Concrete
ASTM C1583 pull-off tests: what failed, and what does it mean?
Read the fracture plane, calculate measured-area stress, and connect each pull-off result to the repair sampling and acceptance plan.
Direct answer
ASTM C1583/C1583M-25 identifies the weakest tensile plane in the tested concrete repair assembly: substrate and repair fractures measure that material's resistance, while an interface fracture measures local bond strength. Read the stress and fracture plane together, then apply the specified strength, age and acceptance rules. ASTM requires a failure mode for every result and permits averaging only results with the same mode.
On this page
What does an ASTM C1583 pull-off test measure?Is C1583 the right test for this repair or coating?What must the acceptance specification say?How many tests do you need, and where should they go?How do core depth, adhesive, and loading affect the result?How do you calculate pull-off stress?Failure-plane atlas: what broke?Which tests are invalid, and which are valid failures?Is 250 psi a passing result?Worked example: read the fracture before averagingWhen do you retest, reprepare, or investigate?What should testing cost, and how should it be scheduled?Field checklist and acceptance recordCitation and review recordOpen the related tool for the calculation, checklist, or packet.
What does an ASTM C1583 pull-off test measure?
ASTM C1583/C1583M-25 identifies the weakest tensile plane in the tested concrete repair assembly: substrate and repair fractures measure that material's resistance, while an interface fracture measures local bond strength. Read the stress and fracture plane together, then apply the specified strength, age and acceptance rules. ASTM requires a failure mode for every result and permits averaging only results with the same mode.
The same equipment answers two different construction questions. On bare prepared concrete, it tests the near-surface material before repair placement. Through a completed repair, it loads the repair, interface and substrate together, so the fracture determines which strength was measured. ASTM C1583/C1583M-25
A fracture identifies the failed layer, not its unique cause. Use the wider survey to diagnose and scope the concrete repair, and retain corrosion, drainage and structural findings in the parking structure restoration program. Pull-off strength does not establish chloride extent, structural capacity or remaining service life.
Is C1583 the right test for this repair or coating?
Choose the property before choosing the tester. ASTM C1583/C1583M-25 addresses prepared concrete surfaces and the tensile or bond strength of repairs and overlays. ASTM D7234-22 addresses coating pull-off strength on concrete; its §5.5 identifies C1583 as potentially more suitable for cohesive evaluation of cementitious surfacers or weak substrates. ASTM C1583/C1583M-25 ASTM D7234-22
An instrument that accepts both sets of accessories does not make the methods interchangeable. A coating contractor should follow the specified coating procedure and the resinous floor coating installation requirements. A repair contractor needs a cut and load path that actually reach the repair interface under examination.
Keep the meaning of bond strength attached to its test. Compression measures crushing resistance, while direct tension pulls perpendicular to the interface. Louisiana DOTD's 2020 patching qualification procedure lists C1583 direct tension and C882 slant shear separately; a high compressive or slant-shear value cannot be substituted for an installed repair's pull-off result. Louisiana DOTD patching qualification (July 1, 2020)
| Assembly/question | Appropriate evidence | Result interpretation | Source |
|---|---|---|---|
| Prepared concrete before repair | C1583 surface tensile test | Strength of the prepared near-surface zone | ASTM C1583/C1583M-25, §5.1 |
| Cementitious patch or overlay | C1583 through repair into substrate | Identify interface, substrate, or repair failure | ASTM C1583/C1583M-25, §5.2 |
| Protective coating on concrete | Specified D7234 test | Coating system pull-off strength and failure mode | ASTM D7234-22, §§1, 5 |
| Repair mortar's compressive capacity | Specified compressive test | Crushing resistance, not installed interface adhesion | Louisiana DOTD (2020), qualification tables |
| Epoxy bond in slant shear | Specified C882 test | Shear/compression test result, not C1583 tensile strength | Louisiana DOTD (2020), qualification tables |
What must the acceptance specification say?
Write the acceptance rule before the first production pull. Anvilfield's test-plan checklist includes the method and edition, purpose, zones, acceptance age, valid count, stress criterion, treatment of failure modes, individual or average requirement, and authority for additional testing. FHWA tensile-bond guidance
Separate four stages: material qualification, preparation approval, mockup approval and installed-work acceptance. A laboratory product value qualifies performance under its stated conditions. A prepared-substrate trial answers whether the production preparation leaves sound concrete; a mockup tests the proposed assembly; production sampling checks the installed work represented by the agreed lot.
