Hydraulic Actuation Inside Moulds: Force, Heat, Fluid, and Fit

A core the size of your thumb has to hold its position against thousands of pounds of plastic trying to shove it out of the way — and then, a second later, break free from that same plastic as it shrinks and grips like a clenched fist. It has to do this thousands of times a shift, in a steel pocket barely larger than the part itself, in heat that punishes everything you put near it, without leaking a drop of oil into a cavity where one droplet means a scrapped part.

That is the problem hydraulic actuation solves inside a mould. And it is worth understanding from first principles, because the difference between a tool that runs for years and one that fights you every cycle usually comes down to decisions made before a single cylinder was ordered — decisions about force, heat, fluid, and fit.

This is a vendor-neutral engineering guide. It covers what an in-mould hydraulic actuator actually is, the jobs it does, how to size one without guessing, and — just as importantly — where hydraulics stop being the right answer and something else should take the load. Wherever the numbers matter, they are here. Where the industry’s own sources disagree, you will see the disagreement rather than a tidy figure that hides it, because a sizing decision built on a number you cannot defend is a decision waiting to fail.

If you design tools, build them, or run them, this is written for you.

Why hydraulics live inside moulds at all

Start with the one equation that explains everything: Force = Pressure × Area.

In-tool hydraulic supply pressures run high — commonly 150 to 320 bar. Feed that pressure into even a small-bore cylinder and you get tonnes of force out of an envelope that fits in your palm. That is the entire reason hydraulics dominate high-force, tight-space mould work: nothing else delivers that much force density in that little room. A core that must resist injection pressure, or wrench itself free of a part that has shrunk onto it, needs exactly that — large force in a small package.

But force density is only the first of three reasons. The second is holding force without continuous energy. Locking-style cylinders preload a core and then hold it after locking, even when supply pressure is removed entirely — some designs claim full preload to rated load at zero pressure. That property turns out to matter enormously, for a reason we will come back to: many presses drop hydraulic pressure to the core circuits during injection. A non-locking cylinder can get pushed back at the worst possible moment. A locking one will not.

The third reason is smoothness and control. Oil is far less compressible than air, so it gives steady, controllable motion and high, controllable force, with speed governed cleanly by flow rate. Air springs; oil holds. Inside a mould, where motion has to be timed against an injection cycle to the millisecond, that difference is not academic.

The anatomy, briefly

A hydraulic cylinder in a mould is a linear actuator: a bore housing a piston joined to a rod, with one or two ports admitting pressurised oil, sealed by piston seals, rod seals, and wiper seals, and guided by wear bands. It mounts into a mould plate or pocket — as a cartridge, a block or monobloc, a threaded body, or a flange style — and converts oil pressure into a controlled push or pull that drives a mould element.

Two pieces of physics you will use constantly:

• Extend force = bore area × pressure. Retract force = (bore area − rod area) × pressure. Retract is always weaker than extend, because the rod subtracts area on the return side. This is the “differential area” trap that catches people who size a cylinder on its extend force and then find it can’t pull the core back hard enough.
• Cylinder speed = flow rate ÷ piston area. The pressure gives you force; the flow gives you speed. A press or auxiliary pump that can’t supply enough GPM will leave a long-stroke cylinder crawling, no matter how much pressure is on tap.

Stroke is constrained by the mould’s daylight and plate thickness — specify the movement you need plus margin, and no more. And you choose single-acting (spring or pneumatic return, oil drives one way) or double-acting (oil both ways, full force in both directions) based on whether you need real force on the return as well as the push.

The space-saver: pressure intensifiers

There is one more building block worth knowing before we get to the jobs, because it solves the space problem in a clever way. A pressure intensifier — or booster — uses two mechanically linked pistons of different areas. Pressure rises inversely with the area ratio, so a 1:4 area ratio quadruples the pressure. That lets a modest supply produce high local pressure for small oil volumes, which is exactly what you want inside a tool where space is scarce and you don’t want to plumb high-pressure lines across the whole mould.

As a sense of the range available: Enerfluid states that its air/oil multipliers reach intensification ratios from 1:25 to 1:64, producing oil pressures up to 512 bar from just 8 bar of air input, while its oil/oil multipliers span 1:2 to 1:10 with a maximum outlet of 320 bar. Those are the supplier’s published figures — useful as a feel for what the technology can do, and worth verifying against your own conditions.

