Guide
Rolled or Cut Threads on a Custom Turned Screw: Does It Matter at 10,000 a Year?
The choice of Rolled or cut threads is the choice between squeezing the thread profile into a bar with hardened dies and cutting it out with a tool. Rolling leaves a cold-worked, compressed thread root that resists fatigue, but it needs a ductile material and a clear run of thread, so at 10,000 a year the deciding factor is the fatigue requirement of the screw, not the quantity.
By Saola Manufacturing engineering team · Published · Updated
This guide is for engineers designing special screws, studs and threaded shafts that see alternating loads, and for buyers who have to decide whether the drawing should require rolled threads. It sets out what the published fatigue studies show, what rolling cannot do, how heat treatment order complicates the picture, and how to word the drawing so the choice is deliberate.
A note on evidence. The fatigue data here comes from two laboratory studies on one thread size and one steel, so it shows direction and not a design allowance. No fastener standard is quoted for the fatigue gain. Where the sources give different magnitudes, this page gives the direction only.
On this page
What is the difference between rolled and cut threads?
A cut thread is made by removing metal with a tool, and a rolled thread is made by displacing metal with hardened dies, without chips. The rolled thread keeps a continuous grain flow along its profile and a compressed surface layer.
Wikipedia, summarizing standard manufacturing texts, describes rolling as a cold-forming process that produces no swarf, needs less blank material, and raises yield strength, hardness, wear resistance and fatigue strength because the grain conforms to the thread profile (Wikipedia, Thread rolling, accessed 2026). The Ecoroll technical article reaches the same picture from test data and adds that a cut thread, by contrast, is already in tension close to its surface.
| Item | Cut thread | Rolled thread |
|---|---|---|
| How the profile is made | Metal removed by a tool; chips | Metal displaced by dies; no chips |
| Grain at the root | Cut across the profile | Continuous, deformed along the profile |
| Residual stress near the surface | Tensile after about 50 µm depth | Compressive, about -650 MPa, to about 1 mm depth |
| Fatigue in the Ecoroll test | Baseline | Higher by 84% (Wohler curve knee) |
| Surface finish, typical | Depends on tool and feed | Generally smoother, with burnished flanks |
| Material | Brittle and ductile alike | Ductile materials only |
| Thread up to a shoulder | Possible | Cannot be formed fully up to a shoulder; incomplete threads remain at the runout |
| Tooling and quantity | Flexible, inexpensive for small lots and special forms | Productive in series; single-part tooling is expensive |
Does rolling really improve fatigue life?
In both laboratory studies found, rolled threads on a quenched and tempered alloy steel clearly outlasted cut threads, with the gain tied to compressive residual stress. The studies cover one thread size and one steel, so treat them as direction, not a guarantee.
The Ecoroll article tested M12 rods in 42CrMo4 and found that rolling raised the fatigue strength well above cutting, with compressive residual stresses present to about a millimeter below the surface where the cut thread was already in tension close to the surface (Ecoroll, accessed 2026). The article also tested deep rolling, which gave a smaller gain than thread rolling, and positions it as a more flexible option for single parts and small series.
A 2025 conference paper tested M12 bolts in 42CrMo4+QT under cyclic tension, comparing cut, rolled and deep-rolled threads, and found rolling gave the largest fatigue gain (du Maire et al., ICSP-15, Purdue e-Pubs, 2025). The two studies report different numerical gains, which is normal for different test setups, so this page quotes only the direction.
What the studies do not show is the effect on other sizes, other steels or other load types. Fatigue life also depends on surface condition, notch geometry and preload, so a rolled thread on a poorly designed joint will still fail early. Treat rolling as one of several levers, along with thread runout design and bearing area, and not as a substitute for them.
When is cutting the only option?
Cutting is the only practical choice for brittle materials, threads that run up to a shoulder, tapered threads and special profiles, and for very small quantities where rolling tools do not pay back.
The Wikipedia summary lists the conditions where cutting is preferred: full thread depth is needed, quantities are small, blanks are imprecise, the thread must run up to a shoulder, the thread is tapered, or the material is brittle (Wikipedia, Thread rolling, accessed 2026). Ecoroll describes cutting as very flexible and comparatively inexpensive, especially for small quantities or special thread forms.