The contract's adopted edition governs the test. Use the instrument manual within that method and the repair manufacturer's curing requirements for its material. Resolve incompatible depth, adhesive, sequence or age instructions in writing before collecting acceptance data. Record the tester's force-verification status, chosen accessories and loading-rate setting with that plan.
The owner or design professional sets project acceptance and release of held structural repairs; the technician records measurements and exceptions. Specify whether low individual values hold the lot and when reserve tests are authorized.
How many tests do you need, and where should they go?
Anvilfield's planning allowance starts with three intended valid pulls and two reserve positions per zone. The adopted method and contract determine the required valid count, compatible failure modes and represented area. One pull is one specimen; the production lot is the work assigned to the resulting evidence.
New York OGS Project 45604-C, October 17, 2024 Addendum 1, §030131 1.05.E.5 requires at least three pull-off tests for each patching material. FHWA-HRT-22-087 used three pulls per research region. Neither establishes a universal tests-per-square-foot rule. New York OGS project specification FHWA-HRT-22-087 (2022)
Separate zones when substrate condition, preparation, repair product, placement date, crew, orientation or exposure changes materially. Distribute representative tests within each zone; identify tests targeting hollow or suspect areas as diagnostics. Use sounding and delamination mapping alongside overlay preparation and installation records.
The layout below divides a 30 m by 20 m deck into two 300 m² placements, R1 and R2, with three initial locations and two reserves each. Coordinates in metres start at the lower-left corner. R1 initial points are (3,4), (11,8), (7,16), with reserves (3,16), (12,16); R2 initial points are (18,4), (26,8), (22,16), with reserves (18,16), (27,16).
Record actual coordinates after relocating a point to avoid reinforcement, cracks or free edges. Three attempts do not guarantee three valid results. Use reserves under the approved replacement or investigation plan, retaining every original attempt.
| Zone | Condition kept together | Initial specimen IDs | Reserve specimen IDs | Report separately | Trigger for more investigation |
|---|---|---|---|---|---|
| P | Prepared substrate, one preparation method | P1-P3 | P4-P5 | Substrate tensile results | Weak near-surface breaks or preparation change |
| R1 | Repair placement 1, same product and cure history | R1-1 to R1-3 | R1-4 to R1-5 | Each strength plus fracture plane | Low valid result, mixed mechanisms, invalid attempt |
| R2 | Different placement date or substrate zone | R2-1 to R2-3 | R2-4 to R2-5 | Keep separate from R1 | Changed placement or exposure history |
| D | Targeted hollow or suspect area | Locations chosen from condition map | As directed by investigation | Diagnostic results, not a representative lot mean | Delamination boundary or extent remains unresolved |
How do core depth, adhesive, and loading affect the result?
Core through the repair interface and into the substrate to the depth required by the adopted procedure. DeFelsko's 2017 v5 C1583 kit addendum specifies a 50 mm core and at least 10 mm below the interface in its SI instructions. Its separately printed 0.5-inch value converts to 12.7 mm; preserve the selected unit system instead of mixing those minima. DeFelsko C1583 kit addendum v5 (2017)
Clear the full cutting depth before drilling, including the repair thickness and substrate penetration. Locate tendons, reinforcement and services through the concrete scanning and cutting and coring plan; relocate unresolved conflicts. DelDOT DCT 285 identifies reinforcement damage as a risk when coring approaches the cover depth. DelDOT DCT 285, §8.2.1
Total cut depth includes repair thickness plus penetration below the interface. A 1.5-inch total core is not a 1.5-inch substrate penetration. DelDOT DCT 285 found that coring depth affected reported strength in its UHPC experiments; keep geometry comparable and record drilling damage. DelDOT DCT 285 (2022)
FHWA-HRT-22-087 applied 5 ± 2 psi/s and penetrated 1 inch into substrate in its implementation citing C1583-20. Anvilfield estimates that 5 psi/s on a nominal 2-inch specimen requires approximately 15.71 lbf/s, calculated as stress rate times circular area. Change the force-rate setting when specimen diameter changes. FHWA-HRT-22-087 (2022)
