Is there a non-destructive test to measure stress levels for plastic parts?

I have used this on a number of important projects, including within the last three months. Its value cannot be overstated in troubleshooting of problem parts, with the output being instantly and clearly understood by technical gurus and novices alike. One can even interrogate the images retrospectively with the accompanying software, to check temperatures in any location. No other method can give you all temperatures at all parts of a freshly ejected component at a single point in time. Though the method will not highlight internal stresses imparted during the mould filling process, it certainly does give dramatic visual evidence of uneven part surface temperatures that lead to distortion upon cooling. I have also seen very strong correlation between actual IR color plots and mould cooling simulations, so it can give those funding projects confidence to use predictive tools.

It's crucially important to note the relationship between process-induced stresses and warpage/distortion. A geometrically acceptable part, with no obvious distortion from design intent, can be highly stressed through its wall thickness, if those stresses are evenly balanced. Think of pre-stressed concrete beams as an example: the re-bars are under huge tension, while the actual concrete is in compression and the loads are in equilibrium. Think also of those toys that collapse when the base is pressed and the internal tension member (string!) is loosened. Warpage occurs when stresses in some areas are not balanced and the stronger ones "win," so to speak, as the part cools. The part shape reaches a state of equilibrium only when the stresses balance, in the distorted shape, when cool. Counter-intuitively, you may actually get lower stress levels through the section in an obviously distorted part. Minimize both flow-induced and cooling-induced stresses AND warpage by:

• Following good component design practice on wall thickness relationships and radii;
• Simulation, simulation, simulation (to paraphrase a recent UK Prime Minister); i.e., FEA if necessary to keep load-induced stresses reasonable; flow simulation and simulation of the proposed cooling circuits (with design reiterations to even out shear stress and part surface temperature-upon-ejection variations); and
• Using the same material and moulding conditions as used during the simulation (or as near as is practical).

I concentrate here on injection moulding, because it’s the process that’s been subject to most simulation.