the crack at M2; loading stays constant. The assumed initial width of 1.20 mm keeps the crack open throughout.

Multiply effective length by thermal coefficient and temperature change to obtain the thermal component, then add elapsed days times the opening trend. The calculation uses exact elapsed hours divided by 24: displacement = −0.03 × (T − 10) + 0.02 × t, in mm, with T in °C and t in elapsed days from day 0 at 07:00. Afternoon points occur one-third day later; only displayed results are rounded.

The pure thermal component changes by 0.60 mm; concurrent opening makes the morning-to-afternoon change 0.59 mm. Maximum minus minimum across the sampled series is 0.87 mm. Matched-temperature morning readings reveal +0.28 mm over 14 days, or +0.02 mm/day, while using only the first and last readings yields −0.31 mm and conceals that progression.

Test a remaining trend against independent load or settlement evidence. Time and temperature can rise together and create a convincing scatterplot without separating their effects. Repeated heating and cooling cycles are more useful than an arbitrary correlation cutoff. Define how closely temperatures must match using the observed thermal sensitivity and temperature uncertainty; also match loading and heating or cooling phase. A fitted residual alone does not diagnose settlement.

Separate sensor correction from real concrete movement

Apply the sensor's calibration and any necessary instrument temperature correction before interpreting member behavior. GEOKON manual X.6, section 6.2.1, offers an optional correction for maximum accuracy or temperature changes greater than 10°C. Preserve raw and corrected columns and the formula version. Correcting instrumental thermal response leaves the concrete's physical thermal movement in the record and in the repair design.

Worked calculation: converting GEOKON readings to movement

GEOKON Model 4420 manual X.6, April 13, 2026, sections 6.1 to 6.2.1, publishes the following 50 mm instrument conversion inputs. Subtract the initial raw reading from the current reading and multiply by the calibration factor; calculate the optional sensor temperature correction and add it. The independently calculated linear result is 21.80 mm before correction and 22.19 mm after correction.

The method record below retains calibration precision. Field stations use their own calibration reports; label polynomial and thermal B constants separately.

For the linear calculation, D = G × (R1 − R0). For the polynomial calculation, D = A × R1² + B × R1 + C, with C set so the initial reading produces zero. The sensor correction is K × (T1 − T0), where K = ((R1 × M) + B) × G using the thermal constants. Add the unrounded correction to the chosen displacement result, then round for reporting.

What evidence releases a crack for epoxy injection or a flexible repair?

Release requires a repair objective, cause treatment or a movement-accommodating design, valid measurements through relevant cycles, acceptable product conditions and structural design acceptance where applicable.

Anvilfield uses an assumed 0.10 mm project review trigger and an assumed ±0.05 mm bound on change for the fine-monitoring worked example. Field use must substantiate that bound through calibration and installation checks. It covers the installed system, not just sensor resolution, and expresses an engineering error allowance rather than a statistical confidence interval.

Preserve direction, but compare movement magnitude with the trigger. For magnitude bounds, subtract the change allowance from the absolute reading, stopping at zero, and add it for the upper bound. A range wholly above confirms exceedance; touching or crossing the trigger requires review. A wholly lower range clears only that observation, subject to coverage and design acceptance.

For example, +0.04 mm with a ±0.05 mm change allowance gives a magnitude range of 0 to 0.09 mm. The assumed 0.10 mm trigger is an Anvilfield review input, not an ACI movement limit, epoxy elongation allowance or safe crack width.

At day 7, +0.14 mm at matched temperature spans +0.09 to +0.19 mm and overlaps the trigger. At day 14, +0.28 mm spans +0.23 to +0.33 mm and confirms progression above it. The day-0 cyclic change already exceeds the trigger with the allowance applied, so rigid injection is on hold from that observation.

ACI RAP-1, reapproved 2009, identifies epoxy injection for cracks 0.002 in (0.05 mm) and wider while warning that subsequent movement can make epoxy repair unsuitable. An injectability width establishes neither dormancy nor repair release. Use the movement-joint systems guide where the design accommodates movement and the foundation stabilization guide where continuing displacement requires cause treatment.

