Updated: 2026 Technical Guide · By Sailwin Engineering Team
Extrusion blow mold cooling design determines how much of the cycle you cannot avoid. Extrusion takes a fixed time, clamping and blowing take a fixed time, and de-flashing takes a fixed time — but cooling takes however long the mould and the cooling circuit require. On most blow moulding lines, cooling is the largest single component of cycle time, and on large parts it is not unusual for it to be the majority of it.
The mistakes are consistent and they are all mould-side. Cooling channels are drilled along the mould plate rather than following the part, so the thickest sections — the ones that actually control the cycle — are the furthest from the water. Circuits are connected in series because it is easier to pipe, so the last circuit in the chain receives water that has already absorbed heat and the part cools unevenly. Flow is set by whatever valve was open at commissioning, so turbulence is lost and heat transfer falls away without anyone noticing.
Sailwin builds extrusion blow moulding machines for containers from 0.5 L to 1,000 L, including models such as SW-S30L (up to 30 L, 600 pieces per hour) and SW-S1000L (up to 1,000 L, 250 pieces per hour), with PLC control using Siemens or Mitsubishi hardware, FESTO combined blowing valves and SMC cylinders. Machines are CE marked, built under ISO 9001:2015, and carry a 2-year whole-machine warranty. This guide covers the cooling decisions that decide your cycle time before the mould is cut.
Key Takeaways
- Cool the thick sections first, not the average: a part cools when its thickest controlling section has solidified, so channel placement has to follow the section thickness map rather than the mould plate outline. Getting this right is worth more than lowering the water temperature.
- Parallel circuits beat series circuits: water supplied at 8–12 °C arrives at the last circuit of a series chain already warm, which produces uneven cooling and warpage. Balanced parallel circuits deliver the same inlet temperature to every zone.
- Flow matters as much as temperature: heat transfer in a cooling channel depends on turbulent flow. A circuit that is starved of flow loses efficiency regardless of how cold the supply is, so flow should be specified and verified per circuit at commissioning.
Reducing Cycle Time on an Existing Blow Mould?
Send the part drawing and current cycle time — our engineers return a cooling and machine assessment within 24 hours.
1. Why Cooling Dominates the Blow Moulding Cycle
In extrusion blow moulding, the parison arrives at the mould already hot, and the mould has to remove enough heat for the part to hold its shape when it is ejected. The extrusion stage has a physical minimum set by output rate and parison quality. The clamping, blowing and venting stages are measured in fractions of a second. Cooling is the stage where the timetable is negotiable, and it is therefore the stage where cycle time is either won or lost.
The reason cooling resists improvement is that heat has to travel. It leaves the polymer, crosses the part-to-mould interface, passes into the mould steel, and is carried away by the water. Each step has a resistance, and the largest is usually the plastic itself — which is why thick sections, not large parts, set the cycle. A container with a heavy base and a thin body cools according to the base, because that is the section that is still soft when everything else has set.
This has a direct commercial consequence. Two containers of the same volume can have very different cycle times if their wall thickness distributions differ, which means the same machine model can be quoted at different outputs for different moulds. When you specify a mould, you are also specifying a cycle — and that cycle should be estimated before the mould is cut, not measured after.
Do not ask how fast the machine is. Ask how long the thickest section of your part takes to solidify, and whether the mould is designed to remove that heat from that section.
2. The Six Cooling Faults You Will Actually Meet
Cooling problems rarely present as a cooling problem. They present as warping, dimensional drift, surface marks or an unexplained cycle time, and the maintenance response is usually to lower the water temperature — which rarely helps and often causes condensation problems on the mould face. The table below maps each symptom back to its cooling cause.
| Symptom | Cooling mechanism behind it | Corrective direction |
|---|---|---|
| Part warps after ejection | One area is colder than another, so shrinkage is uneven and the part distorts as it continues to cool outside the mould | Balance the circuits; move cooling to the slow-cooling area rather than lowering the overall supply temperature |
| Shrink marks or sink on a heavy section | A thick section cools last and pulls material from the adjacent wall as it solidifies | Add local cooling at the thick section — bubbler or baffle — before adjusting the parison programme |
| Dimensional drift through a shift | Water temperature or flow changes as the chiller loads up or a filter blinds, so the effective cooling rate changes | Verify supply temperature and per-circuit flow at intervals; trend them rather than relying on a single commissioning reading |
| Cycle time longer than predicted | The controlling section is not adequately cooled, so ejection has to wait for the whole part rather than the part being ejected when it is rigid | Map section thickness and check channel proximity to the thickest areas; usually a mould design fix, not a water temperature fix |
| Condensation on the mould face | Supply water is colder than the ambient dew point, so moisture forms on the mould surface | Raise the supply temperature and improve flow instead; more flow at a moderate temperature beats less flow at a lower one |
| Base corners thin and crack in service | Base corners are both the thickest section and the hardest to cool, so the material stretches thin and remains soft longest | Dedicated base cooling plus a parison programme that carries material to the leading end |
3. Circuit Design Principles That Decide the Cycle
Six design principles cover most of what makes a blow mould cool well. They are inexpensive to apply at the design stage and expensive to retrofit, because retrofitting means re-machining a mould that is already in production.
| Principle | What it means in practice | Why it matters |
|---|---|---|
| Follow the part, not the plate | Channels positioned relative to the cavity surface, at a consistent distance from the part wall wherever geometry allows | A channel that drifts away from the wall loses effectiveness quickly; distance is the dominant variable in how fast that section cools |
| Parallel, balanced circuits | Manifold feeding each circuit directly, with flow balancing so every circuit sees a similar pressure drop | Series chains deliver progressively warmer water; the last circuit cools worst, which is a warpage pattern built into the mould |
| Bubblers and baffles at deep features | A bubbler directs water up the inside of a core feature so it reaches the tip; a baffle splits flow so both sides of a blade are cooled | A straight drilled channel cannot cool the tip of a deep, narrow core; without a bubbler the tip is effectively uncooled and controls the cycle |
| Turbulent flow, not just flow | Channel diameter and flow rate selected so the water is in turbulent flow rather than laminar | Laminar flow insulates the channel wall with a slow-moving boundary layer, which sharply reduces heat transfer even with cold water |
| Load-aware temperature | Chiller capacity sized for the total heat load at the target cycle, not for a single-part average | As cycle time shortens, heat load per hour rises; a chiller sized for the slower cycle becomes the new constraint |
| Commissioning record | Record supply temperature, return temperature and flow per circuit on the day the mould enters production | Without a baseline, a slow drift in flow or temperature is invisible until dimensional quality falls |
Cooling Design Review Before the Mould Is Cut
Send the container drawing with wall thickness sections — we return a cooling and machine recommendation within 24 hours.
4. Case Study: Cycle Time Reduced by Mould Cooling, Not by the Machine
A packer producing containers up to 30 L on an SW-S30L at 600 pieces per hour was running a cycle time well above the specification, with warping on the larger sizes and inconsistent base dimensions.
- Container mould with a heavier base than the body, so the base controlled the release point rather than the average wall
- Cooling circuits piped in series because it was simpler at the mould shop, so the last circuit ran warmer than the first
- Response to the problem had been to lower the supply water temperature, which produced condensation and no cycle-time gain
- Circuits re-piped onto a manifold so each zone receives water at the same inlet temperature, replacing the series chain
- Dedicated base cooling added using a bubbler arrangement, so the thickest section received water at the point where the heat actually is
- Flow verified circuit by circuit with supply held at 8–12 °C at the mould, and the readings recorded as a commissioning baseline
- Parison programme reviewed alongside the cooling work so material distribution and cooling load were addressed together
- Process settings stored on the PLC, which monitors 40+ parameters in real time, so the improved cycle could be recalled rather than rediscovered
- Warping resolved by evening out cooling rather than by cooling everything harder
- Cycle time brought back towards specification because the controlling section was finally being cooled at the rate it needed
- Condensation eliminated by returning the supply temperature to a sensible level and using flow and circuit balance to do the work
- Repeatable start-up, because the corrected settings were stored as machine parameters instead of living in an operator’s notebook
Scenario based on a Sailwin customer project; site-specific figures available on request during engineering review.
5. Water Temperature, Flow Rate and Turbulence
Most plants treat cooling water temperature as the cooling lever. It is the least effective one. Lowering the supply temperature increases the temperature difference between the water and the steel, which does increase heat transfer — but it also increases the risk of condensation on the mould face, can produce a cold skin that traps heat in the section behind it, and does nothing at all if the water is not flowing fast enough to carry the heat away.
Flow rate is the better lever. In a cooling channel, water in turbulent flow mixes continuously and keeps the hot boundary layer thin, so heat transfers from the steel into the water much more effectively than in laminar flow. Increasing flow within a correctly sized circuit therefore raises cooling capacity substantially, whereas reducing temperature has a linear and comparatively modest effect. Sailwin lines are specified around chilled water at 8–12 °C at the mould; the design question is how that temperature is delivered at every circuit, and at what flow.
This is also why return temperature tells you more than supply temperature on a running line. A circuit with a small difference between supply and return is moving little heat, which usually means low flow rather than a cool part. A circuit with a large difference is doing work and may be the circuit that is setting the cycle. Read the pair together and the mould starts to explain its own bottlenecks.
| Adjustment | Effect on cooling | Side effect to watch |
|---|---|---|
| Lower supply temperature | Moderate, roughly proportional to the increase in temperature difference | Condensation on the mould face; a cold skin that slows cooling of the section behind it; higher chiller load |
| Higher flow rate | Large, because turbulent flow removes the insulating boundary layer that laminar flow creates | Pumping energy and pressure drop; circuits must be sized for the flow you intend to run |
| Channel closer to the cavity wall | Large, because heat has less steel to cross before reaching the water | A mould design decision — it cannot be changed after the steel is machined without compromising the cavity |
| Bubbler at a deep core | Large and localised, converting an uncooled feature into a cooled one | Bubbler tube clearance and alignment must be correct, or flow short-circuits and the tip stays hot |
| Parallel manifold instead of series | Even cooling across the part, which usually improves dimensional consistency as well as cycle time | Requires balancing; an unbalanced parallel set can starve one circuit, so flows must be measured |
6. Cooling in the Context of the Whole Machine
Cooling does not exist in isolation from the rest of the line, and two machine-side effects are worth quantifying because they change the plant’s utility balance. Sailwin servo-driven machines reduce power consumption by up to 30%, and high-pressure exhaust recovery reduces compressor load by roughly 20%. On a line where compressed air is the largest utility cost, that recovery figure can be more significant than a marginal cooling improvement — which is why cooling and utility planning should be reviewed together rather than by different departments.
Temperature control elsewhere on the machine also matters. Sailwin machines hold the heating zone within ±1 °C using PID control, and the parison arrives at the mould at a consistent temperature as a result. A parison temperature that varies also varies the heat the mould has to remove, so cooling stability starts upstream. Machines process PE, PP, ABS, EVA, PC and PA, and where a barrier is required, multilayer structures using PE with PA and EVOH are used; both the resin and the structure affect the heat load the mould must handle.
Mechanically, Sailwin extrusion blow moulding machines use Siemens or Mitsubishi PLC control, FESTO combined blowing valves, SMC cylinders and Schneider electrical components, and complete a mould change in under 30 minutes. Standard lead time is 30–45 days, extending to 45–60 days for custom configurations, with a full-load factory acceptance test before shipment and on-site installation and commissioning in 3–7 days. Common wear parts are dispatched within 48 hours, remote support is available 7×24, and machines carry a 2-year whole-machine warranty with CE marking and production under ISO 9001:2015.
The practical conclusion is that cooling is a design conversation, not a maintenance adjustment. If you are specifying a mould, bring the section thickness map to the discussion and ask how the thick sections will be cooled. If you already have a mould and the cycle is too long, measure per-circuit flow and return temperature before touching the water temperature — the answer is usually in the circuit layout rather than in the chiller.
Frequently Asked Questions
Design the Cooling Before You Cut the Mould
Send your bottle drawing, container sample or target output. Our engineering team replies with a machine recommendation, mould assessment and factory-direct quotation within 24 hours.
Related Reading:
• Extrusion Blow Molding Machines — 0.5 L to 1,000 L
• Extrusion Blow Molding Cycle Time Explained
• Clamping Force in Extrusion Blow Molding
• Parison Swell and Die Gap Optimization
• HDPE Jerrycan Blow Molding Machines
• EBM vs ISBM: Bottle Manufacturing Guide