Keep the loading axis perpendicular and supports stable, using the equipment's specified arrangement for the surface orientation. Dry the disk attachment surface as the adhesive requires, separately from the repair's placement moisture condition. Identify retained coatings or sealers in the tested assembly. Germann catalog 2026 DeFelsko kit addendum v5
Repair age, test-glue cure and traffic reopening are three different clocks. Germann's 2026 catalog gives GRA adhesive 2 to 5 minutes at 15 to 25°C, with longer curing in colder conditions. DelDOT's laboratory procedure used Sikadur-32 with at least 24 hours; its field specimens used Sikadur 30 with at least 72 hours. Track concrete curing and protection independently of those adhesive waits. Germann Instruments Catalog 2026 DelDOT DCT 285 (December 2022)
OSHA 29 CFR 1926.1153(c), Table 1 item (vi), specifies integrated water delivery to the cutting surface for rig-mounted core saws or drills, with operation and maintenance following the manufacturer. Section (c)(2) also requires sufficient water flow and exhaust indoors as needed to limit visible dust; equipment outside that entry needs the applicable controls or the paragraph (d) exposure-assessment route. OSHA silica standard, checked November 5, 2026
OSHA's December 2017 core-drilling fact sheet calls for slurry cleanup before it dries and GFCIs with watertight electrical connectors when using water around electric equipment. Keep those controls in the coring work plan. OSHA FS-3929
| Source/edition | Geometry and sequence | Adhesive/cure | Practical rule |
|---|---|---|---|
| DeFelsko C1583 kit, manual v5, 2017 | Core first; 50 mm; minimum 10 mm below interface in its SI instructions | Manufacturer cure; calls for C881 Type IV, Grade 3; supplied ResinLab epoxy does not meet that specific requirement | A kit's supplied glue is not automatic method compliance; prevent glue entering annular cut |
| Germann catalog 2026, BOND-TEST pp. 12-13 | Bond disk then core; penetration at least 25 mm or half core diameter, whichever is greater | GRA: 2 to 5 minutes at 15 to 25°C | 75 mm core requires 37.5 mm penetration under this procedure |
| FHWA-HRT-22-087, 2022 | 2-inch disk; core 1 inch into substrate after glue cure; 5 ± 2 psi/s | Cure before coring | Study implementation citing ASTM 2020 |
| DelDOT DCT 285, December 2022 §6.2, laboratory | 2-inch cores; research varied penetration; Proceq DY-216; 5 psi/s | Sikadur-32; at least 24 hours | Laboratory protocol; keep separate from its field preparation |
| DelDOT DCT 285, December 2022 §7.4, field | Three cores per shoulder; 1.5-inch total depth; 2-inch steel pucks on polyester polymer concrete (PPC) over ultra-high performance concrete (UHPC) | Sikadur 30; at least 72 hours | Field preparation for the published worked example |
How do you calculate pull-off stress?
Germann Instruments' 2026 BOND-TEST procedure calculates pull-off strength by dividing the failure force by the circular cross-sectional area of the partial core. DeFelsko's v5 addendum calls for two diameter measurements at right angles and their average. Germann Instruments Catalog 2026 DeFelsko C1583 kit addendum v5 (2017)
Use average diameter d = (d1 + d2)/2, area A = πd²/4, and stress f = F/A. Newtons divided by square millimetres give MPa; pounds-force divided by square inches give psi. Convert kilonewtons to newtons before dividing by square millimetres, and retain unrounded area until the final displayed result.
Anvilfield estimates 2.04 MPa, approximately 295 psi, for 4,000 N on a measured 50.0 mm specimen. At the same force, 52.0 mm gives 1.88 MPa, approximately 273 psi. Entering 50 mm for an actual 52 mm specimen overstates stress by 8.16% because diameter is squared in the area calculation.
Use the measured specimen cross-section, excluding the annular kerf. Neither disk sidewall area nor mixed-fracture percentages change that denominator. Method: the Anvilfield estimates below use Germann's equation, the listed diameters and 4,000 N; psi values convert the unrounded MPa results, with rounding only at the end.
| Input/result | Value | Basis |
|---|---|---|
| Calculation-case peak load | 4,000 N | Anvilfield calculation input |
| Diameter readings | 50.0 mm and 50.0 mm | Anvilfield calculation inputs |
| Average diameter | 50.0 mm | Arithmetic |
| Area | 1,963.50 mm² | πd²/4 |
| Nominal tensile stress | 2.04 MPa, approximately 295 psi | F/A; conversion for reader reference |
| Same force on 52.0 mm specimen | 1.88 MPa, approximately 273 psi | Area 2,123.72 mm² |
| Overstatement if 50 mm is used for actual 52 mm | 8.16% | Ratio of areas |
| Same 4,000 N on 75.0 mm specimen | 0.91 MPa, approximately 131 psi | Area 4,417.86 mm² |
Failure-plane atlas: what broke?
