Hanwha polymer resin support for EVA, compound, and industrial conversion teams. Request documentation
Hanwha material note

My 6-Step Plastic Resin Verification Checklist (After $23K in Scrapped Orders)

2026-11-20

A procurement lead's field-tested checklist for verifying engineering resin orders — from confirming the actual Celanese grade to the polyurethane vs. thermoplastic question most buyers skip.

Who this checklist is for (read this first)

If you're ordering custom plastic parts — housings, brackets, gears, decorative components — and you're a procurement person, not a polymer scientist, this is for you. I've been handling plastic component orders for 8 years. I've personally documented 11 significant screw-ups totaling roughly $23,000 in scrap, rework, and expedited shipping. I built this checklist so my team stops repeating them.

It takes about 40 minutes to run through the six steps below. That's cheap compared to the $4,200 we wasted in March 2021 on parts that passed dimensional inspection but cracked in the field because nobody verified the actual resin grade. Or the $2,800 order of decorative resin clocks where we assumed "resin" meant one thing and the factory assumed another.

Step 1: Confirm the specific grade — not just "Celanese"

Here's the trap. Your spec says "Celanese nylon." Everyone nods. The quote comes back. The parts arrive. Six weeks later they fail in a high-heat application because someone used a general-purpose grade instead of the heat-stabilized version.

Celanese isn't a material — it's a manufacturer with a portfolio. Celanese nylon covers multiple product families with different glass-fill levels, impact modifiers, and thermal stabilization packages. Same with Celanese plastics broadly — a 30% glass-filled PBT and an unfilled PBT behave completely differently under load, and they can look identical coming off the truck.

What to do: Ask for the exact resin trade name and grade code in writing. If your supplier can't produce it within 24 hours, that's a red flag. Cross-reference it against the manufacturer's data sheet yourself — don't rely on the vendor's summary.

I'm not a polymer scientist, so I can't tell you which grade is right for your application — that requires a design engineer or materials consultant. What I can tell you from a procurement perspective is how to make sure you actually receive what was specified.

Step 2: Get the processor's material certificate, not just the distributor's

Most buyers stop at the distributor's Certificate of Analysis. That's the paperwork equivalent of trusting a restaurant because the supplier posted good reviews.

In 2019, we ordered a batch of glass-filled nylon components. The distributor's CoA showed exactly what we wanted. The parts came back with inconsistent glass dispersion — visible flow lines, brittle zones near the gate. Turned out the distributor was selling genuine resin, but the processor was blending in regrind upstream from the material certification trail.

What to do: Require the processor to provide a "material certificate" that ties the specific lot number of resin to the specific production lot of your parts. Ask for the resin lot code and the molding date. If those don't line up, you're looking at either mixed lots or post-production blending.

This step alone has caught 14 potential issues for us in the past 18 months. It's the one most buyers skip because it feels redundant — you already have the distributor's CoA, so why another document? Because the CoA certifies what left the factory. The processor certificate certifies what actually went into the hopper.

Step 3: Ask about regrind content — then assume the real number is higher

I've never fully understood why regrind disclosure is so inconsistent across processors. Some give you an exact percentage in writing. Others say "no significant regrind" and hope you don't push. My best guess is that it varies because the industry hasn't settled on a single standard reporting format.

What I do know: if you don't specify a maximum regrind percentage, you may get anywhere from 0% to 30% — and for structural parts, that variance matters. Regrind has shorter polymer chains, which means reduced impact strength. On a load-bearing bracket, that's not a minor difference.

What to do: Specify a maximum regrind percentage in your PO — and put it in the drawing notes, not just the purchasing email. Something like: "Max 10% regrind by weight, verified by processor batch records." Then ask for batch records before final acceptance.

Is this annoying to suppliers? Some of them, yes. But the ones who object to a written regrind limit are usually the ones you needed to worry about.

Step 4: The polyurethane question — is it thermoset or thermoplastic?