The practical test is on the drawing. If the thread runs right up to a head or shoulder, ask whether a relief groove can be added; the thread relief groove guide covers the geometry. If the part uses a hardened or low-ductility material, rolling may be ruled out before fatigue is even discussed, and the cut thread then needs a more generous root design instead.
Does 10,000 a year justify rolled threads?
Volume rarely rules rolling out at that level; the fatigue requirement, the material and the thread geometry decide. If the thread is not fatigue-critical, cutting is simpler and keeps the drawing flexible.
Ecoroll calls rolling very productive and cost-effective in series production and says rolling tools are very expensive for single parts. Ten thousand pieces a year is closer to the series side of that comparison than to single parts, so the cost argument against rolling is weaker than it is for prototypes, though the tooling is still specific to each pitch and thread form. A family of screws in many thread sizes multiplies the tooling, and that is where cutting can win on cost.
A workable decision order is: first ask whether the thread carries alternating load or is a critical joint; then check that the material is ductile and the thread geometry allows rolling; only then compare the cost of the two routes with quotes. If the answer to the first question is no, rolling brings little and a cut thread on a CNC lathe is the lower-risk default.
How does heat treatment order change the picture?
The studies do not say how much of the rolled thread's benefit survives if the part is hardened after rolling, so the order of operations must be a deliberate drawing decision.
The Purdue study designates its steel as 42CrMo4+QT, a quenched and tempered condition, but its abstract does not describe the sequence of threading and heat treatment. Do not assume that a published fatigue gain applies to your sequence. Ask the thread supplier which order its data came from, and write the order on the drawing.
Two other points follow from the order. Heat treatment after threading can move the pitch diameter, so the thread class and allowances must allow for distortion (see heat-treat distortion allowance). And if the screw is later plated, hydrogen embrittlement baking becomes part of the process, as described in the hydrogen embrittlement guide.
How should the drawing and RFQ call out and accept rolled threads?
State the thread, the class, the method required (rolled or supplier's choice), the sequence relative to heat treatment, and how the thread is accepted. Require rolling only when the function depends on it.
- The thread callout and class, written as in the thread callout guide.
- "Thread rolled" if the fatigue behavior depends on it, or "method at supplier's option" if it does not.
- The sequence: thread before or after heat treatment, and before or after any plating.
- Acceptance by thread gauges as described in the go/no-go gauge guide. Gauges check fit, not fatigue, so if fatigue life is a requirement, agree a test or process evidence with the supplier.
- The blank diameter before rolling is a process variable, because metal is displaced rather than removed and the blank size carries through to the major diameter, so let the thread maker set the blank and control it by lot.
Cut threads on turned screws are CNC lathe work, described on the CNC turning page. Thread rolling is not part of the CNC turning page, so tell the quote desk the drawing requires rolled threads: the quote then states whether rolled threads can be supplied for that part and, if so, where they are made and whether that site is in Vietnam. Send the drawing through the quote form.
Frequently asked
Are rolled threads stronger than cut threads?
For fatigue, the studies found them clearly better on one steel and one thread size. The case for rolling rests on alternating load, which is what the test data cover.
Can any screw material be thread rolled?
No. Rolling is limited to ductile materials because the thread is cold formed, and brittle materials such as cast iron are excluded.
What is deep rolling and is it a substitute?
Deep rolling works a cut thread with a roller to add compressive stress. Both studies found it gave a smaller gain than full thread rolling, and Ecoroll positions it for single parts and small series.
Can I thread roll right up to the head of a screw?
Not with a self-opening die head, which cannot form the last threads against a shoulder. Add a relief groove or accept a cut thread near the head.
Should the drawing always say rolled?
Only when the function depends on it. If the thread is not fatigue-critical, leaving the method to the supplier keeps cost and options open.
Sources
- Thread rolling or thread burnishing, Ecoroll technical blog, accessed October 2026 (M12 rods in 42CrMo4, tested on the DIN 969 basis)
- Investigations on the fatigue strength of threads produced by different fabrication techniques, du Maire, Hoffmeister, Oechsner, Johlitz, 15th International Conference on Shot Peening, Purdue e-Pubs, 2025
- Thread rolling, Wikipedia, citing Machinery's Handbook (25th ed., 1996) and other manufacturing texts, accessed October 2026