The linear module is the most common actuator in automation, but how people pick its transmission is often "use what everyone else uses" or "buy ball screw if the budget allows". The real trouble shows up only after the machine is running: the cycle time is short, the accuracy drifts, or a long stroke starts to resonate. By then, changing the transmission means changing the whole mechanism.

This guide puts the three mainstream linear transmissions — timing belt, ball screw, and rack-and-pinion — side by side, explains the principle behind each, their strengths and limits, and ends with a decision table you can use directly. This is not another "belt vs screw" two-way comparison; it is all three inside one decision framework, so you get it right the first time.

1. The essence first: the three differ in how force is carried

Choosing a transmission is really choosing how the motor's rotation becomes straight-line motion, and how the force is carried. Hold on to that and every later parameter falls into place.

  • Timing belt: the motor drives a pulley, the toothed belt meshes with the pulley, and the carriage is fixed to the belt. Force is carried through belt tension — a flexible transmission.
  • Ball screw: the motor turns the screw, the ball nut travels along the screw axis, and the carriage is fixed to the nut. Force is carried through balls rolling between the screw and nut — a rigid, rolling transmission.
  • Rack-and-pinion: the motor turns a gear that meshes with a rack; the gear's rotation moves the rack (or the gear walks along the rack). Force is carried through tooth engagement — a rigid, meshing transmission.

In one line: belt carries by tension (flexible), screw by rolling (rigid and precise), rack by meshing (rigid and long).

2. Timing belt: fast, long, cheap — the price is stiffness and tension

The strengths of a belt module are clear:

  • High speed: up to 1500–2000 mm/s with fast acceleration, ideal for high-frequency reciprocation;
  • Long stroke: a single unit reaches the 4-metre class and can be joined, one of the easiest choices for long stroke;
  • Low cost, simple maintenance: simple structure, generic spares; the belt is a consumable but cheap to replace;
  • Controllable noise: a rounded tooth profile (e.g. HTD5M) plus silent design keeps high-speed noise down — the Zhongbai ZW series silent belt modules follow exactly this path.

The limits are just as clear:

  • Limited stiffness and load: force travels through belt tension, so a heavy load or a long overhang produces measurable deflection at the carriage and in the belt;
  • Tension decays: over time the belt slackens and must be re-tensioned, or accuracy drifts;
  • Belt breakage is an extreme but real risk under overload or an aged belt.

Best fit: light-to-medium load, high-frequency reciprocation, cycle-time priority, medium-to-long stroke — pick-and-place, sorting, labelling, tray handling, inspection transfer. See the ZW series synchronous belt silent modules.

3. Ball screw: precise, rigid, self-locking — the price is speed and stroke

The ball screw module is the default for accuracy-first duty:

  • High accuracy and accuracy retention: balls roll between screw and nut, giving better transmission accuracy and stiffness than a belt, with less drift over time;
  • Good rigidity: the screw is a rigid shaft, stronger against axial force and vibration than a belt;
  • Self-locking: with a small lead angle the screw self-locks, and with a brake it is the natural choice for a vertical axis against dropping;
  • High thrust: more axial thrust for the same size.

The limits are speed and stroke:

  • Speed is capped: the screw's length-to-diameter ratio sets a critical speed; spin too fast and it resonates and judders, so high speed is not its strength;
  • Stroke is capped: a long screw sags under its own weight, typically within 2–3 metres before it needs support or a different structure;
  • Heating: sustained high-speed running heats the screw, and temperature rise matters;
  • Higher cost: usually more expensive than a belt at the same size.

Best fit: medium-to-low speed, high accuracy, medium-to-heavy load, vertical anti-drop, high positioning demand — fastening, pressing, dispensing, positioning fixtures, Z-axis lifting. See the fully enclosed ball screw modules.

4. Rack-and-pinion: long, rigid, heavy — the price is backlash and accuracy

Rack-and-pinion is the answer to "long stroke plus heavy load":

  • Stroke is practically unlimited: the rack is joined segment by segment;
  • High stiffness, no elasticity: tooth engagement is rigid, unlike a belt, so positioning is crisp;
  • High load capacity: carries more than a belt and most screws;
  • High speed: with a gearbox it reaches a high linear speed.

The limits:

  • Backlash: meshing always has clearance, so a backlash-eliminating mechanism (dual gear, etc.) is needed for positioning;
  • Repeatability usually slightly below a screw: depends on rack grade and the anti-backlash design;
  • Noise: metal-on-metal meshing is loud at speed, and needs lubrication and guarding;
  • Higher cost: rack, anti-backlash mechanism and lubrication add up.

Best fit: extra-long stroke, heavy load, high rigidity, gantry X/Y axes — large gantry frames, transfer robots, heavy load transfer tables. See the rack-and-pinion module range.