Inspect both mating faces and the specimen edge before classifying a fracture. Identify old concrete, repair, primer and test adhesive from the assembly record; color alone does not establish material identity. Germann catalog 2026 DelDOT DCT 285
A cohesive fracture passes through a material; an adhesive fracture follows an interface. Shallow substrate failure through a bruised or microcracked preparation zone needs a different investigation from a deeper cohesive break. Compare fracture depth with prepared-substrate results and preparation records. A rough-looking surface can still contain a weak near-surface layer.
An interface fracture directs attention to preparation, contaminants, bond-coat condition, moisture, placement and cure, but does not identify which caused the separation. Match material on both faces to distinguish a split inside a primer from separation at its boundary. Use slab moisture testing when the moisture question calls for its own measurement rather than treating the fracture as a moisture test.
Record mixed fractures with estimated area percentages totaling 100%, the estimation method and changes in fracture depth. DelDOT DCT 285 Figure 38 uses paired-face image analysis to identify 22.54% substrate, 59.06% interface and 18.40% overlay failure. Visual field estimates should use precision consistent with the photographs. DelDOT Figure 38
Failure-plane exercise: open the published photographs in DelDOT DCT 285, Table 10, printed page 71. Cover the strength labels and classify W1 and W3 first. W1 is cohesive polyester polymer concrete (PPC) overlay failure at 556.6 psi; W3 is mixed substrate/interface failure at 593.3 psi. The assembly is PPC over ultra-high performance concrete (UHPC) on bridge BR 1-251. Table 10 shows one fracture face per specimen; Figure 38 supplies the paired-face analysis.
Both exercise specimens exceed the 250 psi individual minimum used in the retest table, but their different failure modes still need separate interpretation. A substrate fracture at 180 psi would leave the interface's ultimate strength unmeasured and fall below that specified minimum.
Photograph each specimen with its ID, scale, top/bottom labels and edge view before moving fragments. Caption the observed plane separately from suspected cause. ICRI lists 210.3R-2022 for in-situ tensile pull-off evaluation. ICRI technical guidelines catalog
| Observed break | What the faces/section show | What the number measures | What it does not prove | Next check |
|---|---|---|---|---|
| Cohesive in substrate | Old concrete attached to repair; break below original interface | Substrate tensile resistance in tested zone | Ultimate interface strength or automatic pass | Depth, soundness, strength versus specification |
| Repair/substrate interface | Separation follows original interface; little transfer for a pure break | Direct tensile bond at that interface | A unique preparation error | Prep, primer, moisture, placement, cure records |
| Cohesive in repair | Repair material on both mating faces | Repair material's in-place tensile resistance | Ultimate interface strength | Test age, mixing water, consolidation, cure |
| Within bonding-agent layer | Identifiable primer/bond coat splits internally | Resistance of that layer within assembly | Failure of substrate or test glue | Product identity, thickness, application window |
| Test adhesive or disk attachment | Disk/glue separates from specimen or glue splits | Limit of test attachment | Repair interface strength | Correct adhesive/preparation/cure and repeat |
| Mixed substrate/interface/repair | Several materials and depths on paired faces | Assembly result with multiple mechanisms | Pure interface strength or whole-area condition | Map fractions, retain individually, apply agreed grouping |
Which tests are invalid, and which are valid failures?
A low stress alone is not grounds to discard a test. Retain a valid substrate, interface or repair fracture and compare it with the acceptance rule. Germann's 2026 catalog calls disk attachment failure inconclusive for bond evaluation; DelDOT DCT 285 §6.2 rejects and repeats attachment failures at the puck/overlay. Germann Instruments Catalog 2026 DelDOT DCT 285 (December 2022)
A pre-test core break supplies no loaded result; adhesive bridging defeats isolation; eccentric loading introduces bending. Use the following procedure controls to identify exceptions and their disposition.
A fracture partly involving test adhesive remains an attachment exception. Preserve all material fractions and apply the adopted method's replacement rule. A coating method's attachment-area cutoff does not automatically govern a C1583 repair test.
