Five material mistakes that make plastic parts fail early

Almost every plastic part failure material selection problem we see traces back to the same five errors, and none of them are exotic. They are the ordinary shortcuts taken in a specification meeting: a temperature read off a datasheet, a chemical chart consulted for ten seconds, a load assumed to be safe because the stress is low, a dimension that moved for reasons nobody checked, and a wear rate judged by how far something slid.

Each error has a signature on site. Learn the signatures and you can usually name the cause by looking at the failed part rather than by sending it away for analysis. What follows is how each mistake presents, and the question that would have caught it before the part was made.

1. Continuous temperature mistaken for peak temperature

Datasheets quote a continuous service temperature and a short-term or peak figure, and the two can differ by more than a hundred degrees. Unfilled PEEK holds about 250 °C continuously and withstands far more for short excursions; PPS sits lower. If a process runs at 260 °C for two minutes every cycle, the part is not living at its continuous rating, it is living at its peak rating, and the two have different consequences for creep and ageing.

The check is simple: write down the temperature the part sees for the majority of its life, and separately the highest excursion and its duration. Then match each against the correct rating. Our temperature classes note explains how the industry brackets these figures so you compare like with like.

2. Chemical compatibility read off a one-page chart

A compatibility chart answers one question: does this polymer resist this chemical at roughly room temperature and zero stress? Real service adds concentration, temperature and, crucially, mechanical stress. A material that shrugs at a dilute acid at 20 °C can craze in the same acid at 60 °C while under a bolt load, because stress accelerates environmental attack.

So the second failure is reading the chart and stopping. The correct approach is to identify the worst combination of medium, concentration and temperature the part will meet, then check that the polymer is rated there and not merely at ambient. Where the medium is aggressive, PTFE sets the benchmark for inertness, but even it deforms under load, which is the next mistake.

3. Creep ignored under sustained load

Creep is slow deformation under a load that never goes away. A bracket that carries a constant weight can deflect for months even though the stress is far below the material’s yield point, because polymers are viscoelastic. This is where plastic part failure material selection goes wrong most quietly: the part passes a short bench test, ships, and sags six months later.

The relevant figure is the long-term or creep modulus at the service temperature, not the short-term modulus on the front of the datasheet. A rule of thumb is that sustained-load stiffness can be a fraction of the quoted short-term value, and the fraction shrinks as temperature rises. Design the support spacing around that lower number, and add ribs where the geometry allows.

4. Moisture uptake mistaken for wear

Nylons take up water. PA66 absorbs roughly 2.5 to 3% by weight when saturated, which translates into a dimensional change large enough to close a designed clearance. POM takes up far less, typically 0.2 to 0.3%, which is why it holds size so well in wet or humid conditions. When a nylon bushing tightens in service, the culprit is often moisture rather than the wear the operator assumes.

The check is to work from the saturated figure, not the dry-moulded figure, for any part that will sit in humid air or see washdown. Compare candidates across the material families using their moisture uptake, and where the geometry is tight, favour the polymer that moves least.

5. Wear counted in sliding distance, not in load and speed

Wear is not a function of how far a part slides. It is a function of pressure times velocity, the PV product, and the pair of materials in contact. A strip that lasts years in a lightly loaded guide can vanish in weeks when the same slide carries more load at higher speed, because the interface heats up and the wear mechanism changes from mild to severe.

The check is to state the real contact pressure and the real sliding speed, then compare the pair against the material’s PV limit at the working temperature, remembering that a rough counterface or a dry start can double the effective severity. If the numbers are tight, consider a filled or lubricated grade rather than simply a thicker strip. A glass-filled grade changes the wear pair itself, which is often the real fix.

Questions buyers ask about plastic part failure

Can one material avoid all five mistakes at once?

Rarely, and never cheaply. PEEK is strong across temperature, creep and chemicals but costs far more than POM or PA66. The skill is matching the material to the two or three loads that actually dominate, then accepting a compromise on the rest.

How do I tell creep apart from wear on a failed part?

Creep shows as smooth, permanent deformation with no material removed and no dust. Wear removes material, leaves a track and usually produces debris. If the surface is intact but the shape has changed, you are looking at creep.

Do I need a full test programme before choosing a grade?

Not for most parts. A short immersion test, a sustained-load check at temperature and a PV estimate will catch the common failures. The frequently asked questions cover the handful of cases where a formal programme is worth the cost.

What if two mistakes point to different materials?

Then the geometry or the process usually has to change, not just the grade. Talk the duty through with a supplier who machines the material daily, because a short conversation is cheaper than a second failed batch.