The jobs hydraulics actually do

Force figures in this section depend heavily on geometry. Treat every number as a starting point and back-calculate for your own tool. Injection moulding and high-pressure die casting (HPDC) are separated wherever they genuinely differ — and they differ more than most guides admit.

Core pulls are the flagship job. A movable core retracts before ejection to free undercuts, side holes, or internal threads. Hydraulics are the default for medium and large, high-force cores, because they deliver high retract force with controllable speed. The forces here are real: retract force runs “typically 5 to 50+ tons” by one moulder’s account. The limiting case is almost always initial breakaway — the moment you break the core free from the plastic that has shrunk and clamped onto it. The rest of the stroke needs far less force. Size for the breakaway, not the running force, or you will under-spec the one moment that matters. For die casting specifically, one source recommends hydraulic core pulling for pull distances of roughly 30 to 100 mm, with a pressure relay tripped at about 1.2× the rated core-pull force for overload protection. Threaded cores are a different animal — they often need rotary unscrewing action rather than a linear pull.

Slides and side-actions carry side cores to form side features. Hydraulics replace or supplement the old angle-pin cams when the force, stroke, or timing exceeds what mechanical geometry can deliver — and crucially, they let the action run independent of mould open and close. Where several small cores must nest and time independently, compact narrow-format cylinders earn their place, because they pack into tight spaces several at a time.

Lifters free internal undercuts, and hydraulics step in where cam geometry can’t deliver the needed stroke or force, or where independent timing is required.

Ejector assist comes in where the press’s own ejector stroke or force isn’t enough, where multi-stage sequenced ejection is needed, or where the ejector system has to be drawn on a specific cadence.

Valve-gate actuation drives hot-runner valve pins — a piston reciprocates the pin to open and close the gate. This is the one job where hydraulics are not the market default. By one trade account, hydraulic valve-gate systems make up less than 20 percent of the market, while pneumatic systems hold over 80 percent. Hydraulics get chosen here only where high, fast pin force is needed: high oil pressure produces strong stem force from a small piston, and oil’s low compressibility gives a crisp open and close. The trade-off is honest and worth stating — putting hydraulic fluid, cooling, and seals into the hot manifold environment is a recognised drawback, not a free win.

Mould locking and clamping is the classic compact-cylinder job: short stroke, high force, used to preload and hold cores or plates against injection or casting pressure. Large force, minimal movement.

Injection versus HPDC: why the context changes everything

The forces tell the story. Injection nozzle pressures run roughly 10,000 to 25,000 psi (about 690 to 1,725 bar), and the holding force a core needs follows directly from projected area × pressure. (One industry source puts the current average injection pressure even higher, at 25,000 to 30,000 psi — worth knowing, even if the conservative range is the right one for a worked sizing example.)

HPDC is harsher in a different way. Injection and intensification pressures there are commonly quoted from 1,500 psi to well over 25,000 psi, with clamp forces reaching around 4,000 tons, and the cores must be locked against being blown out during the fast shot. HPDC therefore demands more robust locking and hotter-rated seals and fluids than injection moulding does. When you read a force figure, always ask which world it came from — a number that is comfortable in injection can be dangerously optimistic in die casting.

Getting the oil there: supply, sequencing, and standards

A cylinder is only as good as what feeds it. There are three broad ways to supply in-mould hydraulics, and the choice has real consequences.

Press-integrated supply uses the machine’s own hydraulics through auxiliary valve banks. It is simple — but it shares flow with everything else and, as noted, may drop pressure during injection. External standalone HPU gives you a dedicated power pack with independent pressure and flow, which is the answer when press supply is insufficient or drops at the wrong time. In-mould intensifier circuits put a small booster on or in the tool, producing high local pressure from a modest supply and saving both lines and space.

There is a fourth approach worth its own mention: pneumo-hydraulic units, which generate oil pressure from compressed air. They suit shops with plenty of compressed air already plumbed, need no electric HPU, and deliver immediate high pressure for small oil volumes. Their limits are equally clear — they displace only a small oil volume per stroke, so they’re best for short-stroke and locking duty, their output depends on air-supply quality, and they’re not suited to high-flow, long-stroke work. As a sense of scale, Enerfluid states its pneumo-hydraulic power units provide effort from 1 to 500 kN, total strokes of roughly 32 to 500 mm with working strokes of 6 to 45 mm, controlled at 6 to 9 bar of air, up to 400 bar.