A specimen that reaches instrument capacity without breaking has sustained the recorded stress, but its ultimate strength remains unknown. DeFelsko lists upper ranges of 3.3 MPa/480 psi for its AT-M and 3.8 MPa/560 psi for its AT-A 50 mm C1583 configurations. Record the no-fracture status and keep the result outside a failure-strength mean; use it as proof only when the contract expressly permits that decision. DeFelsko PosiTest AT dollies
| Observation | Classification | Keep in report? | Action/source |
|---|---|---|---|
| Pure test-glue or disk attachment failure | Inconclusive for repair bond; replace test | Yes, mark attachment failure | Correct glue/cure/cleaning; new location, Germann 2026/DelDOT DCT 285 |
| Core breaks during drilling or handling | No valid loaded pull-off result | Yes, record pre-test break | Document cause and new location; Anvilfield procedure control |
| Glue fills kerf and bridges surrounding repair | Compromised isolation | Yes, exception | Correct technique; replacement specimen, DeFelsko v5 (2017) |
| Tilted fixture, bending, damaged core | Procedural exception requiring review | Yes, photographs and setup | Correct setup and repeat under approved procedure, Germann 2026 |
| Wrong specimen area entered | Potentially correctable calculation error; also check actual loading rate | Yes, original and correction | Recalculate only when raw force, dimensions, geometry and actual loading rate establish a conforming test; otherwise repeat. Germann 2026/DeFelsko v5 plus Anvilfield procedure control |
| Valid low substrate/interface/repair fracture | Valid material result | Yes | Investigate against project criterion; no deletion, ASTM C1583-25 |
| Load limit with no fracture | Specimen withstood applied nominal stress; ultimate not measured | Yes, raw force, dimensions, limit, and no-fracture status | Report a lower bound, exclude from a failure-strength mean, and apply only a contract-authorized proof rule; DeFelsko ranges plus Anvilfield interpretation |
| Mixed material fracture without attachment defect | Mixed assembly result | Yes, faces and fractions | Do not automatically invalidate or pool with pure modes, ASTM C1583-25/DelDOT DCT 285 |
| Fracture partly in test adhesive/attachment | Attachment participates; repair strength inconclusive | Yes, attachment area and all material fractions | Apply the adopted replacement rule; retain attachment involvement. Germann 2026/DelDOT DCT 285 |
Is 250 psi a passing result?
250 psi passes only when it satisfies the project's strength, age and failure-mode rules. ASTM C1583 supplies the test method; product data and agency specifications supply application-specific acceptance values. ASTM C1583/C1583M-25
SikaQuick Smooth Finish's March 2020 data sheet reports greater than 250 psi with substrate failure at 28 days and 73°F/50% relative humidity, a product result rather than a universal ASTM acceptance limit. Louisiana DOTD's July 1, 2020 qualification procedure requires at least 250 psi at 28 days by ASTM C1583 for Type B cementitious and polymer-modified vertical or overhead patching materials. SikaQuick Smooth Finish PDS (March 2020) Louisiana DOTD patching qualification (July 1, 2020)
A substrate fracture below a specified stress floor does not automatically waive that floor. SCDOT/Clemson's July 2025 HFST report uses 250 psi or 100% substrate failure for its research evaluation. That application-specific alternative belongs to its acceptance context; it is not an exemption for every repair. SCDOT/Clemson FHWA-SC-25-02 (July 2025)
Near a threshold, retain raw force, measured diameter, displayed stress, units and the contract's rounding rule. A display rounded to 250 psi does not settle a borderline decision by itself. Apply the agreed rule without inventing a tolerance or averaging away a valid low result where the contract requires an individual minimum.
| Authority/document | Application | Number and age | Meaning | Use on this job |
|---|---|---|---|---|
| ASTM C1583-25 public scope/§5 | Concrete surfaces, repairs, overlays | No general acceptance stress supplied by these provisions | Test and interpretation framework | Identify the project's separate acceptance requirement |
| SikaQuick Smooth Finish PDS, March 2020 v01.03 | Vertical/overhead reprofiling mortar | >250 psi at 28 days, substrate failure, 73°F/50% RH | Published product performance | Compare like conditions; do not convert into blanket field requirement |
| Louisiana DOTD, July 1, 2020, Table 3 | Type B cementitious/polymer-modified patch qualification | Minimum 250 psi at 28 days | Qualification requirement | Apply only when that material qualification is relevant |
| SCDOT/Clemson FHWA-SC-25-02, July 2025 §5.3.4 | HFST research evaluation | 250 psi or 100% failure in substrate | Report's account of its agency acceptance criterion | Verify contract adoption; not a generic repair exemption |
| Anvilfield worked acceptance case | Worked repair lot | Every valid pull ≥250 psi at specified age; no substrate exemption | Worked example specification | A 180 psi substrate break is below this case's floor |
Worked example: read the fracture before averaging
Group comparable results by failure mode before calculating a mean, as ASTM C1583/C1583M-25 §5.4 requires. An acceptance group also needs compatible lot, age and procedure. ASTM C1583/C1583M-25
DelDOT DCT 285, December 2022, Table 10 reports six field results from PPC over UHPC on bridge BR 1-251. W denotes the west shoulder and E the east shoulder. The table below retains every observation, including two mixed A/B results and the discrepancy between E1's 468.3 psi in Table 10 and 463.3 psi in §8.3. These letters are the report's descriptive labels, not universal ASTM mode codes. DelDOT DCT 285 (December 2022)
Anvilfield estimates a 606.53 psi descriptive mean for W1, W2 and E2, the three Mode C results without conflicting source values. Keep the two mixed results separate. These selected observations cross shoulder locations; use the contract's lot definitions when grouping results for acceptance.