5. The decision table: tick the boxes and read off the answer

Split the decision into five dimensions, tick each, and the intersection is your answer:

DimensionTiming beltBall screwRack-and-pinion
StrokeMedium-long (4 m class, joinable)Short-medium (2–3 m)Extra-long (segmented, near-unlimited)
LoadLight to mediumMedium to heavyHeavy
SpeedHigh (1500–2000 mm/s)Medium-low (critical-speed limited)High (with gearbox)
AccuracyMedium (±0.05 mm class)High (up to the screw grade)Medium (backlash-limited)
CostLowMedium-highHigh

Then a few one-factor rules that override everything:

  • Vertical axis, anti-drop required → ball screw (self-locking) or a braked scheme, not a belt;
  • Cycle-time priority, high-frequency reciprocation → timing belt;
  • Extra-long stroke (>3 m) and heavy → rack-and-pinion;
  • Accuracy must hold long-term without re-calibration → ball screw;
  • Budget-sensitive, light load → timing belt.

6. The three most common mistakes

Mistake 1: "accuracy is off, so buy a more expensive transmission." Most "accuracy problems" are structural — beam deflection, tooling overhang, mounting flatness. Fix those first, or a better screw drifts just the same. Do the error budget before the transmission.

Mistake 2: treating repeatability as accuracy. They are different numbers. A belt module datasheet often quotes ±0.05 mm repeatability, but that is run-to-run consistency, not absolute positioning accuracy.

Mistake 3: assuming a belt is always cheaper than a rack on a long stroke. Up front, yes. But tension decay, accumulated error and maintenance cost on a long stroke eat the difference back. Rack-and-pinion costs more initially and less over the long run.

7. What Zhongbai offers

Zhongbai Robot carries all three lines — timing belt, ball screw and rack-and-pinion — with module widths of 40 / 45 / 50 / 60 / 80 mm, effective stroke 50–3800 mm, and repeatability up to ±0.05 mm. Because the three lines sit in one system, selection never has to "settle for a single transmission"; you work back from the duty to the right type, then to the right model.

This is exactly what separates a non-standard integrator from a supplier that only sells one kind of module: the former helps you pick the right transmission, the latter can only sell you the one they have.

Not sure which one fits? Contact us with stroke, load, speed, accuracy and mounting, and an engineer will work through the framework above with you. For full specifications, download the Zhongbai product catalogue (PDF).

---

FAQ

Q1: What is the essential difference between a belt and a ball screw module?

A: How the force is carried. The belt carries force through tension (flexible) — fast, long, cheap, but limited stiffness and load. The screw carries force through rolling balls (rigid) — accurate, stiff, self-locking, but limited in speed and stroke. The choice depends on which requirement ranks first.

Q2: What is the difference between rack-and-pinion and ball screw?

A: Rack-and-pinion carries force through tooth engagement — near-unlimited stroke, heavy load, high rigidity and speed, but with backlash (needs anti-backlash) and usually slightly lower repeatability. The ball screw is more accurate and self-locking, but limited to roughly 2–3 m. Extra-long heavy stroke → rack; short high-accuracy stroke → screw.

Q3: Which parameter do you fix first when selecting a linear module?

A: The envelope (stroke and mounting space) first, then load and centre of gravity, then speed and accuracy. Doing it backwards — picking an accuracy grade first, then finding the stroke or fit is wrong — forces a redesign.

Q4: Can a belt module be used on a vertical axis?

A: Yes, but not bare. A vertical axis has an anti-drop requirement, and a belt does not self-lock — it can drop on belt breakage or power loss, so it needs a brake or counterweight. For anti-drop vertical axes, a ball screw (self-locking) is the safer choice.

Q5: Which lasts longer, a belt or a screw module?

A: It depends on maintenance. The belt is a consumable — replace it at end of life, cheaply. A screw body lasts long but needs periodic lubrication; high-speed heating and dust shorten its life. Maintained well, both last; neglected, a screw can fail earlier through poor lubrication.

Q6: What do you use for a very long stroke (5 m+)?

A: Rack-and-pinion, or a segmented timing belt. A ball screw is not suited to extra-long stroke (sag and critical-speed limits).

Q7: What accuracy can each transmission reach?

A: The Zhongbai ZW series belt modules reach ±0.05 mm repeatability; screw modules depend on the screw grade and are usually better than a belt; rack-and-pinion repeatability is typically in the ±0.05 mm class, limited by backlash. Confirm against the actual selection and test report.

Q8: How do I quickly decide which type I need?

A: Give three numbers — stroke, load, speed. Light-to-medium load + high frequency + medium-long stroke → belt; medium-low speed + high accuracy + vertical anti-drop → screw; extra-long + heavy → rack-and-pinion. Not sure? Have an engineer run the decision table.