ASTM D1781
Climbing Drum Peel Test Fixture
Downloads: ASTM D1781 PDF (PDF) · ASTM D1781 Peel Fixture – Universal Grip Data Sheet (PDF)
Written by Chris Thomas, Universal Grip Company · Reviewed by Frank Bertini · Updated October 23, 2026
ASTM D1781 is the standard test method for climbing drum peel for adhesives. A flanged drum, pulled by straps wound around its flanges, rolls up a bonded facing and peels it away from a rigid adherend or a sandwich core. The result is average peel torque per unit width, a comparative measure of how well the facing is bonded.
This guide covers the drum geometry, specimens, test speed, the peel torque formula with a worked example in both unit systems, failure modes, common errors, how D1781 compares with EN 2243-3 and other peel tests, and the climbing drum fixture Universal Grip builds for it.
Note: Current edition: ASTM D1781-98 (Reapproved 2021), ASTM Committee D14 on Adhesives, Subcommittee D14.80 on Metal Bonding Adhesives. Buy the current text from ASTM and work from the edition your specification names.
What ASTM D1781 measures and who uses it
D1781 measures the peel resistance of adhesive bonds in two situations: between a relatively flexible adherend and a rigid adherend, and between the relatively flexible facing of a sandwich structure and its core. The typical specimen is a honeycomb or foam-core sandwich panel with thin metal or composite skins.
The test is used by aerospace interiors and floor-panel makers, rail and transit panel suppliers, marine and truck-body builders, adhesive formulators and honeycomb producers. Most of them run it for process control: to confirm that surface preparation, primer, adhesive and cure produced a sound facing-to-core bond, batch after batch.
A comparative test, not a design value
The standard is explicit that climbing drum peel gives comparative adhesion data, not a fundamental property. Results depend on facing thickness and stiffness, core type and density, and test conditions, so values can be compared directly only when specimen design and test conditions are identical. The test is very sensitive to adherend surface preparation, which is exactly why it suits process control and primer checks.
One note from the standard is easy to miss: with brittle adhesives, unusually high peel values can indicate incomplete cure. A higher number is not automatically a better bond.
ASTM D1781 at a glance
Standard: ASTM D1781-98 (Reapproved 2021), Climbing Drum Peel for Adhesives
Result: average peel torque per unit width, in N·mm/mm or in·lbf/in
Drum: 102 mm (4.000 in) diameter; outside radius 51.00 ± 0.13 mm (2.000 ± 0.005 in)
Flanges: radius 12.70 ± 0.13 mm (0.500 ± 0.005 in) larger than the drum radius, about 127 mm (5 in) diameter
Drum assembly mass: no more than 3.6 kg (8 lb), lighter preferred
Sandwich specimens: 76 mm (3 in) wide, at least 305 mm (12 in) long, facing overhanging about 25 mm (1 in) at each end
Laminate specimens: 25 mm (1 in) wide, at least 254 mm (10 in) long
Reference construction: 0.51 mm (0.020 in) clad 2024-T3 aluminum facing (ASTM B209) on a 12.7 mm (0.5 in) core
Specimens: at least six per adhesive; honeycomb split between L and W directions
Crosshead speed: 25.4 ± 2.54 mm/min (1.00 ± 0.10 in/min)
Load range: peak loads between 15 % and 85 % of full scale; machine verified to ASTM E4
Formula: T = (RO − RI)(FP − FO) / W
How the climbing drum works
The fixture is a hollow drum with a larger-diameter flange on each side. The free end of the specimen facing is clamped to the drum surface. Flexible straps or cables are wound around the flanges and attached to one crosshead, while the other end of the specimen is held in a grip on the opposite crosshead. Counterweights opposite the drum clamp balance the drum about its axis so it has no tendency to turn on its own.
When the crossheads separate, the crosshead force F acts through the straps at the flange radius RO, while the pull of the facing acts at the drum radius RI. Because the two radii differ, the force produces a net torque of F × (RO − RI) that turns the drum. As it turns, the drum winds the facing onto its surface and climbs up the specimen, peeling the facing from the core at a controlled, constant bend radius.
Why the torque arm is 12.7 mm (0.5 in)
D1781 sets the flange radius 12.70 mm (0.500 in) larger than the drum radius, measured to the centerline of the straps. The facing wraps around the drum, so its centerline sits half a facing thickness outside the drum surface. The effective torque arm is therefore:
RO − RI = 12.70 mm − t / 2 (or 0.500 in − t / 2), where t is the facing thickness
For the reference 0.51 mm (0.020 in) facing, the arm is 12.70 − 0.254 = 12.446 mm (0.490 in). Ignoring the half-thickness term overstates torque by about 2 % for that facing, and by more for thicker skins.
