ASTM F606/F606M Mechanical Testing of Fasteners, Washers and Rivets

What the test does

Most of the work is done on the fastener as manufactured, not on a coupon cut from it. A bolt is engaged by its threads in a hardened fixture or a nut and drawn along its axis to destruction; in the wedge version a hardened block of specified angle sits under the head, loading it out of square with the shank. The proof-load test stops short of failure: the fastener is measured for overall length, loaded once to a specified load, released and measured again. Nuts are run onto a hardened mandrel or cone and loaded until the proof load is reached or the threads strip. Hardness and rotational-capacity tests are covered too, the latter on a bolt tension calibrator.

What it measures, and why it matters

Proof load is a pass or fail on elastic behaviour: a fastener that returns to its original length has not yielded at the load its joint will hold it at — the basis of preload design. Wedge tension gives breaking strength while testing the head-to-shank junction: a bolt that passes in straight tension but breaks under the head on the wedge has a fault an axial pull cannot see — a bad fillet, a quench crack, the wrong heat treatment. Nut and cone proof load confirm a nut can develop its bolt's strength, so an overloaded joint fails in the bolt rather than by stripping. Hardness screens heat treatment, and a fastener hardened beyond its grade's range is generally regarded as more susceptible to hydrogen embrittlement even when its tensile figures pass.

The fastener itself

The important idea here is that the product is the specimen. A coupon machined from a bolt would lose the thread form, the head fillet and the heat-treatment gradient — which is where most fastener failures start.

Full-size testing

The fastener as manufacturedThreads engaged in a hardened fixture or nut, and drawn along its axis.

Proof load

Load once, release, and re-measureLength measured before and after. Return to the original length means it has not yielded.

Wedge tension

A hardened block of specified angle under the headLoading the head out of square with the shank — which is what tests the head-to-shank junction.

Nuts

Run onto a hardened mandrel or coneLoaded until the proof load is reached or the threads strip.

Hardness

Screens heat treatmentA fastener hardened beyond its grade's range is generally regarded as more susceptible to hydrogen embrittlement, even where its tensile figures pass.

Machined specimens

Only where full-size testing is impossibleAnd the report must say which was used, because they answer different questions.

Loading

Every speed in F606/F606M is a free-running crosshead speed, and it differs by test rather than being set once for the standard. The figures below were read in the -21 text and are unchanged from -16 and -14.

Axial tension, full-size product

Not more than 1 in./min (25 mm/min)

Wedge tension

Not more than 1 in./min (25 mm/min), to destruction

Proof load, by length measurement

Not more than 0.12 in./min (3.0 mm/min), held 10 sApplied once and released — not cycled, because the question is only whether the fastener yielded. The 10 s hold is part of the method, not a laboratory habit.

Nut proof load

Not more than 1.0 in./min (25 mm/min), held at load 10 s minimum

Cone proof load

Maximum 0.12 in./min (3.0 mm/min), proof load applied for 10 s

Single shear

Not less than 1/4 in./min (6.0 mm/min) nor more than 1/2 in./min (12.0 mm/min)The only test in the method with a floor as well as a ceiling.

Rotational capacity

On a bolt tension calibratorTurning the nut while tension is measured, which is a different machine again.

What the machine must be capable of

Capacity follows the product, not the method: an M12 property class 8.8 bolt fails near 65 kN and a 1 1/2 in. ASTM A490 bolt approaches 1 000 kN, so fastener laboratories run frames from roughly 300 kN for small sizes to 1 500 kN for large structural product. Practice E4 governs the force verification.

Rate is set per test, each figure a free-running crosshead speed. Full-size axial tension and wedge tension run at no more than 1 in./min (25 mm/min). Proof loading by length measurement is far slower — no more than 0.12 in./min (3.0 mm/min) — and the load is held 10 s before release. Nut proof load allows 1.0 in./min (25 mm/min), held 10 s minimum; cone proof load reverts to 0.12 in./min (3.0 mm/min) for 10 s. Single shear has a floor as well as a ceiling: 1/4 to 1/2 in./min (6.0 to 12.0 mm/min).

Extensometry is less settled. Practice E83 is among the referenced documents, but which class F606/F606M demands is not publicly stated; machined-specimen yield work inherits E8/E8M's Class B-2 in practice, and proof loading by length measurement needs a length gauge, not an extensometer.

Fixturing decides whether a result means anything. Threaded holders, nut and stud fixtures and hardened wedge blocks are all needed, and the wedge angle is not a free choice: 10 degrees is usual, reduced for heat-treated bolts threaded close to the underside of the head and again for embrittlement testing, with the angle for each diameter tabulated. Use the standard angle where a reduced one is called for and the bolt fails under the head for reasons that have nothing to do with the product. Testing is in ambient air; no temperature or humidity band is set.

How the tests run

1. 01Identify the grade and the tests its specification requires.
2. 02For proof load: measure the fastener's overall length precisely.
3. 03Engage the threads in a hardened fixture or nut and load once to the specified proof load.
4. 04Release, and re-measure the length; compare with the original.
5. 05For wedge tension: fit the hardened wedge block of the specified angle under the head.
6. 06Draw the fastener to destruction and record the breaking load and where it broke.
7. 07For nuts: run onto a hardened mandrel or cone and load to proof load or thread strip.
8. 08Take hardness readings at the locations the specification names.
9. 09Report each result against the grade requirement, and state whether full-size or machined specimens were used.

What each test yields

What the report has to contain

• Reference to ASTM F606/F606M
• Fastener identification, grade, size, thread form and coating
• Whether FULL-SIZE or machined specimens were tested
• Proof load applied, and the length before and after
• Wedge angle used, breaking load, and the location of the break
• Nut proof load and whether the threads stripped
• Hardness values and where they were taken
• Rotational capacity results where required
• Number of fasteners tested and the sampling basis

What goes wrong in practice

The failure that wastes most time is the fixture giving way instead of the fastener. Holder threads wear or soften with use, engagement is too short for the grade, and the assembly strips before the bolt breaks; the load recorded is then a property of the fixture.

Galling hits coated and stainless product hardest. A seized thread ruins a rotational-capacity result outright, and in tension it can carry load in a way that flatters the fastener.

Proof loading has a trap of its own: the permanent set looked for is very small, so measurement resolution and repeatable seating points dominate. Published sources are not consistent about the resolution required, so the laboratory's own procedure has to settle it.