Spring-Back and Dimensional Control in Thermoformed Parts

Two different things happen after demould

A thermoformed part leaves the tool at the wrong size. Not because the tool is wrong, but because two independent effects act on the part between the forming station and the trimming fixture.

الانكماش is thermal contraction as the part cools from forming temperature to room temperature. It is uniform in direction, proportional to the temperature drop, and predictable from the material's coefficient of thermal expansion.

Spring-back is elastic recovery. While the part is on the tool it is held against the surface by vacuum. When the vacuum is released, the material — which was stretched during forming — relaxes towards its unstretched shape. The result is that radii open up, walls straighten slightly, and the part moves away from the tool surface.

Shrinkage: the predictable part

Table 1 — Thermal shrinkage by material. The amorphous resins are half the value of the crystalline ones, which is why PS and ABS dominate dimensionally critical parts.

Crystalline materials shrink more and shrink directionally, because the crystals form in the direction of stretch. A PP part will be smaller in the machine direction than across it, and the difference can be 0.5% or more. That is why PP is rarely used for parts where two dimensions have to be equally accurate.

The compensation is straightforward: scale the tool by the shrinkage factor. A 1.5% shrinkage means the tool is made 1.5% larger than the nominal part. The complication is that shrinkage depends on the actual cooling history, so the number from a data sheet is a starting point, not a specification.

Spring-back: the shape change

Spring-back happens because the sheet is stretched during forming and the stretch is elastic to some degree. When the vacuum releases, the elastic component recovers.

Table 2 — What spring-back does to features. Radii and wall angles are the most affected and the most visible on an assembled part.

The magnitude depends on how much the material was stretched. A shallow part formed with little stretch springs back very little; a deep draw with high stretch springs back significantly. That is why the same material and the same thickness can produce parts with different spring-back on different tools.

There is also a time element. Some spring-back is immediate (elastic recovery on vacuum release), and some develops over hours as residual stress relaxes. A part measured five minutes after demould and the same part measured a day later will not be identical.

Compensating in the tool

Table 3 — Tool compensation methods. In practice every tool needs a trial-and-correct cycle; the question is how well you can predict the first attempt.

The economics of tool correction are worth understanding. A first tool that is 95% right needs a small revision, which costs a fraction of the tool. A first tool that ignores spring-back entirely can need a re-make. Spending an extra day on the design to predict spring-back is usually cheaper than the revision.

Process levers that change spring-back

Table 4 — Process levers. Temperature is the strongest: a hotter sheet and a hotter tool both reduce spring-back, at the cost of cycle time.

Hotter is not always better. A sheet that is too hot sags before forming and thins excessively at the bottom of the draw, which creates its own dimensional problems. The window is the same one that governs forming quality generally, and spring-back is a secondary consideration within it.

Measuring and predicting

Table 5 — Measurement schedule. The 24-hour dimension is the one that matters; anything measured sooner is optimistic.

Building a correction record is the most valuable thing a thermoformer can do. For each material, thickness and draw depth, record the measured dimension at 24 hours against the tool dimension. After a dozen parts you have a correction table that lets the next tool be right first time.

Design rules that reduce the problem

Table 6 — Design rules. The tolerance allocation rule is the most commercially useful: putting a tight tolerance on a feature that cannot hold it causes most of the arguments.

The draw ratio rule is worth stating as a number. Above about 1:1 (depth equal to width), wall thinning and spring-back both increase sharply, and the part becomes hard to control. If a design needs a deeper draw, pressure forming or a different process is worth considering.