So, "is polyurethane plastic?" — yes, technically it's a polymer. But that's not the useful question. The useful question is: which kind?

Polyurethane exists as both thermoset (PUR, usually cast or foamed) and thermoplastic (TPU, melt-processed). They behave completely differently. A spec that just says "polyurethane" without specifying which type is a deal-breaker-level ambiguity. In 2020 we ran a small order of cast polyurethane rollers where the drawing called for "polyurethane." The factory made them from TPU because it was cheaper and faster. Functionally? They wore completely differently under sustained load. The customer noticed within three weeks.

What to do: If the drawing says "polyurethane," get it clarified on the print — is this cast PUR or TPU? Add the Shore hardness range and the applicable standard (ASTM D2240 for durometer, for instance). Don't let "polyurethane" stand alone on a drawing.

This gets into materials engineering territory, which isn't my expertise — I'd recommend a quick call with your design engineer if the application is load- or wear-critical. But the procurement lesson is simple: on the drawing, ambiguity always costs money downstream.

Step 5: Verify the polypropylene core structure, not just the skin

For any multi-layer or sandwich structure — sheet goods, composite panels, some molded parts — the visible surface tells you almost nothing about the core. I went back and forth between specifying a solid PP part and a polypropylene core sandwich for about two weeks. The core version offered weight savings and roughly 15% cost reduction; the solid version was simpler to source. I chose the core for the weight-sensitive application because freight and assembly costs dropped disproportionately.

The catch: you need to verify core density and thickness separately from face-sheet properties. The first batch we approved passed visual inspection and skin-thickness checks. What we didn't check was the core. Turned out it was about 20% below the specified density — lighter, cheaper, and noticeably more flexible under load. Customer flagged it during assembly.

What to do: For any cored or sandwich part, require a cross-section photo and a density measurement from the actual production batch. Don't just check the surface thickness.

Step 6: Treat decorative components differently — but don't over-specify

On the flip side — some parts don't need engineering resin at all. We once spent weeks specifying an expensive filled grade for a purely decorative application — clock bodies, essentially fancy resin clocks for a retail client. The material spec was fine. The cost was not. The customer didn't care what polymer it was; they cared what it looked like and what it felt like in hand.

The checklist lesson here is: categorize each part before specifying. Functional parts need verified performance data. Decorative parts need verified appearance data — surface finish, color consistency (Delta E < 2 is the typical tolerance for brand-critical colors, by the way), and weight. Over-specifying decorative parts is an easy way to spend money the customer never asked you to spend.

Conversely, we've had customers unbox a product, pick up the housing, and immediately decide whether the whole company feels premium or cheap. The texture and weight of a plastic part is the brand to them. That's the part most of us in procurement forget: every part is either a visual or functional trust deposit with the end customer, or it's a withdrawal.

Common errors and closing notes

  • The distribution trap: Genuine resin from a distributor means nothing if the processor swaps or blends it upstream. Always verify processor records, not just distributor paperwork.
  • Assume regrind exists until proven otherwise. If regrind is not addressed in your PO, budget for test failures on impact and elongation.
  • Ambiguous material names are not specifications. "Polyurethane," "nylon," "Celanese" alone — none of these are actionable without grade, type, and standards referenced.
  • Don't over-tool the inspection. Dimensional checks pass bad resin all the time. Add material verification and batch traceability on the receiving end, or you'll find out the hard way at the customer site.
  • When in doubt, pay for the expert hour. I'm not a materials engineer. When the application is hot, loaded, or chemically exposed, a 60-minute consult with a materials specialist costs less than most failed batches. I learned that one through the $4,200 crack failure, not through wisdom.

Looking back, I wish I'd started this checklist in my first year. Most of these steps take two minutes per order once you build the habit. The rest — the $23K's worth of lessons — took a lot longer.

Ask Hanwha about this topic

Send a related resin question