These field specimens used the §7.4 preparation listed above. E2's 722.3 psi exceeds the listed limits of the DeFelsko configurations: select tester capacity for the expected strength while retaining the adopted specimen geometry.
Anvilfield estimates W1's corresponding force as 1,749 lbf by multiplying its reported stress by the nominal 2-inch circular area. Method: calculate the three-result mean and nominal force from Table 10 stresses, retaining full precision until final rounding. Mode C measures cohesive overlay failure; the interface's ultimate strength remains unmeasured.
| Item | Input or reported observation | Result/interpretation | Source |
|---|---|---|---|
| W1 | 556.6 psi, Mode C | Cohesive overlay failure | DelDOT DCT 285 Table 10 |
| W2 | 540.7 psi, Mode C | Cohesive overlay failure | DelDOT DCT 285 Table 10 |
| W3 | 593.3 psi, mixed A/B | Keep separate from Mode C mean | DelDOT DCT 285 Table 10 |
| E1 | Table 10: 468.3 psi; §8.3: 463.3 psi; Mode C | Source discrepancy; omit from derived mean | DelDOT DCT 285 Table 10 versus §8.3 |
| E2 | 722.3 psi, Mode C | Cohesive overlay failure | DelDOT DCT 285 Table 10 |
| E3 | 567.0 psi, mixed A/B | Keep separate from Mode C mean | DelDOT DCT 285 Table 10 |
| Three unambiguous Mode C observations | (556.6 + 540.7 + 722.3)/3 | 606.53 psi descriptive same-mode mean | Anvilfield estimates from DelDOT DCT 285 |
| W1 nominal diameter reconstruction | 2.00 in; circular area π in² | Approximately 1,749 lbf corresponding force | Anvilfield estimates, nominal 2-inch geometry in DelDOT DCT 285 |
| Decision | Mode C governs selected observations | Interface held through those tests; ultimate interface capacity unknown | ASTM C1583-25/DelDOT DCT 285 interpretation |
When do you retest, reprepare, or investigate?
Give every retest a question: replace a defective attachment, verify corrected preparation or establish the extent of weak work. Preserve the original specimen IDs and identify added locations on the zone map. Additional high results do not erase valid low results.
This Anvilfield decision table uses a repair specification requiring every valid result to reach at least 250 psi at the specified age, without a substrate-failure exemption.
Map the affected work before deciding removal limits. For low substrate results, diagnose concrete surface defects and compare bare-substrate trials with the failed depth. For interface results, connect added pulls with the preparation and placement history. One local fracture does not define the boundary of a whole-deck replacement. ASTM C1583/C1583M-25 FHWA, Tensile Bond Tester DelDOT DCT 285 (December 2022)
| Result/condition | Immediate status | Next test or investigation | Release condition |
|---|---|---|---|
| Glue failure at 180 psi | Attachment failed; no valid repair bond result | Correct adhesive/cure; test reserve location | Required valid results obtained |
| Interface failure at 180 psi | Valid low result; hold represented work | Check preparation/bond coat/cure; map surrounding condition and define added pulls | Written disposition plus specified evidence after correction |
| Substrate failure at 180 psi | Valid low substrate result; below this example's floor | Compare bare-substrate results; investigate weak depth/extent | Criterion met or authorized written acceptance change |
| Repair failure at 180 psi before scheduled age | Retained early diagnostic result; not a substitute for the scheduled result | Check mixing/cure; test at already specified acceptance age | Age and strength requirements satisfied |
| Substrate failure at 320 psi | Above example floor with substrate governing | Complete remaining required valid pulls and records | Whole lot acceptance plan satisfied |
| Three intended pulls with one pure interface and two substrate failures | Different mechanisms | Report individually; obtain additional compatible results where required by the adopted method and sampling plan | Required valid count and individual criteria satisfied; do not average different modes |
| Mixed fracture | Mechanism not represented by one pure mode | Photograph both faces; classify areas and evaluate individually | Project's mixed-mode treatment applied |
| Instrument limit, no break | Ultimate strength unknown | Check whether contract permits limit result; otherwise higher-capacity setup | Required failure measurement or expressly permitted proof result |
| Documented preparation correction | New process condition | New representative test area, followed by specified production verification | Corrected work demonstrates required performance |
What should testing cost, and how should it be scheduled?
