ASTM F606 / F606M — how fastener mechanical properties are actually measured

Scope — and what it does not cover

ASTM F606 / F606M is a test method standard: it tells a laboratory how to measure the mechanical properties of externally and internally threaded fasteners, washers, direct tension indicators and rivets — proof load, wedge and axial tension, tension on specimens machined from the part, nut proof load on a hardened mandrel, hardness and where the reading is taken. It sets no property value of its own. What a grade 5 bolt or a grade A nut has to achieve is written in the product specification — SAE J429, ASTM A563, ASTM F3125, ASTM F436 — and F606 is the procedure those documents call up. Inch and metric requirements are published today under the combined F606/F606M designation, while older drawings still cite F606 for inch and F606M for metric. It does NOT cover dimensions (the ASME B18 series), coatings, corrosion testing, or hydrogen-embrittlement detection (ASTM F1940, ASTM F1624, ISO 15330).

Standards are copyrighted documents. We explain scope, intent and practical application in our own words; obtain the current edition from ISO, ASTM, DIN or your national standards body.

What actually bites in production

• 1The split between value and measurement is the whole point of this number. In the inch system the grade lives in one document and the way it is proved lives in another, which is why a correct inch certificate quotes both. The metric ISO system does the opposite: ISO 898-1 carries its own test annexes, so a metric report rarely names a separate method standard.
• 2Proof load is not a tensile test. The fastener is measured, loaded once to the proof load, held briefly, released and measured again — and it passes if it came back to the length it started at. The tolerance on those two readings exists to absorb measurement error, not to allow a little permanent stretch. Method 1, the length-measurement procedure, is the referee when two laboratories disagree.
• 3Wedge tension loads the head deliberately off-axis. A hardened wedge under the bearing face bends the head relative to the shank, so the test asks about the head-to-shank junction and about ductility, not only about thread strength. A part that survives a straight axial pull can still lose its head in the wedge, and that is the result the specification wants to see.
• 4Full-size and machined-specimen results are not the same evidence. Machining a test piece from a bolt removes the thread root and the head fillet — the two places fasteners actually break — so the specimen describes the steel, not the finished part. It is the fallback when the part is beyond the frame's capacity or its geometry will not allow a full-size pull, and the standard fixes where in the cross-section the specimen has to come from.
• 5Hardness on this axis is a location question, not a scale question. The product test calls for a transverse section through the threaded portion at a defined distance from the point, read at mid-radius, so the number describes the core of the thread rather than a wrench flat or a decarburized skin. And a fastener too short to grip in tension is tested for hardness instead — that is the standard's own provision, not a shortcut by the laboratory.
• 6Internally threaded parts and washers are tested differently again. A nut is run onto a hardened threaded mandrel and loaded axially: it must neither strip nor rupture, and after the load is released it must come off by hand. If the mandrel threads are damaged the test is void rather than failed. Hardened washers are proved by a cone proof load, so "tested to F606" means something different on a washer line than on a bolt line.

What to write on the drawing or PO

• The product specification and the method together — "SAE J429 grade 5, mechanical testing to ASTM F606/F606M". A method number on its own specifies nothing, because F606 contains no acceptance values.
• Which tests are required per lot and how many pieces: proof load, wedge tension, axial tension, hardness; for nuts, proof load and hardness; for washers, the cone proof load.
• Whether testing is on the full-size fastener or on a machined specimen, and who is allowed to make that substitution — otherwise it will be made by whoever owns the tensile frame.
• Whether the parts are tested before or after coating and baking, since a plated thread and a bare thread do not behave the same in the fixture, and state the wedge angle explicitly if anything other than the table default is intended.
• Where hardness is read and on how many pieces, which laboratory issues the report, and whether the report has to quote the F606 revision as well as the product specification.

What goes wrong when it is cited loosely

• Writing "tested to ASTM F606" as the entire requirement. With no grade, no size range and no sample size it is neither quotable nor inspectable, and two suppliers will hand back two different sets of numbers, both defensible.
• Reading a proof-load pass as a strength figure. It says the part took that load without a measurable permanent set. It says nothing about the load at which it would finally break.
• Offering machined-specimen tensile data against a full-size wedge tension requirement. The specimen cannot fail at the head fillet, because the machining removed it.
• Assuming F606 covers hydrogen embrittlement because both subjects turn up on plated high-strength parts. It does not: that is ISO 15330 or ASTM F1940 / F1624 territory, plus the baking record.