Prismatic Pair Mechanism: How It Works, Diagram, Parts, Friction Formula and Uses Explained

← Back to Engineering Library

A Prismatic Pair is a lower kinematic pair that allows pure translation of one rigid body along a single straight axis relative to another, with all rotation locked out. It solves the problem of constraining a moving element to one degree of freedom along a defined line — no wobble, no yaw, no lift. The mating prism-and-slot surfaces share continuous area contact, which is why Reuleaux classified it as a lower pair. You see it in every machine-tool slideway, hydraulic cylinder rod and Linear Actuator extension tube.

Prismatic Pair Interactive Calculator

Vary normal load, gib preload, friction coefficient, and slide speed to see sliding friction and power loss in a prismatic pair.

Contact Normal
--
Friction Force
--
Friction Force
--
Power Loss
--

Equation Used

F_friction = mu * (W + F_N,gib); P_loss = F_friction * v

The prismatic pair friction force is the friction coefficient multiplied by the total normal force on the sliding surfaces. The total normal force is the payload normal load plus the gib preload. Power loss is friction force multiplied by slide speed.

  • Normal load and gib preload act perpendicular to the sliding surfaces.
  • The friction coefficient represents the selected lubrication and speed regime.
  • Steady sliding is assumed, with friction opposing the translation axis.
Watch the Prismatic Pair in motion
Video: Eliminating backlash in prismatic joints by Nguyen Duc Thang (thang010146) on YouTube. Used here to complement the diagram below.
Prismatic Pair Cross-Section A cross-sectional view of a dovetail prismatic pair showing how the non-circular profile prevents rotation while allowing linear translation. Way (fixed) Slide (moves) Gib Translation Axis Non-circular profile Rotation blocked
Prismatic Pair Cross-Section.

How the Prismatic Pair Actually Works

A Prismatic Pair — the P-joint in robotics shorthand — gives you exactly one degree of freedom: translation along a single straight axis. Two mating bodies share a non-circular cross-section (a dovetail, a square prism, a Vee, a flat-and-gib) and slide along that shared profile. Because the cross-section is not a circle, rotation about the slide axis is geometrically impossible. That is the entire trick. Reuleaux placed it in the lower pair group because the contact is surface-on-surface, not point or line, which spreads load and gives long wear life when the surfaces are properly fitted.

The geometry has to be right or the joint binds or rattles. On a typical machine-tool slideway you are aiming for 0.01-0.02 mm of clearance per 100 mm of travel — tighter than that and thermal growth seizes the slide on a hot afternoon, looser and you lose positional accuracy under cutting load. Surface finish matters too. A scraped cast-iron way at Ra 0.4 µm holds an oil film better than a ground surface at Ra 0.8 µm, which is why high-end Moore Tool jig borers still use hand-scraped ways. If your dovetail gib is over-tightened you will measure a dead-band on reversal — the slide simply will not move until the screw torque overcomes static friction. Loose gib and you get stick-slip chatter at low feed rates, visible as periodic surface marks on a finished cut.

Failure modes are predictable. Galling on the wear surface from oil starvation, edge-loading from a bent rod, and brinelling at the ends of stroke from repeated impact at hard stops. On a Linear Actuator, the most common field failure is side-load fatigue at the rod-end bushing — the prismatic constraint holds, but the bushing wears asymmetrically and the rod develops measurable yaw under load.

Key Components

  • Slide (moving member): The translating body that carries the payload along the prism axis. Typically hardened steel at 58-62 HRC for machine tools, or anodised aluminium for light-duty linear guides. Straightness must be held to roughly 0.005 mm per 300 mm of length to avoid binding.
  • Way or guide (fixed member): The stationary mating profile that defines the line of travel. On a Bridgeport-style knee mill this is the cast-iron dovetail way; on a Linear Actuator it is the inner extrusion tube. The way must be straight, parallel and flat to the same tolerance class as the slide.
  • Gib: A tapered or parallel wedge that takes up clearance between slide and way. You adjust it with set-screws to hit a target drag torque — typically 0.3-0.6 N·m of resistance on a manual mill cross-slide. Too tight and the lead-screw stalls, too loose and you lose accuracy.
  • Wipers and seals: Felt or polyurethane wipers keep grit and chips off the wear surface. On a CNC machine running flood coolant, missing wipers will destroy a hardened way in under 200 hours. They are cheap insurance.
  • Lubrication film: An oil or grease film of 2-10 µm thickness separates the surfaces and converts boundary friction to mixed-film friction. Way oil (ISO VG 68 with tackifier) is the standard choice because it resists squeeze-out under load.

Real-World Applications of the Prismatic Pair

The Prismatic Pair is everywhere you need straight-line motion under load. Any time a designer needs to constrain a body to one axis — and only one axis — they reach for some form of P-joint. The implementation changes (dovetail, ball rail, hydraulic cylinder, telescoping tube) but the kinematic classification is identical.

