Testing standard
ISO 204
Metallic materials — Uniaxial creep testing in tension — Method of test
Written and technically reviewed by Dak System Inc. engineeringLast reviewed
ISO 204 is the international method for uniaxial creep testing of metals in tension. A test piece is held at temperature under a constant force in an aligned load train and left to extend. It gives creep strain at nominated times, creep rates, times to specified strains and — where the run goes to fracture — rupture time with elongation and reduction of area.
At a glance
- Test type
- Creep & relaxation
- Published by
- ISO
- Edition
- ISO 204:2023
- Material
- Metals, alloys & welds
- Runs on
- Series 7200 and Series 9000
What the test does
A threaded or shouldered test piece is built into an aligned load train — loading bars, universal joints, grips — inside a multi-zone furnace, heated to a specified temperature and soaked. The force is applied without shock, held constant, and the piece left to extend. Two regimes exist: an uninterrupted test runs to its end with extension logged continuously, while an interrupted test is periodically unloaded, cooled to ambient, measured off the machine, then reheated and reloaded — a cycle repeated many times over a run approaching 100 000 h.
What it measures, and why it matters
Creep elongation against time is the raw output, giving creep strain at nominated times, creep rates, times to specified strains and, where the test runs to fracture, rupture time with elongation and reduction of area.
For a component that spends its life hot and loaded, strain accumulated over service life decides whether a rotor still clears its casing or a bolted joint still holds preload; rupture data decides whether the part survives at all. The same data ranks candidate alloys and welded joints against parent metal, and feeds remnant-life assessment on plant already in service.
Test piece and load train
- Test piece
- Threaded or shouldered
- Load train
- Loading bars, universal joints, grips — alignedThe universal joints are there to remove bending, which over thousands of hours would otherwise accumulate damage the test does not intend to measure.
- Furnace
- Multi-zoneZones are what make a long gauge uniform. A single-zone furnace has a hot middle and cold ends.
- Uninterrupted test
- Runs to its end, extension logged continuously
- Interrupted test
- Periodically unloaded, cooled, measured off the machine, reheatedA cycle repeated many times over a run approaching 100 000 h. It buys precision in the strain measurement at the cost of disturbing the specimen each time.
- Record every interruption
- As part of the historyDakAn interrupted test's curve has a thermal cycle at every measurement point, and comparing it with an uninterrupted one requires knowing that.
Load and time
- Force
- Applied without shock, then held constant
- Soak
- Until the gauge is at temperature throughout
- Duration
- Up to about 100 000 h
- Temperature
- Held to a tight tolerance for the whole runThe binding requirement, as in every creep method.
What comes off the curve
ε_t = extension at time t / original gauge length
Design codes commonly specify a permissible creep strain in a stated number of hours, and this is the figure checked against it.
ε̇ = dε / dt
The secondary-stage rate is the one normally quoted and the basis of extrapolation.
t_u = elapsed time to fracture at one stress and temperature
With elongation and reduction of area from the broken halves, which say whether the failure was ductile or brittle.
How the test runs
- 01Machine the test piece with threaded or shouldered ends and measure the gauge.
- 02Assemble the load train with universal joints and verify alignment.
- 03Install the piece in the multi-zone furnace.
- 04Attach thermocouples along the gauge length.
- 05Heat and soak until the gauge is uniform at temperature.
- 06Apply the force without shock and hold it constant.
- 07For an uninterrupted test, log extension continuously to the end.
- 08For an interrupted test, unload, cool, measure off the machine, reheat and reload on the defined cycle.
- 09Continue for the planned duration or to fracture.
- 10Measure elongation and reduction of area on the broken halves where it fractured.
- 11Report every result against its stress and temperature, and state which regime was used.
What the report has to contain
- Reference to ISO 204
- Material identification, cast, product form and heat treatment
- Test piece type and gauge length
- Applied stress
- Test temperature and the tolerance held
- WHICH REGIME — uninterrupted or interrupted
- Number and timing of interruptions where applicable
- Creep strain at nominated times and the creep rate
- Rupture time, elongation and reduction of area where it fractured
- Any excursion in temperature or load
What the machine must be capable of
Two certificates come before any capacity figure. Force verification must reach at least class 1 of ISO 7500-2 — the creep-specific part of that series, not ISO 7500-1; specifying ISO 7500-1 on a creep frame is the common procurement error. Extensometry must be class 1 or better of ISO 9513, contacting or non-contacting, recalibrated at intervals no longer than three years and re-verified before any test expected to outlast its certificate. No extensometer is needed where only elongation after creep fracture, or creep elongation over a stated duration, is wanted.
Capacity follows from stress and section, nothing else. On a 28 mm² section — a 6 mm round piece — stresses between 50 and 400 MPa land between roughly 1.4 and 11 kN; step up to 10 mm at 300 MPa and the requirement is about 24 kN. Lever and spring-loaded frames sold against this standard are commonly rated to 50 kN, the arm doing the multiplication. There is no speed to set; what the frame owes the test is a steady force and a load train that keeps inadvertent bending and torsion to a minimum — alignment and joint quality rather than stiffness.
Temperature control is stated as a permitted deviation between corrected and specified temperature, banded by level: ±3 °C up to 600 °C, widening in steps to ±6 °C at 1 100 °C, with the same limit on variation along the test piece and the surrounding air within ±3 °C. Above that, tolerances are agreed between the parties; no lower bound is stated, and humidity is not controlled.