The press-side trap

Here is the failure that surprises people, so it is worth stating plainly: many presses drop hydraulic pressure to the core circuits during injection. A non-locking cylinder, holding a core against the incoming melt, can be pushed back exactly when it must not move. The fixes are known — preloading or locking cylinders that hold at zero pressure, or a dedicated supply that never drops. But you have to design for it. Assuming press pressure stays constant through the cycle is one of the most common and most expensive wrong assumptions in tool design.

Sequencing and the scope boundary

Hydraulic actions are timed against the press cycle: core set before injection, core pull before or with mould opening, ejector after opening. Position is confirmed by limit switches, inductive sensors, end-of-stroke detection, and pressure-decay monitoring — die-casting core pulls typically use a pressure relay tripping around 1.2× rated force plus end-position confirmation, so the tool never closes on a core that failed to pull.

Know where the scope line sits. The hydraulic vendor supplies the cylinders, any integrated position sensing, and the intensifier or power-unit hardware. The press and its controller own the cycle logic, valve switching, and interlocks. One useful, low-commitment move: position sensors can be added to hydraulic actuators at build, letting you gain control later without committing to a full electric system now.

The EU safety baseline

For machines placed on the European market, EN ISO 4413:2010 is the harmonised Type-B standard for hydraulic fluid power, giving presumption of conformity with the Machinery Directive 2006/42/EC. One nuance that catches people: it is not harmonised to the Pressure Equipment Directive. Around it sit ISO 12100 (risk assessment), ISO 4406 (fluid cleanliness), ISO 6743-4 and ISO 12922 (fire-resistant fluid classification), ISO 1219 (symbols), and ISO 13849 with IEC 60204-1 (control and electrical safety). None of this is optional reading if you’re building for Europe.

The hard-engineering core: heat, fluid, and cleanliness

This is where tools quietly succeed or fail, long after the force calculation is done. Three variables — temperature, fluid, and cleanliness — decide how long your cylinders last and whether they leak where they must not.

Temperature drives seal choice

Operating temperatures split sharply by process. In injection moulding, mould and cavity control temperatures commonly run 20 to 120°C depending on the polymer; the melt that touches the cavity is far hotter (180 to 350°C), but the cylinder body sits in cooled plate steel, so its temperature is usually well below the cavity surface. In HPDC, the die surface is held relatively low against the melt — “normally, around 200°C” by one account — while molten aluminium at 660°C and above enters the cavity. Cylinders near an HPDC die live in a hotter, harsher world than their injection counterparts.

Seal material follows directly from the temperature the cylinder body actually sees:

• NBR (nitrile) is the general-purpose hydraulic seal: continuous service to about +110°C (short-term ~130°C), down to roughly −35°C. Cheapest, with good abrasion and oil resistance.
• FKM (Viton-class) handles continuous service to approximately +205°C by a conservative standard rating — and this is grade-dependent, with some sources citing +230°C and special grades (such as Viton GLT, rated roughly −45°C to +275°C) extending the window further. FKM brings superior oil, chemical, and ozone resistance, making it the default near hot tools and aggressive mould-release exposure. One caveat: many FKM dynamic seals are pressure-limited and not ideal for water-based fluids.
• PTFE-based seals offer the widest temperature and chemical range with low friction, used for the most demanding dynamic seals and back-up rings.

The discipline that matters more than any single number: specify the seal by grade, and verify it against the cylinder-body temperature, not the cavity temperature. Speccing seals to cavity heat over-builds them; assuming ambient under-builds them. Measure the body. (The industry’s own sources genuinely disagree on the FKM ceiling — anywhere from ~205°C to ~275°C depending on grade — so resolve it by grade and use the conservative figure for standard continuous duty.)

Fluid: standard oil, until fire enters the picture

For injection tooling away from fire risk, standard mineral oil (HLP/HVLP) is the default — best lubricity, best cost. The picture changes near molten metal.