The moisture and application limits in Table G come from Sika's Sikadur-35 Hi-Mod LV data sheet, July 2021 version 01.02. Its 6 mm value is the manufacturer's rounded pairing with 1/4 in. Verify the current product instructions at application; failed conditions require an approved revised product or detail.

Reclamation 2015 distinguishes ordinary polyurethane leak-sealing foam from epoxy structural rebonding and identifies specialized two-component polyurethane systems. Specify the selected system's structural function and movement capability. A tensile elongation percentage does not automatically establish a cyclic crack-movement rating. Coordinate water control and structural work where the crack both leaks and requires rebonding.

When does the investigation remain on hold?

The field reader validates and records measurements; the monitoring reviewer interprets movement and exceptions; the structural repair designer releases structural repair. Hold when the cause remains active, relevant extremes are missing, measurement integrity fails or product limits are unmet. Record the failed item and the next observation or design action.

How much does crack monitoring cost, and what is included?

Anvilfield estimates $2,595 to $3,005 for six manual stations and 18 to 22 paid reading rounds, excluding structural design, electronic monitoring, lifts and traffic control. The location plan's electronic M2 and R1 stations, calibration and readout require a separate supplier quote.

Method: Anvilfield estimates this program using its November 5, 2026 labor, material and local-travel allowances, multiplying each listed quantity by its rate and adding reporting. Each round is a separate trip. Trained owner staff collecting routine readings changes paid labor and travel without removing observations. Replace these editable allowances with actual bids. Structural analysis/design, electronic equipment and logging, specialist surveys, access equipment, protection/closures, tax and repair labor are outside the total.

The optional microscope price in Table I was checked on November 5, 2026; GEOKON supplies equipment quotations.

Specify station installation, reading count, data format, interpretation, replacement of damaged gauges and ownership of follow-up readings. Separate access and protection costs, particularly where closures or repeat mobilizations govern the scope. An estimator scoping public-building crack investigations can use PorchRocket Bid Control to find city, county and school-district repair bids and early signals from capital plans.

What should be monitored after repair?

Retain mapped stations where practical and begin a separately dated post-repair baseline. Inspect the repaired line and adjacent concrete for new cracks, debonding, leakage and repeated displacement through the driver that governed the repair.

Use the Anvilfield follow-up schedule in Table C after the specified cure and service release. Verify the repair through the load, water or temperature condition that originally drove the problem.

For a water-control repair, record the water head or service condition during verification. A dry wall during dry weather supplies a different test from the same wall under the previously leaking condition.

For structural repair, retain the execution and penetration checks in the injection guide alongside monitoring and design acceptance. Continue any spall and corrosion investigation; renewed movement or leakage reopens diagnosis with the original baseline intact.

Field checklist and handover record

Handover includes the location plan, baseline photos, installation and calibration records, raw and corrected readings, environmental log, plots, invalid-data notes, diagnosis, approved repair detail and signed disposition.

• Define the repair objective and identify the member, exposure and loading.
• Map each crack and station with a unique ID, face, chainage and photographs.
• Record actual width at a repeatable mark separately from displacement-gauge zero.
• Select opening, shear and out-of-plane measurements to match the mechanism.
• Select range, calibration mode and installed resolution; establish the uncertainty of a change against the project's decision criterion.
• Verify sound mounting concrete and approve any drilling near embedded reinforcement or tendons.
• Record installation, adhesive cure and removal of preset pegs; retain the initial reading.
• Verify repeated readings and document uncertainty or invalid stations before interpreting change.
• Log concrete temperature, air temperature, moisture, rainfall/service water and loading at each reading.
• Assign the reader, monitoring reviewer and repair decision owner; fund the full visit schedule.
• Cover the suspected daily, event and seasonal drivers; identify any missing extremes.
• Plot net change, full excursion and comparable-temperature readings separately.
• Retain raw readings and show sensor correction, calibration and units in the calculation record.
• Record the diagnosed cause, its treatment and the selected product's current application limits.
• Document the repair-release or hold decision, with structural design acceptance where applicable.
• Establish the post-repair baseline, next readings and event-based verification.

Carry the filled hold record into the next review; when release becomes appropriate, record every satisfied criterion, remaining limitation and the approved repair detail.