What goes wrong in practice
Off-axis loading heads the list, and the standard's own introduction concedes that quantitative data on its influence is still being sought. A load train out of line, or a joint that binds, adds bending to the axial stress; strain read on one side then mixes creep with bending, and rupture life comes out short.
Instrument drift is the slow failure, and it can outrun the paperwork. A thermocouple that read true on installation may not a year later, and the calibration ceiling this standard puts on extensometry exists because the instruments age faster than the tests finish. Extensometer zero drift is the companion: a trace that wanders over months cannot be told apart from slow creep.
On interrupted tests the reload is where results are lost: cooling, remounting and reheating carry their own strain history, and a piece not returned to the same alignment and temperature puts a discontinuity into the curve.
ISO 204 and ASTM E139
| ISO 204 | ASTM E139 | |
|---|---|---|
| Scope | Uniaxial creep in tension | Creep, creep-rupture and stress-rupture |
| Regimes | Uninterrupted and interrupted | Continuous measurement or rupture-only |
| Load train | Universal joints specified | Aligned train, lever-arm dead weight usual |
| Durations | Up to about 100 000 h | Up to about 100 000 h |
The two are close in substance and a laboratory equipped for one is equipped for the other. What must not be mixed is the regime: an interrupted curve carries a thermal cycle at every measurement point, and comparing it with an uninterrupted one without saying so hides a real difference in the specimen's history.
Questions we are asked about this test
What is ISO 204?
It is the international standard for uniaxial creep testing of metallic materials in tension. A test piece is held at temperature under constant force in an aligned load train, and the method yields creep strain at nominated times, creep rates, times to specified strains and rupture data where the test runs to fracture.
What is the difference between an interrupted and an uninterrupted test?
Whether the specimen stays loaded. An uninterrupted test runs to its end with extension logged continuously. An interrupted test is periodically unloaded, cooled to ambient, measured off the machine and then reheated and reloaded — which gives a very precise strain measurement at the cost of putting a thermal cycle into the specimen's history at every measurement point.
Why are universal joints in the load train?
To remove bending. Over thousands of hours even a small misalignment accumulates damage on one side of the specimen that the test is not intended to measure, and unlike a short tensile test there is no opportunity to notice and repeat. The joints let the train find its own axis and stay there.
Why a multi-zone furnace?
Because a single-zone furnace has a hot middle and cold ends, and the specimen's gauge length spans that gradient. The hottest section creeps fastest, so the reported result belongs to that section rather than to the nominal temperature. Multiple zones with independent control are what make a long gauge uniform.
What is the difference between ISO 204 and ASTM E139?
Very little in substance — both cover constant-force creep of metals over the same kinds of duration, and a laboratory equipped for one is equipped for the other. ISO 204 is explicit about the interrupted regime and about universal joints in the train. What matters commercially is that the report names which document and which regime, since an interrupted curve is not directly comparable with an uninterrupted one.
Why does temperature control matter more than force control in a creep test?
Because creep rate is far more sensitive to temperature than to stress. A few degrees of drift over hundreds of hours changes the accumulated strain more than a small error in the applied load would, and unlike a load error it is cumulative and invisible in the data. That is why multi-zone furnaces, several thermocouples along the gauge and tight control bands are specified.
What is the difference between creep strain and creep rupture data?
They answer different design questions. Creep strain data — usually the time to reach a specified plastic strain — governs components that must hold dimensional tolerance, such as turbine casings and bolted flanges. Creep rupture data governs components designed to a life, where the question is when the part will actually fail. A material can be good at one and unremarkable at the other.
Running ISO 204 on the Series 7200 and Series 9000
Dak verifies against whichever standard the method names, and where a class applies our frames sit a class tighter than it asks.
| The method asks for | Dak supplies | |
|---|---|---|
| Capacity | The standard prescribes an accuracy class rather than a capacity. Sizing follows the test piece: a 6 mm diameter round bar of about 28 mm² at creep stresses of 50–400 MPa needs roughly 1.4–11 kN, and a 10 mm bar at 300 MPa about 24 kN. Lever and spring-loaded creep frames built to ISO 204 are commonly rated to 50 kN, the lever arm supplying the multiplication so dead weights stay manageable. | Load cells from 1 kg to 60 ton on the Series 7200, and 0.5 to 100 kN on the Series 9000 |
| Force accuracy | ISO 7500-2 Class 1 | ISO 7500-1 Class 0.5, verified to ASTM E4, DIN 51221 and BS 1610 |
| Strain measurement | An extensometer to ISO 9513 Class 1, gauge length Extensometer gauge length Le not less than 10 mm, and as long as practical; original gauge length Lo generally at least 5D on round test pieces | Certified to ISO 9513 Class 1 and ASTM E83 — non-contact video, clip-on and high-elongation |
| Gripping | Aligned creep load train — loading bars, universal joints and threaded or shouldered grips — inside a multi-zone furnace on a lever or spring-loaded creep frame | Our a fixture built for this method, built to the specimen |
| Environment | Constant specified temperature, with permitted deviation between corrected measured and specified temperature of ±3 °C up to 600 °C, ±4 °C to 800 °C, ±5 °C to 1 000 °C and ±6 °C to 1 100 °C, matched by the same limit on variation along the test piece; air around the machine should stay within ±3 °C. No humidity requirement. | 3009 series chambers, −150 °C to +400 °C — temperature only |
This page describes the method as practised. The governing text is the current edition from the issuing body. Tell us what you are testing and we will answer with the machine, the fixture and a quotation.