How much thread is enough
Every bolted joint is designed around one rule: the bolt must break before the threads strip. A broken bolt is visible, ductile and gives warning. A stripped thread is silent — the nut still looks tight and the joint has already failed. How deep the bolt has to go to guarantee the right failure depends on what it is screwed into. Move the slider and watch which one gives way.
The rules everyone uses, and where they come from
There is no ISO or ASME standard that publishes a minimum engagement in diameters. What the standards do publish is the principle — ASME B1.13M Appendix B: engagement is chosen so the bolt reaches its full tensile strength before the internal threads shear — and the method for calculating it, in FED-STD-H28/2B and the Machinery's Handbook. The diameter rules below are the industry's long-standing shorthand for the result of that calculation, and they are what most drawings, procedures and inspectors actually use.
| Bolt screwed into | Minimum engagement | Why |
|---|---|---|
| Nut, or steel of equal strength to the bolt | 1.0 × d | Matched strength. This is what a standard nut gives you — ISO 4032 and A194 2H heights are set at about 0.8–1.0 d for this reason. |
| Mild steel, with a high-strength bolt (10.9, 12.9) | 1.5 × d | The bolt is stronger than the hole. More thread has to share the load. |
| Cast iron | 1.5 × d | Weaker in shear than steel and brittle. |
| Aluminium | 2.0 × d | Roughly half the shear strength of steel. |
| Plastic | 2.5 × d or more | Low shear strength and creeps under sustained load; often better with a threaded insert. |
Some references quote 1.5 d as the general rule for static loads and 2 d for dynamic loads regardless of material. Both are conservative readings of the same principle. And a floor that applies everywhere: never fewer than three full threads engaged.
The calculation behind the rule
For a tapped hole in a material weaker than the bolt, the engineer works out the length at which the internal thread's shear strength equals the bolt's tensile strength. The FED-STD-H28/2B form is:
That is what sits behind the diameter rules. It is also why a rule of thumb is not enough for a critical joint in an unusual material: the calculation uses the actual thread dimensions and the actual strengths, and the answer for a 12.9 bolt in a low-grade casting can be a good deal more than 1.5 d.
Questions we are asked
Why is a standard nut only about 0.8 to 0.9 d high if the rule is 1.0 d?
Because the nut is made of steel matched to the bolt class and its thread is fully formed, and ISO 898-2 tests the nut to a proof load at which the bolt is meant to fail first. The 1.0 d rule is for a tapped hole in structural material, where the thread is cut on site, the material is whatever the drawing says, and nobody proof-tests it. Use the nut the standard pairs with the bolt and the engagement question is already answered.
Does more engagement keep helping?
Only up to a point. Load is not shared evenly along a thread — the first few engaged threads carry most of it — so once the engagement passes roughly 1.5 to 2 d the extra threads are doing little. Beyond that, more depth buys you little strength; what it does buy is tolerance for damaged or incomplete threads at the top of the hole.
Can I use a longer bolt in an aluminium housing instead of an insert?
Sometimes, if there is 2 d of sound material to tap. But aluminium threads also gall, wear on repeated assembly, and creep under load, so a steel threaded insert is often the better answer where the joint will be undone more than a few times or carries a sustained load. It also lets you keep a standard bolt length.
What if the engagement on the drawing is less than the rule?
Then someone has either calculated it properly for the specific materials — in which case there will be a number to point to — or nobody has thought about it. Ask. The calculation exists precisely so that the rule of thumb can be departed from with justification, and a joint below the rule with no justification is a question for the engineer, not a reason to proceed.
Related
Head markings chart · Torque is not preload · Why bolts fail · Back to the Bolt Lab