Reviewed 2026-11-05. Sources checked are listed below with their document versions. Update log: 2026-11-05, reviewed primary sources and calculations; clarified station capability, safety response, corrosion exclusion, temperature matching and uncertainty decisions; corrected the commercial link.

How to cite this page: Anvilfield, How to monitor concrete crack movement using ACI guidance, 2026, https://anvilfield.com/field-guides/concrete/concrete-crack-monitoring-movement/. Citable tables: gauge resolution, location plan, environmental log, movement series and repair-release criteria. Add “reviewed 2026-11-05” and the table's named authorities and editions.

Related tools

FAQ

People also ask

• Why does concrete crack?
• Epoxy or polyurethane crack injection: which do I use?
• Why does parking garage concrete spall?
• What is concrete slab curling?
• What is a building expansion joint?

Sources checked

• P1, ACI Committee 224, 224.1R-07 publisher abstract, March 1, 2007 Publisher abstract, March 2007: evaluation and repair scope. Opened with curl 2026-11-05.
• P2, ACI RAP Bulletin 1, Structural Crack Repair by Epoxy Injection, 2003, reapproved 2009, public PDF 2003, reapproved 2009, p. 4: cause treatment, movement and epoxy selection. Opened with curl 2026-11-05.
• P3, ACI 364.9T-03(11), Cracks in a Repair, public benchmark PDF 2003, reapproved 2011: historical repair-cracking benchmark. Opened with curl 2026-11-05.
• P4, ACI PRC-364.9-21, Cracks in a Concrete Repair, 2021 publisher page 2021 publisher description: cause, consequences and cracking mitigation. Opened with curl 2026-11-05.
• P5, Bureau of Reclamation, Guide to Concrete Repair, second edition, 2015 Second edition, 2015, printed I-75 to I-79 and I-144 onward: monitoring and repair objectives. Opened with curl 2026-11-05.
• P6, Avongard, FAQs, current undated primary web page Undated current FAQ: readable increments, installation and material temperature. Opened with curl 2026-11-05.
• P7, GEOKON, Model 4420 product specifications, current undated web page Undated current specifications: ranges, resolution and laboratory accuracy. Opened with curl 2026-11-05.
• P8, GEOKON 4420 instruction manual, revision X.6, release April 13, 2026 Revision X.6, April 13, 2026, sections 6.1 to 6.3 and Appendix A: conversion, correction and accuracy. Opened with curl 2026-11-05.
• P9, FHWA, Thermal Coefficient of Portland Cement Concrete, undated research page Undated research page: approximate typical concrete thermal-expansion range. Opened with curl 2026-11-05.
• P10, FHWA-HRT-04-150, Petrographic Methods of Examining Hardened Concrete, Chapter 4, July 2006 FHWA-HRT-04-150, Chapter 4, July 2006: crack interpretation and material evidence. Opened with curl 2026-11-05.
• P11, Sika, Sikadur-35 Hi-Mod LV Product Data Sheet, July 2021, version 01.02 July 2021, version 01.02: substrate, temperature, pressure and crack-width limitations. Opened with curl 2026-11-05.
• P12, Elcometer USA, Elcometer 900 illuminated ×50 microscope, current sales page Current US sales page: magnification and $399 listed purchase price. Opened with curl 2026-11-05.
• P13, Concrete Masonry & Hardscapes Association, Maintenance of Concrete Masonry Walls, current primary resource at legacy TEK 08-01A URL Current masonry maintenance guidance: supplemental brittle-patch indication. Opened with curl 2026-11-05.
• P14, Avongard, Plus Tell-Tale Crack Monitoring Record, current PDF hosted under June 2023 directory Current PDF: signed movement, date, location and gauge-reference fields. Opened with curl 2026-11-05.
• P15, ACI 546R-14, Guide to Concrete Repair, September 2014, public preview Public preview, section 1.2.1.1: unsafe conditions, owner notification, access restrictions and stabilization. Opened with curl 2026-11-05.
• P16, FHWA-HRT-13-028, Guidelines for Sampling, Assessing, and Restoring Defective Grout in Prestressed Concrete Bridge Post-Tensioning Ducts, October 2013 Chapter 6: locate internal bridge tendons with GPR, avoid reinforcement damage and do not rely on as-built plans alone for tendon location. Opened with curl 2026-11-05.