
Where the cycle time actually goes
People new to thermoforming assume the forming step is the slow part. It is not. Forming takes two to ten seconds. Cooling takes two to four times as long, and it is the only part of the cycle that cannot be shortened by doing it faster.
| Phase | Share of cycle | Typical time (2 mm sheet) | Can it be shortened? |
|---|---|---|---|
| Load and clamp | 5 – 10% | 2 – 4 s | Automatización |
| Calefacción | 15 – 25% | 8 – 15 s | Better heaters, closer spacing |
| Forming (vacuum + draw) | 3 – 8% | 2 – 5 s | Faster vacuum, more vent area |
| Cooling in the tool | 50 – 70% | 15 – 40 s | Mould temperature, air assist |
| Demould and unload | 5 – 10% | 2 – 5 s | Automatización |
Table 1 — Cycle breakdown for thin gauge. On heavy gauge the cooling share climbs towards 70%: heat enters the sheet quickly but has to leave it through the thickness.
Cooling is the constraint because the part has to be stiff enough to demould without distorting. That is a material property, and the time to reach it is set by heat transfer — which follows a relationship you can calculate.
The relationship that governs everything
Cooling time in thermoforming follows the same square law as injection moulding: it scales with the square of the thickness. Double the sheet thickness and the cooling time roughly quadruples.
For a sheet cooled from both faces, the time to reach a given average temperature is approximately:
t ≈ (s² / (π² · α)) · ln( (8/π²) · (Tsheet − Tmould) / (Tdemould − Tmould) )
Where s is thickness, α is thermal diffusivity (about 0.10 mm²/s for most plastics), and the temperatures are sheet, mould and demould.
| Sheet thickness | Theoretical cooling | Realistic with air assist | Realistic without |
|---|---|---|---|
| 0.5 mm | 0.6 s | 2 – 3 s | 4 – 6 s |
| 1.0 mm | 1.4 s | 4 – 6 s | 8 – 12 s |
| 2.0 mm | 3.6 s | 10 – 15 s | 20 – 28 s |
| 3.0 mm | 7.0 s | 18 – 26 s | 35 – 45 s |
| 4.0 mm | 12 s | 28 – 40 s | 55 – 70 s |
| 6.0 mm | 26 s | 55 – 80 s | 110 – 140 s |
Table 2 — Cooling time by sheet thickness. The gap between theory and practice is the tool interface: air gaps, contact quality and mould temperature control.
The gap between the theoretical and realistic numbers is where the money is. A 2 mm part that calculates at 3.6 s and runs at 25 s has 20 seconds of avoidable time in it — usually from poor contact between sheet and mould, or from a mould that is running too hot.
The four levers you actually have
| Lever | Effect | Practical limit |
|---|---|---|
| Mould temperature | Lower mould = faster cooling | Below about 40 °C causes chill marks and stress |
| Air assist (blown air on the part) | Adds convective cooling on the free face | Needs a nozzle design that does not distort the part |
| Contact quality | Vacuum holds the sheet against the tool | Limited by part geometry and vent area |
| Demould temperature | Eject sooner, at higher temperature | Part must be stiff enough not to distort |
Table 3 — The levers. Air assist is the biggest single gain available on an existing tool, and the cheapest to retrofit.
Air assist works by blowing ambient or chilled air onto the exposed face of the part while it is still on the tool. It roughly doubles the heat transfer coefficient on that face, which cuts cooling time by 25–40%. A simple manifold of nozzles is enough; it does not need to be a precision system.
The limit is distortion. Blowing air on a part that is still soft will push it out of shape, particularly on thin walls. The usual approach is to delay the air until the part has set enough to hold its shape, then blow hard.
Mould temperature: the quality constraint
| Mould temperature | Tiempo de enfriamiento | Part quality |
|---|---|---|
| 20 – 30 °C (cold water) | Fastest | Chill marks, high stress, poor detail, possible distortion |
| 40 – 60 °C | Moderado | Good balance for thin gauge |
| 70 – 90 °C | Lento | Best surface detail, lowest stress |
| 100 – 120 °C | Slowest | Needed for PC and heavy gauge; texture reproduction |
Table 4 — Mould temperature trade-off. Running a cold mould to save cycle time costs surface quality and dimensional stability, which usually costs more downstream.