Why a resisting load FO is subtracted
Part of the crosshead force does no peeling at all. It goes into turning and lifting the drum, flexing the straps, and bending the facing around the drum. That part is called FO, the load required to overcome the resisting torque of the drum. Subtracting it from the average peel load FP leaves the force that actually separated the bond. Labs obtain FO in different ways, for example by running the drum without a bonded specimen or by re-running over the already-peeled facing, which also captures facing bending. Use the method in the current edition and record which one you used.
Specimens
Laminated (flexible-to-rigid) assemblies
Width: 25 mm (1 in), cut between saw kerfs
Length: at least 254 mm (10 in)
Sandwich constructions
Width: 76 mm (3 in)
Length: at least 305 mm (12 in)
Overhang: about 25 mm (1 in) of the facing to be peeled extends beyond the core at each end for clamping
Core thickness: not critical, provided the specimen does not bend during the peel
Reference metal-faced construction: 0.51 mm (0.020 in) clad 2024-T3 aluminum facing per ASTM B209 on a 12.7 mm (0.5 in) core
Honeycomb ribbon direction
Honeycomb and other orthotropic cores bond differently along and across the ribbon. D1781 calls for half the specimens to be peeled parallel to the core length (L, or ribbon) direction and half parallel to the width (W) direction. Mark the direction on each specimen before cutting, and report L and W results separately; averaging them hides a real difference.
Number of specimens
Test at least six specimens per adhesive, for example two or more from each of three separately bonded panels. Specimens from more than one panel catch panel-to-panel variation in surface preparation and cure, which is usually what process control is looking for.
Preparation
Prepare and bond the panels following the adhesive manufacturer's recommendations for surface preparation, primer, bond-line thickness, pressure and cure. Cut specimens with a saw or router that does not heat or delaminate the edges, and remove the core from the overhang regions carefully so the facing is not kinked. A pre-bent or creased facing at the drum clamp gives a high starting spike and can tear the skin.
The climbing drum fixture and top grip
Detailed drawings of the D1781 drum are published as an ASTM adjunct, and a compliant fixture must hold the drum and flange radii to the tolerances above. Key points when choosing or checking a fixture:
Drum diameter 102 mm (4.000 in) with the flange radius 12.70 mm (0.500 in) larger, measured to strap centerline.
Total mass of drum, flanges, drum clamp and counterweights no more than 3.6 kg (8 lb). A lighter drum means a smaller FO and less inertia.
A drum clamp that holds the full facing width squarely without crushing it.
Counterweights that balance the drum so it does not rotate on its own at rest.
Straps or cables that are flexible, inextensible and wound evenly on both flanges so the drum climbs level.
The upper end of the specimen is held in a separate grip. For 76 mm wide sandwich specimens with a thick core you need a wide jaw with plenty of opening, such as our extra wide vise grips; standard vise grips handle 25 mm laminates. The grip must hold the specimen vertical and centered over the drum so the facing peels square.
Note: A D1781 fixture is not an EN 2243-3 fixture. The European aerospace method uses a different drum and flange geometry, so the torque arm and the results differ.
Testing machine, load cell and speed
Any universal testing machine that holds a constant crosshead rate under a fluctuating load will run D1781. The standard calls for load accuracy within 1 %, verification to ASTM E4, and autographic recording of the load-displacement curve, which today means software that captures the full peel trace.
Load cell sizing
Peel loads are modest, typically tens to a few hundred newtons, but the standard requires peak loads to fall between 15 % and 85 % of full scale. A 267 N (60 lbf) peak is only 5 % of a 5 kN cell but 27 % of a 1 kN cell, so a smaller cell, or a well-characterized lower range, is often the right choice. A 5 kN or 10 kN frame has ample capacity; the load cell is what matters.
Crosshead speed
Run the crosshead at 25.4 ± 2.54 mm/min (1.00 ± 0.10 in/min). Some peel guides quote 50.8 mm/min (2 in/min) for D1781; that is not the standard's rate. Speed changes peel values, especially with tough or viscoelastic adhesives, so hold it within tolerance for every specimen.
Daylight
Check vertical space before you buy or set up. The machine must fit the top grip, a specimen of at least 305 mm (12 in), the drum and flanges, the straps and the lower coupling, plus room to peel the full length.