Anvilfield estimates $1,480 for ten completed attempts on the accessible R1/R2 deck, or $247 per result when six valid results are obtained. Plan eight field hours and two report hours at the specified repair age, using an adhesive that cures within the visit.
Field time covers setup, coring, adhesive waiting, pulling and routine patch labor. Add separate quantities for substrate approval trials, diagnostic testing, tendon scanning, lifts, night traffic control, structural hole repairs and return visits.
Schedule repair cure before adhesive cure. DelDOT's laboratory and field adhesive waits require separate visits compared with Germann's short GRA cure. Include scanning, setup, coring, documentation and reinstatement in the access window. Germann catalog 2026 DelDOT DCT 285
Municipal and school repair bids should identify testing lots, access windows, closure responsibility, hole reinstatement and rates for authorized additional attempts. Estimators use past award amounts from PorchRocket Bid Control to compare testing and closure allowances with earlier public repair contracts.
Method: Anvilfield estimates use November 5, 2026 planning inputs of labor hours, equipment days, attempts and travel. Multiply each quantity by its rate, sum, and divide by valid results. Changing ten labor hours to $75 to $125/hour gives $1,280 to $1,780 total, or $213 to $297 per result with six valid pulls. Unused reserves reduce consumables; attachment failures increase the cost per valid result.
| Cost item | Quantity and assigned rate | Extended cost | Inclusion |
|---|---|---|---|
| Field technician | 8 hours × $95/hour | $760 | Setup, coring, testing and routine patch labor |
| Equipment allowance | 1 day × $180 | $180 | Tester and coring equipment allowance |
| Consumables/reinstatement material | 10 attempts × $20 | $200 | Adhesive, wear, basic hole patch materials |
| Report preparation | 2 hours × $95/hour | $190 | Map, result review, photos and report |
| Travel/mobilization | 1 allowance × $150 | $150 | Accessible local visit |
| Total | Sum of five lines | $1,480 | Ten-attempt allowance |
| Six valid reportable results | $1,480/6 | $247 each | Four additional attempts allowed for replacements |
| All ten attempts valid | $1,480/10 | $148 each | Same total if all reserve tests are authorized and completed |
| Labor-rate sensitivity | 10 hours at $75 to $125; other costs unchanged | $1,280 to $1,780 total | $213 to $297 per result if six valid |
Field checklist and acceptance record
Keep the zone, coordinates, edition, repair age, conditions, dimensions, raw force, fracture faces and exceptions together. Give each attempt its own ID and connect any replacement to the original.
Retain low and invalid attempts in the construction QA/QC records. The release entry identifies the accepted lot, criterion, corrective work and decision authority. Restore test holes with the approved compatible repair and finish system, recording cure and return to service.
Want this checklist to run itself on every job — with photo proof and a signed record crews can hand the customer? That's FieldOS.
Citation and review record
Atlas citation: Anvilfield, ASTM C1583 pull-off tests: what failed, and what does it mean?, Failure-plane atlas, reviewed 2026-11-05, https://anvilfield.com/field-guides/concrete/concrete-repair-pull-off-testing-c1583/#failure-atlas, based on ASTM C1583/C1583M-25 and the identified manufacturer and agency sources.
2026-11-05 update: Rechecked publisher editions, manufacturer and agency procedures, fracture data and arithmetic; corrected links, the layout markup and calculation-error retest guidance.
How to cite this page: Anvilfield, ASTM C1583 pull-off tests: what failed, and what does it mean?, 2026, https://anvilfield.com/field-guides/concrete/concrete-repair-pull-off-testing-c1583/.
FAQ
How do you interpret an ASTM C1583 concrete repair pull-off test?
Read stress and fracture plane together, then apply the project's acceptance rule. ASTM C1583/C1583M-25 §5 explains why the weakest plane controls the measured property.
What is the C1583 failure plane?
The failure plane is the layer or interface where separation occurs. Cohesive failure passes through material; adhesive failure follows an interface. The atlas distinguishes repair bond failure from test attachment failure.
How many repair pull-off tests are required?
New York OGS Project 45604-C's 2024 specification requires at least three pulls per patching material. Anvilfield budgets three intended valid pulls plus two reserves per zone; the adopted method and contract govern the valid count, compatible modes and represented area.
