Acme Lead Screw

What an Acme lead screw actually is

A lead screw is the simplest machine for turning rotation into a straight push: spin a threaded shaft inside a captive nut and the nut walks along the thread. The Acme part names the thread form — a trapezoidal profile standardised in ASME/ANSI B1.5, with each flank angled 14.5° from a line perpendicular to the screw axis, so the included angle between opposite flanks is 29°. That stubby trapezoid sits between two extremes: the sharp 60° V-thread of an ordinary bolt (great for clamping, terrible for sliding a nut back and forth thousands of times) and the perfectly square thread (most efficient power form, but a nightmare to cut and impossible to make adjustable).

The 29° compromise is the whole point. The flank angle makes the thread root broad and strong, lets you cut it with a single-point lathe tool or a standard tap and die, and — crucially — allows a split nut to be re-tightened to take up wear, which a square thread cannot do. Acme threads have been the workhorse of the machine shop since the 1890s precisely because they are strong, manufacturable, repairable, and forgiving.

The single property that defines how an Acme lead screw behaves in a mechanism is self-locking: under normal geometry, no amount of axial force on the nut can spin the screw backwards. Push down on a scissor jack and it stays put; cut power to a valve actuator mid-stroke and the gate holds. That comes free from the thread geometry, with no brake, ratchet, or holding current. The cost of getting it free is efficiency, and the rest of this page is really about that one trade.

The geometry: lead, pitch, and lead angle

Three numbers describe the helix, and people mix them up constantly:

• Pitch (p) — the axial distance from one thread crest to the next. On a 10-TPI (threads-per-inch) screw the pitch is 0.1 inch.
• Lead (L) — how far the nut advances in one full turn of the screw. On a single-start thread, lead equals pitch. On a double-start (two intertwined helices) the lead is twice the pitch, so the nut travels twice as fast per turn at the same pitch.
• Lead angle (λ) — the angle the helix makes with a plane perpendicular to the axis. It is set entirely by lead and mean diameter:

tan λ = L / (π · d_m)        d_m = mean (pitch) diameter

The lead angle is the lever the whole mechanism pivots on. A small lead angle means a steep helix that advances slowly per turn, gives big mechanical advantage, and tends to self-lock. A large lead angle advances fast, gives little mechanical advantage, and tends to back-drive. Multi-start threads and "fast" lead screws are simply ways of buying a large lead angle without an absurdly coarse pitch.

Screw	Mean dia d_m	Lead L	Lead angle λ	Self-locking?
1/4-16 Acme, 1-start	0.219 in	0.0625 in	5.2°	Yes (λ < φ)
1/2-10 Acme, 1-start	0.450 in	0.100 in	4.0°	Yes
1-5 Acme, 1-start	0.900 in	0.200 in	4.0°	Yes
1-2 Acme, 4-start (fast)	0.900 in	0.500 in	10.0°	Marginal / back-drives

Why it self-locks — lead angle versus friction angle

Unwrap one turn of the thread and the helix becomes a ramp: a block (the load) sitting on an inclined plane whose slope equals the lead angle λ. Turning the screw is sliding the ramp under the block; the load wanting to back-drive the screw is the block trying to slide down the ramp on its own.

From basic statics, a block sits still on a ramp as long as the ramp angle is below the friction angle φ = arctan μ. The same logic governs the screw, with one correction for the trapezoidal flank: the normal force is spread over an angled face, which inflates the effective friction by 1 / cos αₙ, where αₙ is the normal flank angle (14.5° for Acme). The self-locking criterion is therefore:

self-locking  ⇔   tan λ  <  μ / cos αₙ

with μ ≈ 0.15, αₙ = 14.5°:
  μ / cos αₙ = 0.15 / cos 14.5° = 0.155   →   λ_max ≈ 8.8°

most single-start Acme screws have λ ≈ 4–5°  →  comfortably locked

So a typical single-start Acme screw, with a lead angle around 4–5° and a friction angle around 9–12° for steel-on-bronze, has a wide self-locking margin. You would have to use a 4-start "fast" thread, or run the screw bone-dry-then-suddenly-greased so μ collapses, to make it back-drive. This is exactly why a bench vise stays clamped and a globe valve stays where you cranked it — no detent, no pawl, just geometry and friction.