Conditioning and specimen preparation
Condition specimens before testing as the standard specifies. Labs commonly use the standard laboratory atmosphere of 23 °C (73.4 °F) and 50 % relative humidity; confirm the tolerances and minimum conditioning time in the current edition, or follow the material specification if it sets its own.
Test in the same atmosphere where possible. Moisture absorbed by the adhesive or the core, particularly aramid-paper honeycomb, can change peel values noticeably.
Hot and cold peel
Specifications for aircraft interiors and transport panels sometimes require peel at elevated or low temperature. Fit an environmental chamber around the fixture, let the drum, grip and specimen reach temperature, and remember that FO should be determined at the same temperature as the peel.
Step-by-step ASTM D1781 procedure
Measure the specimen width W at several points and average it; record the facing thickness t and the core direction (L or W).
Condition the specimens as specified.
Install the top grip and the climbing drum fixture; check that the drum is balanced, the straps are wound evenly on both flanges and the specimen will hang vertically over the drum center.
Zero the load cell with the fixture installed and no specimen load applied.
Clamp the facing overhang squarely in the drum clamp, then clamp the opposite end of the specimen in the top grip without preloading it.
Set the crosshead speed to 25.4 mm/min (1.00 in/min).
Start the test and record load against crosshead displacement as the drum climbs and peels the facing.
Peel a sufficient length of bond, typically at least about 150 mm (6 in), and disregard the initial portion of the curve where the drum is starting to climb, as the standard directs.
Stop the test before the drum reaches the far end of the bond or the top grip.
Determine FO by the method in the current edition and record how it was obtained.
Note the failure mode over the peeled length: adhesive, cohesive, core or mixed.
Repeat for at least six specimens, half in each core direction for honeycomb.
Calculate average peel torque per width for each specimen and report the average, minimum and maximum peel loads, failure modes and test conditions.
Peel torque calculation with a worked example
The formula
Average peel torque per unit width: T = (RO − RI)(FP − FO) / W
T: average peel torque per unit width, N·mm/mm or in·lbf/in
RO: flange radius plus half the strap thickness
RI: drum radius plus half the facing thickness
RO − RI: 12.70 mm − t/2 (0.500 in − t/2) on a D1781 drum
FP: average load required to bend and peel the facing, taken over the valid part of the curve
FO: load required to overcome the resisting torque of the drum
W: specimen width
Note the units: N·mm/mm reduces to N, and in·lbf/in reduces to lbf, but report them as torque per width so nobody mistakes them for a peel force. Conversion: 1 in·lbf/in = 4.448 N·mm/mm.
Worked example, inch-pound
Sandwich specimen width W = 3.00 in
Facing thickness t = 0.020 in
Average peel load FP = 60 lbf
Resisting load FO = 15 lbf
Torque arm = 0.500 − 0.020 / 2 = 0.490 in
T = 0.490 × (60 − 15) / 3.00 = 0.490 × 45 / 3.00 = 7.35 in·lbf/in
Same example, SI
W = 76.2 mm; t = 0.508 mm
FP = 266.9 N; FO = 66.7 N; FP − FO = 200.2 N
Torque arm = 12.70 − 0.508 / 2 = 12.446 mm
T = 12.446 × 200.2 / 76.2 = 32.7 N·mm/mm
Check: 7.35 in·lbf/in × 4.448 = 32.7 N·mm/mm
You may also see results quoted per 3 in specimen width (in·lb per 3 in). That is simply T × 3: 7.35 × 3 = 22.05 in·lb per 3 in. Always state which basis you report.
What the half-thickness correction is worth
If the 0.500 in arm is used without subtracting t/2, the same data give 0.500 × 45 / 3.00 = 7.50 in·lbf/in, about 2 % high. With a 0.040 in facing the error doubles to about 4 %. Leaving out FO is far worse: 0.490 × 60 / 3.00 = 9.80 in·lbf/in, which overstates the bond by a third.
Reading the curve and failure modes
A good climbing drum curve rises to a starting peak as the drum begins to turn, then settles into a band of load that fluctuates around a steady average as the peel front moves along the bond. Honeycomb cores produce a regular saw-tooth pattern as the facing lets go of each row of cells. FP is the average over the valid region; the maximum and minimum within that region show how uniform the bond is.
Failure modes
Adhesive (interfacial) failure: the adhesive releases cleanly from the facing or the core. Usually points to surface preparation, contamination or primer problems, which is what process control is meant to catch.
Cohesive failure: the adhesive splits within itself, leaving adhesive on both surfaces. The bond reached the adhesive's own strength; this is generally the desired mode.