Cold moulds are the most common false economy in thermoforming. A mould at 25 °C will cut 30% off the cooling time and produce parts with chill marks, high residual stress and a surface that will not hold a texture. The customer sees the surface; nobody sees the cycle time.
Heavy gauge: where the cycle becomes a different problem
Above about 4 mm the cooling time dominates so completely that the process economics change. A 6 mm part might have a 90-second cooling time against a 30-second heating time, which means the machine is idle for two-thirds of the cycle unless it is designed for it.
| Acérquese a | How it works | Effect |
|---|---|---|
| Accept the cycle | Run slow, one part at a time | Simple, but low output |
| Multiple tools on a carousel | Several tools cycle through one heating station | Machine runs continuously; capital cost |
| Twin-sheet or shuttle | Two stations alternate | Hides cooling behind heating |
| Cooling fixtures | Part moves to a separate cooling jig | Frees the tool for the next part |
Table 5 — Handling long cooling times on heavy gauge. The carousel and the cooling fixture are the two that actually work; the rest just accept the loss.
The cooling fixture is the cheapest of these. The part comes off the tool while still warm and slightly flexible, is transferred to a shaped jig that holds it while it cools, and the tool is free for the next cycle. The jig costs a fraction of a second tool and it can cut effective cycle time by 40%.
Measuring and improving
| Measurement | How | Target |
|---|---|---|
| Part surface temperature at demould | IR gun or contact probe | Below the resin's demould temperature |
| Mould surface temperature | Probe at three points | Uniform within 5 °C |
| Cooling water ΔT | Inlet vs outlet | Under 3 °C |
| Cycle time by phase | Stopwatch or machine log | Cooling under 65% of total |
| Distortion after 24 h | Measure key dimensions | Within tolerance |
Table 6 — What to measure to improve the cycle. The water ΔT is the fastest diagnostic: a high ΔT means the mould is not transferring heat as designed.
Part temperature at demould is the number that tells you whether the cooling time is right. If the part comes off at 70 °C when the resin needs to be below 60 °C, it will distort after demould and the cycle is being pushed too hard. If it comes off at 45 °C, there is cooling time to give back.
How much of a vacuum forming cycle is cooling?
45–55% on thin gauge and around 70% on heavy gauge. It is the single largest phase and the only one that cannot be shortened by running faster — only by better heat transfer or a cooler mould.
Why does cooling time quadruple when thickness doubles?
Because heat has to conduct through the thickness, and conduction distance enters the equation squared. It is the same relationship as injection moulding, and it is why thick-gauge parts have such long cycles.
Does air assist really help?
Yes, 25–40% reduction in cooling time. Blowing air on the free face roughly doubles the heat transfer coefficient there. The constraint is distortion: the air has to be delayed until the part is stiff enough not to be pushed out of shape.
Can I run a colder mould to speed up the cycle?
You can, and you will get chill marks, high residual stress, poor surface detail and dimensional instability. Below about 40 °C the quality loss usually costs more than the cycle time saves. The exception is very thin, non-cosmetic parts.
How do I know when to demould?
Measure the part surface temperature. Every resin has a demould temperature below which it is stiff enough — typically 60–70 °C for PS and ABS, 80–90 °C for PP, 100–120 °C for PC. If it comes off hotter, it will distort.
What is the biggest cycle time mistake?
Setting cooling time by trial and error and then never revisiting it. Most machines run with 20–30% more cooling time than necessary because the setting was made once for a difficult part and never adjusted for easier ones.
Does the sheet thickness tolerance affect cycle time?
Yes. A sheet at the top of its thickness tolerance needs noticeably more cooling than one at the bottom, and if the cycle is set for the thin end, thick sheets will distort. Setting the cycle for the thickest sheet in the batch is the safe choice.
Is there a way to cool both sides?
On female tools with a male plug, the plug can be temperature controlled, which cools the inside. On simple vacuum forming there is only one tool face, which is why air assist exists — it is the only way to cool the free side.
Cycle time longer than it should be?
Send us the part drawing, material, gauge and current cycle breakdown. We will calculate what the cooling time should be and tell you whether the limit is the mould temperature, the contact, or the demould temperature.