  • Machine tools: The cross-slide and saddle on a Hardinge HLV-H toolroom lathe — hand-scraped cast-iron prismatic ways carrying the carriage along the bedways.
  • Linear motion: The extending inner tube of a FIRGELLI Linear Actuator, where the rod-and-housing pair forms a non-rotating prismatic constraint driven by an internal lead-screw.
  • Hydraulics: The piston-and-cylinder bore in a Parker H-Series hydraulic cylinder — prismatic translation under load, with rotation prevented by a keyed rod or external guidance.
  • CNC and automation: THK SHS-series profile rail blocks running on hardened linear rails, used on Haas VF-series machining centre axes for X, Y and Z translation.
  • Furniture and storage: Drawer Slides — Accuride 3832 ball-bearing slides are a pure prismatic pair, constraining a drawer to single-axis extension under up to 100 lb load.
  • Aerospace: The landing-gear oleo strut on a Boeing 737 main gear — a prismatic pair between the inner and outer cylinders, with torque links preventing rotation about the slide axis.

The Formula Behind the Prismatic Pair

The practical question on a prismatic pair is rarely 'does it slide' — it is 'how much force do I waste fighting friction at the operating speed and load I actually run at?' The friction force varies sharply across the operating range. At the low end of typical slide speeds (under 5 mm/s) you are in the boundary-friction regime where coefficient of friction can hit 0.15-0.20 and stick-slip dominates. In the middle (50-200 mm/s on a CNC rapid) you fall into mixed-film at roughly μ = 0.05-0.08, the design sweet spot. Push past 500 mm/s and viscous drag from the oil film starts adding back to the total. Sizing the actuator means knowing where on that curve you actually live.

Ffriction = μ × (W + FN,gib)

Variables

Symbol Meaning Unit (SI) Unit (Imperial)
Ffriction Total friction force opposing motion along the prism axis N lbf
μ Coefficient of friction between slide and way (regime-dependent) dimensionless dimensionless
W Normal load from payload weight on the slide surface N lbf
FN,gib Normal preload from gib adjustment or rail clamping N lbf

Worked Example: Prismatic Pair in a vertical-mill Z-axis quill counterweight slide

You are sizing the drive force needed for a manually-fed Z-axis dovetail slide on a small benchtop mill — a Sherline 5400-class machine — where the slide carries a 12 kg head assembly and rides on a hand-scraped cast-iron dovetail with an adjustable gib. You need to know the friction force the user's hand-wheel must overcome at the typical hand-feed range, so you can spec a lead-screw and avoid the customer complaining the wheel is 'sticky' on slow plunge cuts.

Given

  • W = 12 × 9.81 = 117.7 N
  • FN,gib = 40 N
  • μnominal = 0.10 dimensionless
  • Total normal load = 157.7 N

Solution

Step 1 — at nominal hand-feed speed of around 20 mm/s, the slide sits in mixed-film friction with a coefficient near 0.10 on oiled cast iron:

Fnom = 0.10 × (117.7 + 40) = 15.8 N

That is the force the lead-screw must transmit at nominal feed. Roughly 1.6 kgf at the slide — easy on a 10 mm pitch hand-wheel, the operator barely notices it.

Step 2 — at the low end of the operating range, slow plunge cuts at 1-2 mm/s, the slide drops into the boundary-friction regime where μ rises to about 0.18:

Flow = 0.18 × 157.7 = 28.4 N

Almost double the nominal. This is exactly why slow plunge feels notchy on a tight gib — you are climbing the static-friction hump every time the slide stops moving for even a fraction of a second. Stick-slip chatter shows up here as periodic ridges on the finished surface.

Step 3 — at the high end of useful manual feed, around 100 mm/s during rapid retraction, the oil film fully separates the surfaces and μ drops to roughly 0.06:

Fhigh = 0.06 × 157.7 = 9.5 N

The slide feels glassy at this speed — about 1 kgf at the screw. This is the regime CNC rapid traverses are designed to live in, and it is why a powered Z-axis on the same slide feels dramatically smoother than the hand-wheel at 1 mm/s.

Result

Nominal friction force is 15. 8 N at typical hand-feed. In practice that is a hand-wheel torque the operator describes as 'firm but smooth' — noticeable but not tiring. The full operating range tells the real story: 28.4 N at slow plunge versus 9.5 N at rapid traverse, a 3× swing across the speed band, with the sweet spot sitting between 10 and 50 mm/s. If your measured force is 50% above predicted, the most likely causes are (1) an over-tightened gib screw — back it off until you can just feel drag at the wheel, (2) contaminated way oil with metal fines from a missing wiper, which converts mixed-film back to boundary friction, or (3) a bent or twisted gib that edge-loads the dovetail at one end of travel and binds harder near the stops.