Core failure: honeycomb cell walls, nodes or foam tear and stay attached to the peeled facing. The bond is stronger than the core; the result reflects core strength more than adhesion.
Fillet failure: in honeycomb, the adhesive fillets at the cell walls fracture. Look at fillet size and wetting.
Facing yield or kinking: the skin yields, creases or tears instead of peeling smoothly. The result is invalid as a peel value; a thicker facing or different construction may be needed.
Mixed: estimate and report the percentage of each mode over the peeled area.
Read the number and the fracture surface together. A value that looks acceptable with mostly adhesive failure is a warning sign; a modest value with core failure may simply mean the core is the weak link.
Common errors
Not subtracting FO, or measuring it at a different temperature or speed from the peel.
Using the nominal 12.7 mm (0.5 in) arm without the half-thickness correction.
Reporting peel force per width (N/mm) instead of peel torque per width.
Running at 50.8 mm/min (2 in/min) instead of 25.4 mm/min (1 in/min).
Testing fewer than six specimens, or all from one panel.
Mixing L-direction and W-direction honeycomb results in one average.
Using a fixture built to the EN 2243-3 geometry and reporting the result as D1781.
Cutting 25 mm wide sandwich specimens; sandwich specimens are 76 mm (3 in) wide.
A core so thin or flexible that the specimen bends during the peel.
Specimen not vertical or drum not centered, so the facing peels at an angle and one edge leads.
Including the starting peak in the average, which inflates FP.
Load cell so large that peaks fall below 15 % of full scale.
Using a drum assembly heavier than 3.6 kg (8 lb).
Citing the wrong reference: the clad aluminum specification is ASTM B209, and the designation is D1781, current edition 98 (Reapproved 2021).
ASTM D1781 vs EN 2243-3, DIN 53295 and related peel tests
European drum peel standards
The closest counterparts to D1781 are EN 2243-3, the European aerospace method for peel testing of sandwich structures with a climbing drum, and DIN 53295, the German sandwich drum peel method. Fixture makers list EN-style drums at about 100 mm diameter with 125 mm flanges, a nominal 12.5 mm torque arm, against D1781's 102 mm drum and 12.7 mm arm. The specimen requirements also differ, so check the EN text for its own specimen and speed requirements. Fixtures and results are not interchangeable between the two.
ISO 11339 and ISO 4578 are sometimes listed as D1781 equivalents. They are not: ISO 11339 is a T-peel test, the counterpart of ASTM D1876, and ISO 4578 is a floating roller peel test, the counterpart of ASTM D3167.
Other ASTM peel tests
ASTM D3167, floating roller peel: a thin flexible metal adherend is peeled from a rigid one around a free-turning roller. Used for metal-to-metal bonds; the result is peel force per width.
ASTM D1876, T-peel: two flexible adherends pulled apart in a T. Used for flexible-to-flexible bonds.
ASTM D903, 180° peel, and ASTM D6862, 90° peel: flexible adherend peeled from a rigid one at a fixed angle, without a drum.
D1781 is the choice when the facing is too stiff for a T-peel and the substrate is a sandwich core. See our peel testing page for fixtures for every method.
Sandwich panel companion tests
Climbing drum peel is usually one test in a sandwich panel qualification program. Core shear strength and modulus come from ASTM C273, beam flexure from ASTM C393, flatwise tension from ASTM C297 and edgewise compression from ASTM C364. Our sandwich panel testing guide explains how they fit together.
ASTM D1781 equipment from Universal Grip
Universal Grip designs and builds the ASTM D1781 climbing drum peel fixture shown on this page and supplies the grips and machines to run it:
ASTM D1781 climbing drum peel fixture: 102 mm (4 in) drum per D1781, 5 kN maximum capacity, specimens up to 80 mm wide (covers 76 mm sandwich and 25 mm laminate specimens), about 3 kg so it stays within the 3.6 kg limit, aluminum and steel construction, 5/8 in female coupling with 5/16 in pin, optional adjustable start table. The top grip is sold separately.
Top grip: extra wide vise grips (5 kN, 10 to 100 mm specimens) for sandwich panels; standard vise grips for laminates.
Testing machines: Galdabini universal testing machines, including 5 kN and 10 kN benchtop frames with load cells sized so peel peaks fall within 15 to 85 % of range.
Environmental chambers: temperature chambers for hot and cold peel.
The video on this page shows how to set up and run the fixture, and our climbing drum fixture article covers it in more depth. For other peel methods see the peel fixtures guide and adhesive testing.
