Animatronic Dinosaur Pneumatic Actuators: 5 Key Components

A pneumatic animatronic dinosaur is driven by five components working in a chain: a compressor makes the air, tubing and fittings carry it, solenoid valves decide where it goes, cylinders turn it into force, and a controller decides when. Take any one away and nothing moves.

Before going further, one point of honesty. Most dinosaur models built in Zigong, including ours, are defined as steel-framed, foam-bodied and electronically program-controlled: the drive element in the standard build is an electric motor through a reducer. Air is the alternative, and it earns its place in a specific set of jobs. It gives very high force from a small package, it moves fast, and it holds a load hard without needing a brake or a gearbox to resist back-driving. What it does not give you is fine positioning, quiet running, or a low electricity bill. Everything below is about the five components of that alternative, and what they change about the design.

The Five Components at a Glance

One note on how to use the numbers that follow. The working pressure assumed throughout is 6 to 7 bar (about 85 to 100 psi), which is ordinary shop-air pressure and the range most pneumatic components are built around. Air consumption, force and speed all follow from arithmetic you can check yourself, and I show the arithmetic where it matters, because the specification for a pneumatic exhibit should be reproducible by whoever maintains it. Where a figure depends on the air compressor you choose rather than on the dinosaur, I say so rather than quoting a single number as if it were a rule.

Pneumatic Cylinders

The cylinder is where air becomes movement, and its force output is the easiest thing in the whole system to predict. Multiply the air pressure by the piston area and that is the force you get, minus a little for friction:

F = P × A
Piston area A = π ÷ 4 × bore²
A 63 mm bore has a piston area of 31.2 cm². At 6 bar (600,000 Pa) that gives 1,870 N, or about 190 kgf.

The useful part is what that tells you about sizing. Bore diameter is what you choose, and force rises with its square, so a small step up in bore is a large step up in force:

Two things about that table are worth remembering. First, it is the force on the extend stroke. The retract stroke pushes on the smaller annular area around the rod, so it is always weaker, and on a small bore with a thick rod the difference is significant. Second, the force is probably larger than you need. The mistake I see most often in pneumatic designs is an oversized bore bolted onto a joint that needed half the force: it costs more, it consumes more air on every single cycle, and it makes the movement harder to control smoothly because the cylinder is now fast and violent in proportion to the load. Start from the load at the joint, add the margin you need for acceleration, and pick the smallest bore that clears it.

The details around the cylinder matter as much as its size.

• Cushioning. A cylinder driving a tail or a heavy limb arrives at the end of its stroke with real kinetic energy. Adjustable end-of-stroke cushioning is not a luxury on those joints; without it the impact goes into the frame, the mount and the skin, and it is the sound visitors hear as a knock.
• Mounting. Clevis, trunnion and flange mounts exist to keep the cylinder from being side-loaded. A cylinder that is slightly misaligned against its load will destroy its own rod seal in a fraction of the time it should last, and that failure is usually blamed on air quality. The Zigong general technical specification for animatronic dinosaurs, a draft now out for public comment, makes the same point about electric drives: mount the drive as close as practical to the driven point and keep the fixing accessible for maintenance.
• Rod protection. Outdoor figures need a wiper on the rod. Dust drawn past the seal is what turns into scoring on the rod surface and then into a leak.
• Two positions, or more. A standard cylinder has two positions. If a joint has to stop somewhere in the middle, or hold a partially open jaw, that needs a proportional valve and position feedback, and the cost and complexity jump.

Control Valves

A solenoid valve is a switch for air. A small electrical signal from the controller shifts a spool inside the valve body, and the spool connects the supply port to one end of the cylinder while venting the other. Reverse the signal and the cylinder goes the other way. Almost every animatronic joint in a pneumatic system is driven by a 5/2 valve (five ports, two positions), with the 4/2 variant appearing where the exhaust has to be piped away rather than vented to atmosphere.

There are two families, and choosing between them is the main decision on this component:

• Direct-acting. The coil moves the spool itself. It shifts quickly, commonly in the region of 10 to 40 milliseconds, and it works at any supply pressure because nothing else is driving the spool. The limit is flow, so direct-acting valves suit small and medium cylinders.
• Pilot-operated. The coil only moves a small pilot, and the pilot pressure moves the main spool. These handle large flows with a small coil, but they need a minimum supply pressure to work at all and they shift more slowly, with larger valves taking up to 100 to 150 milliseconds.

That brings up the single most overstated claim in pneumatic animatronic writing: sub-millisecond synchronisation between axes. It cannot happen, and the reason is not the controller. A controller can issue two signals in the same microsecond, but the air still has to shift a spool, travel down a tube and fill a cylinder before anything moves. With direct-acting valves you are in the tens of milliseconds, and with large pilot-operated valves you can be in the hundreds. Air is compressible and it takes time to pressurise a volume. The realistic target for a pneumatic sequence is repeatability within a few tens of milliseconds, and for an animal that is turning its head or opening its jaw, that is imperceptible. If a movement genuinely needs to hit a mark within a millisecond, the right answer is an electric servo axis, not more expensive air valves.

Two practical rules do most of the work on this component. Size the valve by its flow rating, not by its port thread, because two valves with the same 1/4 inch ports can differ by a factor of three in what they pass. Then keep the total pressure drop across the valve down to roughly 10% of line pressure; if it is larger, the cylinder will be slow no matter how big it is. And mount the valve bank close to the cylinders it serves. Every metre of tube between the valve and the cylinder adds both pressure drop and delay, so a valve bank pushed into a corner of the frame for tidy wiring will make the far limbs feel lazy. The same logic appears in the Zigong general specification, which asks for the drive to be mounted as near as practical to the point it drives.

Finally, what actually kills valves in a park is rarely the spool. It is water and oil arriving in the air. A valve that has seized or gone slow is nearly always the messenger, not the problem.

Air Compressors

Sizing the compressor means answering one question: how much free air does the exhibit consume when the worst-case group of movements fires at once? Free air is air measured at atmospheric pressure, and it is always much larger than the swept volume inside the cylinder, because the air was compressed to get there.

Free air per stroke = piston area × stroke × (working pressure + 1 bar)
Take the 63 mm bore from the table, with a 150 mm stroke.
Swept volume = 31.2 cm² × 15 cm = 468 cm³ = 0.47 L
At 6 bar gauge the absolute pressure ratio is about 7:1
Extend stroke = 0.47 × 7 = 3.3 L of free air, retract a little less, so a full cycle is roughly 6.4 L
At 4 cycles per minute that is about 26 L/min, or 0.9 CFM

Six cylinders of that size running together come to around 5 CFM, and once the jaw, neck and tail are added the figure for a mid-size figure typically lands in the single digits to low teens of CFM. Size the compressor at roughly 150% of that peak rather than the average, because the peak is what makes movements look right; a compressor sized to the average will let the pressure sag exactly when the animal is doing its most interesting sequence. Add the small continuous losses too, because a pneumatic installation always has some leakage, and that leakage is drawn from the same compressor.

The receiver tank smooths the difference between the compressor's output and the exhibit's demand. Its useful capacity is the free air released over the pressure band you are willing to let it swing, and the arithmetic is simple:

Stored free air = tank volume × pressure band
A 500 L receiver allowed to swing between 7 bar and 5.5 bar
Holds 500 × 1.5 = 750 L of free air
At an 8 CFM draw (about 227 L/min) that is roughly 3 minutes of ride-through with the compressor off

Heat and water are the two things a compressor brings into the system whether you want them or not. Hot discharge air carries moisture as vapour, and when it cools in the tubing that vapour becomes liquid water sitting in the lowest point of the run, in a valve, or in a cylinder. A park in a humid climate needs a refrigerant dryer, and the size of the dryer should follow the compressor rating with margin for summer conditions rather than a mild average. Siting matters as much as specification here: put the compressor and the dryer in a plant room away from the exhibit, where heat and noise can be handled by the building rather than by the machinery, and where the tank drain can be emptied as part of a routine.

Two environmental notes. Screw compressors are quieter than reciprocating piston units at the same output, but the difference between a well-enclosed machine and an open one is larger than the difference between the two types, so treat the enclosure and the location as the noise control. And air compressors lose capacity with altitude, commonly quoted as about 3% per 300 m (1,000 ft); a park at 1,700 m should expect roughly 15% less output for the same machine. If the site is coastal, specify the corrosion class of the plant rather than assuming a standard industrial finish will survive: grade C4 to ISO 12944, verified by salt-spray testing to ISO 9227, is the usual specification for coastal and offshore environments.

Electronic Controllers

In a pneumatic exhibit the controller does not drive anything directly. It switches valve coils, reads sensors, and holds the sequence. That is a modest job, and it is worth being honest about the size of it: for a fixed show loop running the same choreography every few minutes, a small PLC or a dedicated board is entirely sufficient. Spending on a large controller buys you diagnostic capability and the ability to add axes, not better movements.

The outputs are 24 V DC switched to the valve coils, which draw a few watts each rather than the amperes that non-specialist writing often quotes for them. The inputs are the interesting part. In a simple build these are end-of-stroke sensors, reed or magnetic switches on the cylinder, which tell the controller that a movement has finished before the next one starts. In a more demanding build there are analogue pressure sensors for load feedback, and where a joint has to hold a partial position there is a proportional valve with position feedback and a closed control loop.

Because the achievable timing is set by valves and air rather than by the controller, the useful design target is to make the sequence repeatable and self-checking. Three alarms earn their place on a pneumatic figure, and all three are simple:

• Supply pressure below the working band, which explains almost every symptom of weak or slow movement at once.
• A cylinder that does not reach its end-of-stroke sensor within the expected time, which catches a stiff joint, a failing valve or a partially blocked tube before a visitor sees it.
• Compressor run-hours climbing against cycle counts, which is how you find a leak. If the compressor runs far more often than the show requires, air is escaping somewhere, and the leak will get worse.

The electrical work itself is unglamorous and heavily standardised, which is a good thing. The Zigong general technical specification for animatronic dinosaurs requires program-control enclosures to be insulated, waterproof and ventilated for heat, citing the general machine-electrical safety and earthing standards, and specifies copper core power wiring in the 1 to 2.5 mm² range. It also puts insulation resistance on the list of items checked before a model leaves the workshop, alongside dimensions, electrical connections and appearance. That is the level of rigour a public installation deserves, and it is worth asking any supplier what their pre-despatch inspection actually covers.

Tubing and Connectors

Tubing looks like the least important part of a pneumatic dinosaur and it is the part most often blamed for problems that are really somebody else's fault. Two materials cover almost everything, and they are not interchangeable:

• Nylon (PA11 or PA12) is the choice for long static runs inside the body. It is stiffer, holds its shape, resists pressure well and is less permeable to air than the alternatives. Its weaknesses are that it hates tight bends and that it absorbs a little moisture over time, which changes its dimensions slightly.
• Polyurethane (PU) is the choice for anything that moves. It takes a much tighter bend without kinking and tolerates repeated flexing, which is exactly what happens at a joint that swings all day. It is softer, so it needs more support along its run.

Bore size is decided by pressure drop, not by what the fitting accepts. Every metre of tube and every fitting costs you some of the line pressure, and the flow through a small bore rises steeply as you demand more speed, so the loss rises with it. The design rule is to keep the total drop from the valve bank out to the cylinder at roughly 10% of line pressure or less. Get that wrong and you create the most common fault in the whole system: a limb that is slow and weak at the far end of the machine while everything near the compressor performs perfectly. The instinct is to turn the regulator up, which over-pressurises the whole system to fix one joint and shortens the life of every seal in it. The correct fix is a larger bore on that run, or a shorter one.

Two rules of thumb are worth writing on the inside of the control cabinet. The minimum bend radius is about eight times the tube's outside diameter, so a 6 mm tube needs roughly 48 mm, and a tight bend is not a small loss: it restricts flow badly and it becomes the point where the tube eventually cracks. Note that this is a minimum, not a maximum, and it is the error I see most often in copied design notes. And clamp tube runs at close intervals, because an unclamped length of tubing inside a moving animal will rub against the frame until it fails. At the joints that swing, use swivel or banjo fittings so the tube itself is not the pivot.

Push-to-connect fittings are the standard fixing for this work because they make a dinosaur serviceable in the field. They are also where most leaks start, and the cause is usually the same two things every time: a tube that was not pushed fully home into the fitting, and a tube that was cut at an angle or left with a burr on the end so it never seats properly. When a joint goes weak, check those before you suspect the valve or the cylinder. When you cut tube, use the proper cutter rather than side cutters, and push it in until it stops, then pull back gently to confirm it is gripped. If the system carries oil from a lubricated compressor, specify seals that are compatible with it; a seal that swells in mineral oil will leak and then fail.

Where the Cost Actually Sits

Pricing for a pneumatic exhibit is quoted per project, and it is worth understanding which decisions drive it, because they are not the decisions most people expect.

• The compressor and its air treatment are the largest single item, and the plant room that houses them is usually a building project rather than part of the dinosaur order. This is the cost that surprises buyers who compared a pneumatic figure against a motor-driven one on the model price alone.
• Running cost is dominated by the compressor. A pneumatic exhibit holds pressure all day whether or not it is performing, so the energy bill follows the running hours and the leak rate rather than the show schedule.
• Leaks are the hidden line item. Leakage accumulates quietly: a fitting that was never fully seated, a tube rubbed through, a rod seal on its way out. A compressor that keeps starting when the exhibit is idle is telling you money is going out through the fittings.
• Every axis has a valve, a cylinder and a run of tube. The movement list is therefore close to linear in cost, which is why a restrained movement set that is well programmed nearly always looks better value than a long list of axes that all do something similar.
• Air suits force, not finesse. If you want a slow, precise, silent movement, an electric servo axis will be cheaper to buy, cheaper to run and easier to maintain. If you want a jaw to snap shut, a tail to flick, or a heavy section to be held up hard, air is the simpler answer.

That last point is the honest summary of the whole system. Pneumatics and electric drive are not competing for the same jobs, and the best large figures mix them. On our own builds the drive type is confirmed per model, motor or pneumatic, against the movement list and the site conditions. If you are specifying an exhibit and you are not sure which joints should be which, that is a conversation to have before the frame is designed rather than after. Our article on how animatronic dinosaur movements work covers the mechanism side in more detail, and the animatronic dinosaur range shows the sizes and movement options the drive decision applies to.

Frequently Asked Questions

What air pressure do pneumatic animatronic dinosaurs run on?

Most systems run at 6 to 7 bar, which is about 85 to 100 psi — ordinary workshop air pressure, and the range standard pneumatic components are designed around. The receiver tank is kept at a higher pressure and the working pressure is set by a regulator, so the figures you see on the compressor and the figures at the cylinder are deliberately different.

Why use air instead of electric motors?

Air gives a lot of force from a small, simple package, it moves quickly, and it holds a load without needing a brake because the cylinder is naturally stiff against the load. It is also forgiving of being overloaded, which is an advantage on a figure the public can stand next to. What it does not give you is precise positioning, quiet operation or a low energy bill. Most figures are better served by electric drive on most joints, with air reserved for the joints where force and speed matter more than finesse.

How much compressed air does one dinosaur need?

Work it out from the cylinders rather than from a rule of thumb: piston area times stroke times absolute pressure gives the free air for one stroke, roughly double that for a full cycle, and multiply by the number of cycles per minute. Add every movement that fires at the same time in the busiest part of the sequence, then size the compressor at about 150% of that peak. A mid-size figure usually lands in the single digits to low teens of CFM.

How do you keep pneumatic movements synchronised with the sound?

One controller writes one timeline, and the movement signals and the audio events are triggered from that same timeline, so the two cannot drift apart the way two separate players would. Realistic accuracy on air is a few tens of milliseconds, set by how long the valves take to shift, and that is well inside what the eye notices on a dinosaur. The Zigong general technical specification requires that movement timing, cycle and speed match the design and stay in step with the sound, and that is the standard we build to.

What maintenance does a pneumatic system need?

Most of it is air quality rather than the dinosaur. Drain the receiver, confirm the dryer and filters are working, and check the drain traps at the start of the season and regularly through it. Then look at the mechanical items: tube runs at the moving joints for rubbing and chafing, fittings for security, cushioning settings, and the rod seals on the joints doing the most work. Cylinder seal kits and valve seals are normal service items, and how often they are needed depends on the duty cycle and on how well the air is dried. Our checklist of critical component checks covers the wider figure.

Can a pneumatic dinosaur operate outdoors in winter?

It can, and the limit is water rather than cold. Moisture that condenses in the lines can freeze and block a valve or a tube, and standard pneumatic seals are usually rated no lower than around −20 °C. The general specification for animatronic dinosaurs sets the exhibition environment at −40 °C to +40 °C, so on a cold site the pneumatic drive is the part that narrows that range, not the steel frame. Dry the air properly, keep the plant room above freezing, and the exhibit will run. On sites where winter operation is essential and the movements are not force-critical, electric drive is the easier specification.

If you are planning a pneumatic exhibit, the useful starting point is the movement list and the site conditions: which joints need force, which need finesse, where the plant room will be, and what the ambient conditions are. Send us those and we will tell you honestly which joints should be air and which should not. You can contact us directly, or see how the numbers look on a multi-axis controlled figure.