Choosing the Prismatic Pair: Pros and Cons

A Prismatic Pair is one of three ways to constrain a body to a single axis of motion. The other two — round-rod-and-bushing (cylindrical pair with anti-rotation feature) and recirculating-ball profile rail — solve the same kinematic problem with different cost, accuracy and load profiles. Pick based on what you actually need.

Property Prismatic Pair (dovetail slide) Round shaft with bushing + key Profile rail with ball blocks
Positional accuracy over 300 mm 0.01-0.02 mm (scraped) 0.05-0.10 mm 0.005-0.010 mm
Typical max speed 500 mm/s 1,000 mm/s 5,000 mm/s
Load capacity (radial, 30 mm size) High — full surface contact Low-moderate High but point-contact fatigue
Stiffness under cutting load Highest of the three Lowest — bushing flex High but lower than scraped way
Cost per linear metre (2024) $$ (cast and scraped) $ (commodity shafting) $$$ (THK/Hiwin profile rail)
Service life under flood coolant 20,000+ hr with wipers 5,000-10,000 hr 10,000-15,000 hr
Stick-slip below 5 mm/s Present without way oil Significant Negligible — rolling contact
Best application fit Machine tool slideways, heavy cuts Light automation, pneumatic actuators CNC servo axes, pick-and-place

Frequently Asked Questions About Prismatic Pair

You are seeing non-parallel ways or a tapered gib seated wrong. The prismatic constraint assumes the two surfaces are straight and parallel along the full stroke — if the way has 0.03 mm of bow (common on poorly-stress-relieved castings) the gib clearance closes at the bowed end and the slide binds.

Check it with a dial indicator on a granite surface plate: traverse the slide and watch for variation greater than 0.01 mm. If the way itself is straight, suspect the gib — a tapered gib pushed too far in one direction will load the dovetail flanks unevenly. Back the adjustment screw out, tap the gib back, and re-set drag torque to a uniform value across the stroke.

Choose the prismatic pair (scraped or hardened-and-ground way) when you need maximum stiffness under interrupted cutting loads and you can accept lower top speed. A scraped way has area contact, so a 5 kN side-load spreads across hundreds of square millimetres — the deflection is tiny.

A ball block of the same size carries the same 5 kN through a few dozen point contacts and deflects more, plus the balls fatigue under repeated heavy cuts. Profile rail wins on speed, low friction and ease of replacement. If you are roughing steel with a 50 mm face mill, stay with the prismatic way. If you are cutting aluminium at 10,000 mm/min, go ball rail.

The kinematic constraint is perfect on paper but real hardware has clearance. As the rod extends, the unsupported moment arm grows and the small clearance at the rod-end bushing translates into measurable angular deflection at the tip. A 0.05 mm radial clearance at a 25 mm bushing becomes about 0.2° of yaw, which at 300 mm extension is 1 mm of tip wander.

The fix is either a longer engagement length between the inner and outer tube (more bushing area), tighter bushing clearance, or external guidance such as a parallel rail. This is why heavy-duty industrial actuators have stroke-to-engagement ratios capped at around 4:1.

Run a quick A/B test. First, back the gib adjustment screws off completely and re-measure drag — if the friction drops by more than half, your preload was the problem. If it barely changes, the gib is not the dominant contributor.

Next, flush the way with fresh ISO VG 68 way oil and traverse the slide 20-30 times to flush debris, then re-measure. A 20-30% drop after flushing points to contamination — usually metal fines from a worn wiper that have embedded in the cast-iron surface and converted your mixed-film regime back to boundary contact. If neither test moves the needle, the way surface itself is galled and needs re-scraping.

The prismatic pair constrains motion in 5 of the 6 degrees of freedom but it does nothing to limit travel along the free axis — that is the whole point. Axial load is carried by whatever drives the slide (lead-screw, hydraulic piston, rack-and-pinion) plus whatever resists overtravel at the ends.

Without a hard stop or limit switch, the slide will happily translate itself off the end of the way under any axial force. On a hydraulic cylinder the cylinder cap is the stop. On a machine slide it is a bolted-on travel limit. Skip this and you will eventually fish a 12 kg saddle off the floor.

Stick-slip is a function of the difference between static and kinetic friction coefficients combined with system compliance. Even with fresh oil, if the lead-screw and its support are flexible, the screw winds up under static friction torque, then releases when the slide breaks free, lurches forward, and re-sticks.

You fix it by either reducing the static-kinetic gap (way oil with the right tackifier additive — Mobil Vactra No. 2 is the textbook choice), increasing system stiffness (preloaded ballscrew, stiffer screw mount), or using a Turcite-B or Rulon-laminated way that has near-zero static-kinetic difference by design. Hand-scraped iron with Turcite is the classic Moore Tool solution and it eliminates stick-slip down to 0.5 mm/s.

References & Further Reading

  • Wikipedia contributors. Prismatic joint. Wikipedia

Building or designing a mechanism like this?

Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.

← Back to Mechanisms Index
Share This Article
Tags: