A recurring pattern in moulded-part failure is a part that is brittle for no reason the paperwork can explain. The resin was the specified grade, its certificate of analysis is unremarkable, the tool is correct and the geometry checks out. What the certificate describes, however, is the resin as delivered — not the material as it existed at the moment it entered the cavity. Between the hopper and the tool the polymer is heated, sheared and held, and each of those can reduce molecular weight in ways that never appear upstream and are fully present in the finished part.
Molecular weight is what toughness rests on
The mechanical properties that matter most in a cracking failure — impact resistance and resistance to slow crack growth — depend strongly on molecular weight, and they fall off faster than stiffness or nominal tensile strength as chains are shortened. A modest reduction in molecular weight can therefore leave a part that still passes a dimensional and tensile check while being markedly more brittle than intended.
Hydrolysis at the hopper
Condensation polymers — polyesters, polycarbonate, polyamides — cleave when processed with moisture present, because water participates directly in the reverse of the reaction that formed them. The damage happens in the barrel, in seconds, and is irreversible. Drying is therefore a controlled process with a specification, not a precaution, and inadequate drying is one of the more common causes of an inexplicably brittle part. Dryer records and dew-point data are the relevant evidence.
Residence time compounds temperature
Degradation depends on how hot the melt was and how long it stayed there. A machine substantially oversized for the shot holds material in the barrel across many cycles, so a modest melt temperature combined with a long residence time can degrade a resin more than a brief exposure to a higher one. Interruptions matter for the same reason: material sitting in a hot barrel through a stoppage is degraded, and the parts made immediately after restart carry it.
Regrind accumulates history
Reprocessed material has been through a full thermal cycle already, and material reground repeatedly has been through several. Each pass reduces molecular weight further and, in filled compounds, breaks reinforcing fibres shorter. A specified regrind ratio assumes single-pass material of known origin; regrind of uncontrolled provenance and unknown generation count is a materially different input. Sourcing and generation, not just percentage, are what the record needs to establish.
Contamination and cross-mixing
Small quantities of an incompatible polymer introduced through shared regrind equipment, purge material or a hopper changeover can embrittle a compound out of proportion to the amount present, because the second phase acts as a population of internal defects. Spectroscopic analysis identifies foreign polymer, and the finding is usually decisive because there is no legitimate explanation for its presence.
Detecting it in the finished part
Molecular weight measurement on material taken from the part, compared against virgin resin of the same grade, quantifies degradation directly. Spectroscopy identifies oxidation products and foreign species. Thermal analysis shows shifts in melting behaviour and crystallinity consistent with shortened chains, and — for the additive question — measures how much stabiliser capacity remains. Ash or burn-off testing checks filler content against specification, which also detects excessive regrind in filled compounds.
Visual and process tells
Degradation frequently leaves marks before any testing: discolouration or yellowing, brown streaking, silver streaking from moisture flashing to steam, and visible bubbles or splay. These are not proof, since some can arise from other causes, but their presence in a part that later cracked justifies going straight to material analysis rather than treating the failure as purely mechanical.
Why this often looks like a supplier dispute and is not one
Because the incoming resin met specification, the initial assumption is frequently that the material was adequate and the design or the user is at fault. Establishing that the polymer in the finished part is measurably different from the polymer that was delivered relocates the question to the moulding operation. That is a specific, evidenced finding rather than a general assertion about quality, and it depends on having virgin reference material of the same grade to measure against.
What to preserve
Failed parts and unused parts from the same lot. A sample of virgin resin of the specified grade as the reference. The process record covering drying conditions and dew point, melt and barrel temperatures, cycle and any stoppages, shot size against machine capacity, and regrind percentage with its source and generation. These records are routinely overwritten within weeks, which makes the request urgent rather than routine.
This article is general technical orientation, not a failure analysis, an engineering opinion, or advice on any specific matter. Determining the cause of a particular incident requires hands-on examination by a credentialed expert.