In HPDC, foundries, and steel mills, a leak can ignite, so fire-resistant fluids are the standard, and sometimes the legal requirement. Under ISO 6743-4 / ISO 12922 they fall into families: HFA (high-water synthetic-aqueous), HFB (water-in-oil — not approved in Germany), HFC (water-glycol, the die-casting market standard), and HFD (anhydrous synthetics, including HFDR phosphate esters and HFDU polyol esters with higher temperature capability). The trade-offs are real: water-based fluids reduce lubricity and pressure capability and demand stricter maintenance, while phosphate esters attack some seals and carry a low viscosity index. And the seal must match the fluid — water-glycol, for instance, is unfriendly to some FKM compounds. Pairing an incompatible seal with a water-based or phosphate-ester fluid is a self-inflicted failure.

For medical, packaging, and food-contact work, the priority flips to corrosion-resistant materials, leak-free architecture near the cavity, and cleanable surfaces — and an honest question about whether hydraulics belong there at all versus a sealed or electric alternative.

Cleanliness is the number-one killer

If you take one maintenance lesson from this guide, take this one. Contamination is the leading cause of hydraulic failure — and the industry’s own figures, while they vary, all point the same direction. One Fluid Power Journal source attributes “as many as 80–90% of hydraulic system failures” to fluid contamination; a second Fluid Power Journal source cites “65% to 90%”; a 2021 academic study attributes about 41.1% specifically to contamination and oil pollution. The exact share is contested. The conclusion is not: dirt in the oil is what kills these systems.

The control is ISO 4406, which codes particle counts at ≥4, 6, and 14 µm as a three-number code (e.g., 18/16/13) on a logarithmic scale where each step roughly doubles contamination. Specify a target class for your most sensitive component — sensitive servo and proportional valves often want 16/14/11 or cleaner — and protect it with filtration and clean handling. Ask your supplier what cleanliness the component actually needs, rather than guessing.

One more degradation driver: operating fluid above roughly 60°C accelerates seal breakdown. Heat and dirt are the two enemies, and both are manageable if you design and maintain for them.

The sizing framework: eleven questions, in order

This is the spine of the whole discipline. Work these in sequence — each one constrains the next, and the traps listed are where real tools go wrong.

1. Required force at the part. What force holds the core during injection, and what breaks it free on the pull? Use projected area along the axis of movement × nozzle/cavity pressure, with nozzle pressure around 10,000–25,000 psi, and account for shrinkage and taper at breakaway. Trap: using exposed surface area instead of projected area, or treating running force as the limit when breakaway is far higher.

2. Available envelope. What actually fits the pocket, given plate thickness and clearance to cooling channels and other actions? Trap: a bore that collides with conformal-cooling channels or leaves no room for ports and lines.

3. Stroke. How far must it move — undercut depth plus clearance plus margin? Trap: forgetting eject clearance and under-stroking so the part clips.

4. Operating temperature at the cylinder body. How hot is the cylinder actually, given cooling layout and distance from the hot zone? This drives seal and fluid choice. Trap: speccing seals to cavity temperature, or assuming ambient — measure the body.

5. Duty cycle and cycle-life target. How many cycles over what service life? Trap: ignoring that faster cycles raise seal temperature and shorten life.

6. Single vs double-acting, mounting style, port orientation. Do you need force both ways? Cartridge, block, threaded, or flange? Trap: a mounting that can’t be serviced without pulling the entire tool.

7. Supply pressure available vs intensifier needed. Is there enough pressure and flow — and is it maintained during injection? Trap: assuming press pressure is constant; many presses drop core-circuit pressure during injection, so consider locking cylinders or an intensifier.

8. Cleanliness and fluid class. Medical, food, wash-down, or fire-risk HPDC? Set an ISO 4406 target and a fire-resistant fluid requirement where needed. Trap: pairing an incompatible seal with a water-based or phosphate-ester fluid.

9. CAD model availability. Can the supplier give you a model early in tool design, so you design the pocket around real geometry? Trap: designing the pocket before the component geometry is confirmed. (Enerfluid, for example, supplies CAD on request rather than as a self-service download.)

10. Lead time, serviceability, spares. What’s the delivery, and is there a seal-kit and spares pipeline? Trap: a cheaper cylinder with no spares behind it.

11. Retrofit vs new-build. Fitting an existing tool or designing fresh? Trap: assuming a like-for-like swap fits when bore, port, or mount differ.

The traps that recur, in one box: oversizing “to be safe” (wastes envelope and oil, slows motion); under-spec’d seals for the temperature or fluid; ignoring that retract force is lower than extend force; forgetting return-line routing; and assuming press supply pressure holds constant during injection. If you check nothing else, check those five.

When it breaks: failure modes and maintenance

Tools fail in a predictable order, and knowing it tells you where to look first.

Contamination leads — dirty fluid, by the figures above, is the dominant cause. Particles scar rods and bores, and the scarring then shreds seals. Seal degradation is next and most visible: hardening from heat, chemical attack from the wrong compound or fluid, extrusion from pressure spikes, plain wear. Seals are the weakest, most-replaced part, and an external leak is usually the first symptom you’ll see. Leaks at or near the cavity are their own category — a failed wiper seal admits contaminants, and oil near the cavity is both a quality defect and, in HPDC, a fire risk. Hose and fitting fatigue comes from flexing lines and over-tightened fittings under cyclic pressure. And intensifier wear demands especially clean fluid, with tight-tolerance boosters often calling for 10 µm filtration or better.

A healthy maintenance regime is unglamorous and effective: scheduled visual inspection for leaks at the rod junction — using cardboard, never a bare hand, because a pinhole high-pressure leak is an injection-injury hazard; periodic fluid sampling (roughly every 500 operating hours) tracking the ISO 4406 trend; filter changes on a schedule rather than on symptoms; desiccant breathers; keeping fluid below about 60°C; and keeping seal kits and spare cylinders on hand to cut downtime. The signs that say replace the cylinder rather than nurse it: drift, visible droplet leakage, sluggish or reduced force, a scored rod, or internal scoring deeper than a fingernail catches.

On service life, resist the urge to quote a single MTBF number — there isn’t a defensible universal one. Life is set by pressure, temperature, cleanliness, side-load, and seal compound together. Ask a supplier for a rated figure for your actual conditions, and treat anything more general with suspicion.

The retrofit reality is encouraging if you’ve sized well: many cylinders can be re-sealed in place or swapped without re-machining the pocket — if the replacement matches bore, stroke, mount, and port. A compatible-footprint swap takes minutes. A mid-life replacement that changes the footprint means re-machining and is often not realistic without reworking the tool. Which is the whole argument for fit-first sizing at design time: it’s what makes service realistic years later.

Where the industry is heading

A few currents are reshaping in-mould hydraulics, and they pull in tension with each other.

Tools are getting smaller and more complex — thin-wall and multi-cavity work pack more side-actions into less space, raising demand for compact, narrow-format cylinders that nest and time independently. Thin-wall also needs higher injection pressure, which raises core loads. At the same time, cycles are getting faster, and since cooling dominates the cycle — one instrumented study found 85% of the cycle spent cooling, with sources broadly converging on 60–85% — compressing cycle time raises seal temperatures and shortens seal life. Faster tools are harder on seals, full stop.

Conformal cooling is the quiet space war: 3D-printed cooling channels (cutting cooling time by 10–57% in one review, and more in advanced lattice designs) compete for the same internal real estate as cylinders and lines, forcing tighter integration than ever. Automation raises the bar on verified hydraulic timing and end-position confirmation for unattended running. And sustainability pressures push toward zero-pressure locking, more efficient HPUs, fire-resistant and biodegradable fluids, and leak-free architectures.

Running underneath all of it is a genuine judgment call: sensors can be added to hydraulic actuators at build to gain control later — but a real counter-trend favours keeping in-mould hydraulics simple and avoiding sensor and firmware lock-in. There’s no universal right answer there; there’s only the right answer for your shop.

When not to use hydraulics

A guide that only tells you where its subject wins is a brochure. Here is the honest boundary.

Pneumatic is cheaper, cleaner (no oil), and uses air you probably already have — but it delivers much lower force for a given size and offers less precision and holding capability. It’s the right call for light, fast, simple actions and for cleanroom contexts where oil is unwelcome. It already dominates valve-gate actuation at over 80% of the market for exactly these reasons.

Electric and servo actuation offers the highest precision, programmable position and speed, cleanliness, and energy efficiency — at higher cost, with the commitment made at build time, and with motors that dislike mould heat and need distancing or cooling. It’s the right answer where programmable pin or position control and cleanliness dominate the requirements.

Mechanical cams, angle pins, and springs are simplest and cheapest, need no supply at all, and are reliable for short strokes — but stroke, force, and timing are fixed by geometry and tied to mould open and close. Right for modest undercuts, wrong the moment you need independent timing or more force than geometry can give.

So: choose hydraulic when you need high force from a small envelope, holding force, long or independent strokes, or robust performance in hot HPDC environments. Choose otherwise when force is modest, cleanliness is paramount, or programmable precision is the priority. Knowing both halves of that is what separates an engineer who understands the technology from a supplier who only sells it.

The European context

Two regions concentrate this demand. In Italy, Lombardy is a dense tooling and die-casting cluster — Brescia province, and Lumezzane especially, hosts a concentration of die-casting mould makers serving automotive, lighting, heating, electronics, and design end-markets, with Treviso and the Veneto adding plastics moulding. In Germany, North-Rhine Westphalia and Baden-Württemberg are core automotive tooling regions, where a dense OEM-and-supplier base drives demand for complex, multi-action moulds. Across both, the same currents — miniaturisation, multi-action complexity, faster cycles, automation, giga-casting in HPDC, and reshoring — favour compact, fit-first, CAD-ready hydraulic components backed by a reliable European supply chain.

Frequently asked questions

What is a hydraulic actuator in a mould? A linear cylinder mounted in a mould plate or pocket that converts pressurised oil into a controlled push or pull, driving a core, slide, lifter, ejector, or valve pin.

When should I use hydraulic core pulls instead of mechanical? When the force, stroke, or timing exceeds what cams and angle pins can deliver, or when you need the action to run independent of mould open and close.

How do I size a hydraulic cylinder for a core pull? Start from required force at the part — projected area × cavity pressure, sized for breakaway, not running force — then work through envelope, stroke, body temperature, duty cycle, supply pressure, and fluid class. The eleven-step framework above is the full sequence.

What force does a core pull need to resist injection? It follows from projected area × nozzle/cavity pressure (commonly 10,000–25,000 psi); reported retract forces run from about 5 to 50+ tons depending on geometry.

Why does retract force differ from extend force? The rod subtracts area on the return side, so retract force = (bore area − rod area) × pressure, always less than extend force.

What seal material for hot moulds — NBR or FKM? NBR serves continuously to about 110°C; FKM to roughly 205°C and higher by grade. Near hot tools, FKM is usually the default — but specify by grade and check it against the cylinder-body temperature, not the cavity.

What hydraulic fluid for die casting? A fire-resistant fluid, because leaks near molten metal can ignite. Water-glycol (HFC) is the die-casting market standard, with anhydrous HFD options for higher temperatures.

What ISO 4406 cleanliness should I specify? Set the target for your most sensitive component — sensitive servo and proportional valves often want 16/14/11 or cleaner — and ask the supplier what the component requires.

What is a pressure intensifier, and when is it worth it? Two linked pistons of different areas that raise local pressure inversely to their area ratio, producing high pressure for small oil volumes. Worth it when space is scarce and you don’t want high-pressure lines running across the tool.

Why do presses drop hydraulic pressure during injection? Press hydraulics are shared across the machine cycle, so core-circuit pressure can fall during the injection phase. Use locking cylinders or a dedicated supply so the core can’t be pushed back.

Can I retrofit hydraulics into an existing tool? Often yes, if the replacement matches bore, stroke, mount, and port — a compatible-footprint swap is quick. Changing the footprint means re-machining the pocket, which is frequently impractical.

Which EU standards apply? EN ISO 4413:2010 is the central harmonised standard for hydraulic fluid power under the Machinery Directive, alongside ISO 12100, ISO 4406, ISO 6743-4 / ISO 12922, ISO 1219, and ISO 13849 / IEC 60204-1.

Sizing a specific tool? The decisions above — force, envelope, stroke, body temperature, duty, and fluid class — are exactly the inputs that determine whether a cylinder fits and lasts. Working them out on paper before you machine the pocket is the single highest-leverage hour in the whole design, and it’s what makes later service realistic instead of a tool rebuild.