One subtlety worth knowing: self-locking is a static condition. Under vibration the effective friction drops and a "locked" screw can creep loose over hours — the same reason bolts back off without a lock washer. Safety-critical actuators (aircraft trim, theatre rigging) never rely on screw self-locking alone; they add a brake or a secondary nut precisely because static self-locking is not a dynamic guarantee.

Torque, thrust, and the lifting/lowering equations

Sizing a lead-screw drive comes down to the torque needed to raise a load, the torque needed to lower it, and the torque the load itself produces (which tells you whether you need a brake). For a power screw with an axial load F, mean diameter d_m, lead L, friction coefficient μ, and normal flank angle αₙ:

Raise (lift):
  T_raise = (F · d_m / 2) · (L + π·μ·d_m·sec αₙ) / (π·d_m − μ·L·sec αₙ)

Lower (release):
  T_lower = (F · d_m / 2) · (π·μ·d_m·sec αₙ − L) / (π·d_m + μ·L·sec αₙ)

If T_lower > 0  →  self-locking (you must drive it to lower the load)
If T_lower < 0  →  load back-drives the screw on its own

A concrete example — a 1-inch 5-TPI single-start Acme screw lifting a 1000 lbf load, with d_m ≈ 0.9 in, L = 0.2 in, μ = 0.15, αₙ = 14.5°:

sec 14.5° = 1.033
T_raise = (1000 · 0.9 / 2) · (0.2 + π·0.15·0.9·1.033) / (π·0.9 − 0.15·0.2·1.033)
        = 450 · (0.2 + 0.438) / (2.827 − 0.031)
        = 450 · 0.638 / 2.796  ≈  103 lbf·in

T_lower = 450 · (0.438 − 0.2) / (2.827 + 0.031)
        = 450 · 0.238 / 2.858  ≈  +37 lbf·in   →  positive  →  self-locking ✓

The positive lowering torque is the numerical fingerprint of self-locking: you have to actively spin the screw to let the load down, which is exactly what you want in a jack. The efficiency falls straight out of the raise torque — it is the ratio of the ideal frictionless torque (F·L / 2π) to the actual torque:

η = (F · L) / (2π · T_raise)
  = (1000 · 0.2) / (2π · 103)
  ≈ 0.31   →   31 % efficient

Two-thirds of the motor's work became friction heat in the nut. That is the unavoidable bill for self-locking, and it is why a continuously cycling Acme actuator runs warm and why ball screws take over in high-duty applications.

The efficiency / self-locking trade in one curve

There is a clean, almost philosophical relationship hiding in the equations. Screw efficiency for the simplified case is:

η = tan λ · (1 − μ·tan λ) / (μ + tan λ)

Plot η against lead angle for a fixed μ and you get a curve that climbs from zero, peaks near λ ≈ 45° − φ/2, and falls again. The key fact is where the 50% line sits: efficiency crosses 50% at almost exactly the lead angle where self-locking stops. In other words:

• Any screw below ~50% efficiency is (broadly) self-locking.
• Any screw above ~50% efficiency back-drives.

You cannot have both. A self-locking screw is, by definition, throwing away more than half its input to friction; that surplus friction is precisely what jams the back-drive. If you want a 90%-efficient screw you must accept that the load will run the motor backwards, and you must add a brake to hold position. This is not a manufacturing limitation you can engineer around — it is a statement of the same friction physics seen from two directions.

Acme lead screw versus ball screw

The defining alternative is the ball screw, which replaces the sliding bronze nut with recirculating hardened-steel balls running in a matched groove. Rolling instead of sliding drops the effective friction by an order of magnitude — and with it, both the inefficiency and the self-locking.

Property	Acme lead screw	Ball screw
Contact	Sliding (steel on bronze/polymer)	Rolling (recirculating steel balls)
Effective friction μ	0.15–0.20	≈ 0.01
Efficiency	30–50 %	90–95 %
Self-locking	Usually yes (no brake needed)	No — always back-drives
Backlash	Moderate, grows with nut wear	Near-zero with preload
Contamination tolerance	High — survives dirt, chips, grit	Low — balls jam on debris
Heat at high duty	Significant (friction-limited)	Low
Relative cost	Low (1×)	High (5–15×)
Typical use	Vises, jacks, valve stems, manual slides	CNC axes, EV actuators, flight controls

The decision rule is almost entirely about holding versus moving. If the job is to hold a load most of the time and move it occasionally — a clamp, a jack, a valve, a manually fed machine slide — the Acme screw's self-locking is a feature you get for free, and its low efficiency rarely matters because duty is light. If the job is to move a load fast, continuously, and precisely — a CNC table, a 3D-printer Z-axis under heavy duty, an aircraft horizontal-stabiliser screw jack — the ball screw's efficiency and life win, and you pay for a separate brake to hold position. Many designs split the difference: a high-helix-lead screw with a polymer nut for cheap, lighter-duty linear actuators that need to move quickly and accept that they are no longer self-locking.

Materials, manufacture, and the sacrificial nut

The screw is almost always steel — cold-rolled or precision-ground from medium-carbon or stainless stock, sometimes hardened on the flanks. The nut is deliberately a softer, sacrificial material:

• Bronze (often leaded or aluminium bronze). The classic pairing. Bronze has a low friction coefficient against steel, embeds grit without scoring the screw, and wears as the sacrificial element. Standard on machine tools and valve stems.
• Acetal / PTFE-filled polymer. Self-lubricating, quiet, cheap, corrosion-proof. Dominant in consumer linear actuators, printers, and medical devices. Lower load and temperature limits than bronze.
• Cast iron. Used for large, slow, heavily loaded screws (presses, large valves) where a graphite-rich matrix provides some boundary lubrication.

The rule "never steel on steel" is hard-won: a steel nut on a steel screw galls — the flanks cold-weld microscopically, tear, and seize within a few cycles unless flooded with extreme-pressure lubricant. The bronze (or polymer) nut is engineered to be the part that wears out, because re-cutting a precision steel screw is far more expensive than pressing in a new nut. Anti-backlash nuts take this further: two half-nuts are spring-loaded apart so they take up opposite flanks, and as wear opens the gap the spring closes it, holding lost motion near zero until the nut is finally spent.

Where Acme lead screws show up

• Machine-tool feed and lead screws. The lathe lead screw that drives the carriage for threading, and the feed screws on milling-machine tables and knee. Manually fed, self-locking so the table stays where you crank it.
• Vises, clamps, and presses. Bench vises, C-clamps, arbor and screw presses. The whole point is clamping force that holds with the handle released — self-locking is the entire spec.
• Scissor and screw jacks. Car scissor jacks, machine levelling jacks, theatrical and architectural screw jacks. Self-locking means the load can't drop if you let go of the handle — a safety property, not a convenience.
• Valve actuators. Rising-stem gate and globe valves use an Acme stem so the valve holds its set position with no holding torque, and a handwheel or electric actuator drives it open and closed.
• Linear actuators. Inexpensive electric linear actuators (recliners, hospital beds, satellite-dish positioners, RV slide-outs) use an Acme or polymer trapezoidal screw so the actuator holds extension with the motor off, drawing zero current to maintain position.
• Optical and instrument stages. Micrometer heads, microscope focus, camera focus helicoids — fine-pitch trapezoidal threads for precise, drift-free linear positioning.
• 3D-printer and CNC Z-axes (light duty). Trapezoidal lead screws (the metric "Tr" cousin of Acme) hold the gantry against gravity when the steppers are de-energised — though heavy machines move to ball screws for speed.

Failure modes and design pitfalls

• Nut wear and growing backlash. The expected, designed-in wear path. Sliding contact removes nut material, opening axial lost motion until positioning drifts. Cure: anti-backlash spring-preloaded nuts, scheduled nut replacement, harder nut material for the duty.
• Galling / seizing. Steel-on-steel or starved lubrication welds the flanks and locks the screw solid. Cure: always pair dissimilar materials (steel screw, bronze/polymer nut) and maintain a grease or solid-film lubricant.
• Overheating at high duty. With 50–70% of input power becoming friction heat, a continuously cycling screw can cook its polymer nut or break down its grease. Cure: derate duty cycle, switch to bronze, or move to a ball screw.
• Column buckling. A long, slender screw loaded in compression buckles below its material yield, governed by Euler's critical load P_cr = π²EI / (KL)². Cure: load the screw in tension where possible, support the free end, increase root diameter, or shorten unsupported length.
• Whip / critical speed. A long screw spun fast reaches a resonant "whip" speed where it bows out and thrashes. Cure: keep rotational speed below the critical speed (raise it with larger diameter, shorter span, or end support).
• Vibration-induced creep. Self-locking is a static property; under sustained vibration the effective friction drops and a loaded screw slowly unwinds. Cure: add a brake or locking nut on anything safety-critical — never trust self-locking alone to hold a hazardous load.