How is dolly stress calculated?
Germann's 2026 BOND-TEST procedure divides peak force by the circular area calculated from measured specimen diameter. N/mm² equals MPa; the stress-calculation table shows the arithmetic.
Does concrete failure mean the bond passed?
The intact interface resisted the applied stress; acceptance still follows the specified strength and failure-mode rule. The 180 psi substrate case fails the assigned 250 psi individual floor.
Is 250 psi required by ASTM C1583?
No universal 250 psi criterion comes from the method. The threshold crosswalk separates Sika product performance, Louisiana material qualification and the SCDOT research criterion.
Can glue failure count as repair bond strength?
No. Germann's 2026 catalog calls test attachment failure inconclusive for bond evaluation; DelDOT DCT 285 repeats attachment failures. Retain the original attempt and identify its replacement.
Can different C1583 failure modes be averaged?
ASTM C1583/C1583M-25 permits averaging only results with the same failure mode. Keep unlike lots, test ages and procedures separate even when their fracture modes match.
How long should the dolly glue cure?
Germann's 2026 GRA procedure gives 2 to 5 minutes at 15 to 25°C. DelDOT's December 2022 study used at least 24 hours for laboratory Sikadur-32 and 72 hours for field Sikadur 30. Follow the adhesive protocol separately from repair age.
Does one good pull prove the whole overlay is bonded?
A pull characterizes its local specimen; the defined sampling plan and wider condition survey establish what that result represents.
How deep should a C1583 core go below the repair?
DeFelsko v5 states at least 10 mm below the interface in SI; Germann 2026 uses 25 mm or half the core diameter, whichever is greater. Follow the adopted procedure, without merging these depths.
At what age should a concrete repair be pull-off tested?
The cited SikaQuick Smooth Finish performance is a 28-day value at 73°F/50% RH; field acceptance uses the age in the repair specification. Repair chemistry, temperature, and curing history control strength development.
People also ask
Codes cited in this guide
This guide is written and reviewed against the published standards below. Always confirm the current adopted edition with the authority having jurisdiction.
Sources checked
- ASTM C1583/C1583M-25 Public scope and §5 interpretation; Reopened with curl 2026-11-05.
- ASTM D7234-22 Public scope and §5.5 method distinction; Reopened with curl 2026-11-05.
- FHWA, Tensile Bond Tester Undated equipment overview, not the current ASTM procedure. Reopened with curl 2026-11-05.
- FHWA-HRT-22-087 (2022) Research procedure citing ASTM 2020; three pulls per region and loading rate. Reopened with curl 2026-11-05.
- Germann Instruments Catalog 2026 BOND-TEST, printed pages 12-15; force/area, geometry and GRA cure. Reopened with curl 2026-11-05.
- DeFelsko C1583 kit addendum v5 (2017) Dated manufacturer instructions; SI depth, diameter measurement and epoxy limitation. Reopened with curl 2026-11-05.
- SikaQuick Smooth Finish PDS (March 2020) Version 01.03; 28-day tensile adhesion and conditioning. Reopened with curl 2026-11-05.
- Louisiana DOTD patching qualification (July 1, 2020) Table 3, Type B material qualification, not general field acceptance. Reopened with curl 2026-11-05.
- DelDOT DCT 285 (December 2022) Laboratory §6.2; field §7.4; Figure 38 and Table 10; E1 discrepancy retained. Reopened with curl 2026-11-05.
- SCDOT/Clemson FHWA-SC-25-02 (July 2025) §5.3.4, HFST research evaluation criterion. Reopened with curl 2026-11-05.
- DeFelsko PosiTest AT dollies Undated manufacturer-listed ranges for 50 mm C1583 configurations. Reopened with curl 2026-11-05.
- ICRI technical guidelines catalog Catalog verifies 210.3R-2022 listing; full guideline text not used. Reopened with curl 2026-11-05.
- New York OGS 45604-C, Addendum 1 (October 17, 2024) §030131 1.05.E.5, at least three pull-off tests per patching material. Reopened with curl 2026-11-05.
- OSHA 29 CFR 1926.1153, construction silica standard Current federal text, (c)(1) Table 1 item (vi), (c)(2) and (d); read 2026-11-05 via browser. Direct curl returned 403.
- OSHA FS-3929, Rig-Mounted Core Saws or Drills (December 2017) Wet methods, slurry cleanup and electrical protection; read both pages 2026-11-05 via browser. Direct curl returned 403.