To get a quote, contact us with your panel construction (facing material and thickness, core type and thickness), specimen width, expected peel load and your testing machine model, and we will confirm fit and recommend the top grip and load cell.
ASTM D1781 FAQ
What is the ASTM D1781 climbing drum peel test?
ASTM D1781 is a peel test for adhesive bonds in which a flanged drum, pulled by straps wound around its flanges, rolls up a bonded facing and peels it from a rigid adherend or a sandwich core. The result is average peel torque per unit width. It is mainly used to compare bonds and check surface preparation, primer and cure in sandwich panel production.
What does the climbing drum peel test measure?
It measures comparative peel resistance of the bond between a relatively flexible facing and a sandwich core, or between a flexible adherend and a rigid one. The standard describes the data as comparative rather than fundamental, so values can be compared only for identical specimen designs and test conditions. It is most useful for process control and surface-preparation checks.
What size are ASTM D1781 specimens?
Sandwich specimens are 76 mm (3 in) wide and at least 305 mm (12 in) long, with about 25 mm (1 in) of the facing overhanging the core at each end for clamping. Laminated flexible-to-rigid specimens are 25 mm (1 in) wide and at least 254 mm (10 in) long. The reference construction is a 0.51 mm clad 2024-T3 aluminum facing on a 12.7 mm core.
What test speed does ASTM D1781 use?
The crosshead speed is 25.4 ± 2.54 mm/min (1.00 ± 0.10 in/min). Some peel guides quote 50.8 mm/min (2 in/min) for D1781, but that is not the standard's rate. Speed affects peel values, especially with tough or viscoelastic adhesives, so hold it within tolerance for every specimen and record it in the report.
How many specimens does ASTM D1781 require?
Test at least six specimens per adhesive, for example two or more from each of three separately bonded panels. For honeycomb and other orthotropic cores, peel half the specimens parallel to the core length (L or ribbon) direction and half parallel to the width (W) direction, and report the two directions separately.
How is climbing drum peel torque calculated?
Average peel torque per unit width is T = (RO − RI)(FP − FO) / W. RO − RI is the 12.70 mm (0.500 in) flange offset minus half the facing thickness, FP is the average peel load, FO is the load needed to overcome the drum's resisting torque, and W is the specimen width. Example: 0.490 × (60 − 15) / 3.00 = 7.35 in·lbf/in.
What units are used for climbing drum peel results?
Results are reported as torque per unit width, in N·mm/mm or in·lbf/in. To convert, 1 in·lbf/in equals 4.448 N·mm/mm. Results are sometimes quoted per 3 in specimen width (in·lb per 3 in), which is simply the per-inch value multiplied by three. Always state which basis you are reporting.
Why is FO subtracted in the D1781 calculation?
Part of the crosshead force only turns and lifts the drum, flexes the straps and bends the facing; it does not separate the bond. FO is the load needed to overcome that resisting torque. Subtracting it from the average peel load leaves the part that actually peeled the facing. Leaving FO out can overstate peel torque by a third or more.
What drum size does ASTM D1781 specify?
The drum has an outside radius of 51.00 ± 0.13 mm (2.000 ± 0.005 in), a 102 mm (4 in) diameter. The flanges are 12.70 ± 0.13 mm (0.500 ± 0.005 in) larger in radius, measured to the strap centerline, about 127 mm (5 in) in diameter. The drum, flanges, clamp and counterweights together must weigh no more than 3.6 kg (8 lb).
What is the difference between ASTM D1781 and EN 2243-3?
Both are climbing drum peel tests for sandwich structures. EN 2243-3 is the European aerospace method and uses a different geometry, typically a 100 mm drum with 125 mm flanges, along with its own specimen and test requirements. D1781 uses a 102 mm drum with a 12.7 mm flange offset. Fixtures and results are not interchangeable between the two standards.
How does ASTM D1781 compare with D3167 and D1876?
D1781 peels a facing around a climbing drum and reports torque per width; it suits sandwich panels and flexible-to-rigid bonds. ASTM D3167 peels a thin metal adherend from a rigid one around a floating roller and reports force per width. ASTM D1876 is a T-peel of two flexible adherends. Choose the method that matches your adherends and specification.
What does a high peel value mean for a brittle adhesive?
The standard notes that with brittle adhesives, unusually high peel values may indicate incomplete cure, so peel resistance can serve as a degree-of-cure check. A higher number is not automatically a better bond. Read the result together with the failure mode and the cure record before accepting a panel.