Question 2 of 6: Identifying rolled threads, and process selection for a finned profile

Question 2: Identifying rolled threads, and process selection for a finned profile

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Approach. Parts (a) and (b) are the same question asked twice with different evidence: (a) shows the internal grain flow, (b) shows the starting-stock diameter. Rolling is a deformation process that displaces metal, cutting is a subtractive process that removes it, and each leaves an unmistakable signature.

(a) Figure Ro1 — the grain-flow evidence

Thread A is the rolled thread. In sketch A the flow lines are continuous and curved: they sweep up into each crest and bend around each root without ever being cut off at the flank surface, and they crowd together at the roots where the material has been cold-worked most heavily. That is exactly the fibre pattern produced when a die presses the profile into the surface and the metal flows plastically from the root region up into the crest. In sketch B the flow lines are straight and horizontal — the original drawn-bar fibre — and they are truncated where they meet the thread flanks, because a cutting tool has removed the material that used to continue along them.

The distinction matters far beyond identification. Because the fibres in the rolled thread follow the contour and the roots carry compressive residual stress from cold working, a rolled thread has substantially higher fatigue strength than a cut thread of the same geometry — typically 30 % or more, and considerably greater still if the thread is rolled after heat treatment. The cut thread has both a machined surface finish and the fibre discontinuity acting at the root, which is precisely where the fatigue crack starts.

(b) Figure Ro2 — the blank-diameter evidence

Thread B is the rolled thread. The proof is in the labelling. In B the starting stock is called the “diameter of blank”, and it is drawn between the finished minor and major diameters — approximately at the pitch diameter. No metal has been removed: the dies have pushed material inward to form the roots and that displaced volume has flowed outward to raise the crests, so the finished major diameter is larger than the bar it started from. In A the starting stock is called the “diameter of bar”, and it coincides with the finished major diameter: the crests are the original bar surface and everything below them has been machined away to form the roots.

Quantitatively, the blank for a rolled thread is sized on a volume balance so that the displaced root volume equals the crest volume; in practice the blank diameter comes out just under the pitch diameter, close to d_blank ≈ d_major − 0.65 p for a standard 60° ISO profile. Rolling therefore also saves material and eliminates chips, and it is far faster: a thread-rolling head or flat-die machine produces threads in a fraction of a second per part, which is why virtually all mass-produced bolts and screws are rolled.

(c) Process for the finned cross-section of Figure Pr1

Recommendation: hot extrusion of aluminium (direct extrusion through a shaped die), followed by cutting the extruded length to size.

The part shown is a heat sink: a great many thin, closely spaced radial fins on a common base, and — the decisive observation — that cross-section is constant along the length of the part. A prismatic part of constant complex cross-section produced in high volume is the textbook case for extrusion, because the entire profile, however many fins it carries, is generated in a single pass through one die at rates of metres per minute. The die is the only shape-specific tooling, so once it is paid for the marginal cost per metre is small, and the process scales to any length.

Aluminium is the natural material and the natural choice for extrusion at the same time: it has high thermal conductivity for a heat sink, and its low flow stress at about 450–500 °C makes even a thin-fin section extrudable at practical press pressures. Machining the fins from solid would remove most of the material as chips and take minutes per part; casting cannot fill fins that thin reliably; stamping and bonding individual fins is an assembly operation with a thermal joint at every fin root. Extrusion has none of these penalties.

Two design cautions go with the recommendation. The fins must not be so thin or so tall that the die tongues between them deflect or break, so the fin aspect ratio and the tip radii should be discussed with the extruder, and the die-land lengths must be relieved so metal flows out of the thin fins at the same velocity as out of the thick base — otherwise the section leaves the die twisted or bowed. That is the same flow-balancing issue raised explicitly in Question 3(b).