mixing defect containment sorting of custom rubber parts
Approfondissement / Analyse

Mixing Defect: 12,400 Rubber Parts Audited and Fixed

Zhang
octobre 14, 2026

A mixing defect is the failure mode where parts of one part number end up inside the carton of another. It surfaced on one of our cable seal programs last month: an assembly line opened a box of 32A1PH seals and found 32A3PH parts mixed through the batch. The two parts differ by a few millimeters of body diameter — enough to fail a waterproofing test, not enough to catch by eye.

Mixing defects cost far more than their size suggests. One wrong part in a sealed connector can fail an IP test, stop an assembly line, and trigger a recall investigation that costs more than the entire order.

This report covers what happened, how the batch was contained in 72 hours, the two root causes we found, and the controls that have held for five consecutive production runs since.

What Counts as a Mixing Defect in Custom Rubber Parts

A mixing defect is a part-identity problem, not a compound problem. Two similar SKUs share a press, a trim table or a drying tray, and one ends up in the other's box. It is different from dispersion or scorch issues inside the rubber itself.

It is also the defect most likely to reach the customer undetected, because every individual part passes dimensional inspection. Nothing looks wrong until someone tries to assemble it.

Cable seals are the classic case. 32A1PH and 32A3PH share an outer profile and differ only in body diameter by a few millimeters. In a shared bin, that difference reads as normal molding variation, and operators sorting at line speed will not catch it.

mixing defect containment sorting of custom rubber parts

The Case: Two Cartons, Two Part Numbers

The complaint arrived from a customer's assembly line in Europe. Operators had found 32A3PH seals inside batches of IEC16A3PH and 32A1PH components. The customer stopped the line, quarantined the affected stations and sent the lot back.

For a waterproof cable seal, that is the right reaction. A seal with the wrong body diameter compresses unevenly, and the failure only shows up in the IP test — or worse, in the field after 18 months of thermal cycling.

Our account team authorized the return the same day. The quality team quarantined every finished-goods carton that shared a press, a mold change window or a packing station with the affected lot.

72-Hour Containment, Step by Step

Containment follows a fixed sequence. Skipping a step means defective stock stays in the warehouse and ships again next week.

  1. Stop the line and freeze the lot. All shipments of the affected part numbers were held before the next truck left.
  2. Quarantine every related finished good. Not just the returned lot — every carton produced on the same presses during the same window.
  3. Audit 100% of warehouse stock. Both part numbers were inspected piece by piece against drawing dimensions and visual criteria.
  4. Sort and rework, then re-inspect. Mixed parts were separated, and the reworked lots went through a second 100% inspection before release.

The audit numbers from that week:

Location Material code Quantity Result Disposition
Customer returned lot 32A1PH / 32A3PH Full lot Mixed parts confirmed 100% manual sorting and rework
Warehouse stock A 03-0034 7,400 pcs 100% compliant Released after full audit
Warehouse stock B 03-0037 5,000 pcs 100% compliant Released after full audit

Twelve thousand four hundred pieces inspected by hand in three days. That is what a mixing defect costs in labor, before anyone discusses the customer relationship.

Root Cause 1: Shared Containers at the Press

The mixing defect traced back to the demolding station, not to the operator's attention span.

Every compression press fed a common collection container. Operators demolded 32A1PH and 32A3PH into the same bin during a mold change, then carried the bin to trimming. From that point the two part numbers were physically indistinguishable without a caliper.

This is the part most suppliers get wrong when they investigate. They conclude "operator carelessness" and respond with a memo. The real cause is that the process allowed two similar parts to occupy the same physical space — and any process that permits that will produce the defect again, with a different operator.

Root Cause 2: Moisture in the Air Line

The same week, the lot also showed heavy flash that could not be torn off by hand. That was not a coincidence, and it is worth explaining because it is widely misdiagnosed.

Compressed air in a plant carries condensation. When a blow gun sprays moist air onto a hot cavity, water droplets trap rubber debris in the cavity corners and along the parting line. The debris holds the mold faces apart under clamping pressure, rubber seeps into the gap, and the result is thick flash instead of a clean tear line.

So one root cause was spatial (shared containers) and the other was pneumatic (wet air). Both were process conditions, and both were fixable without asking anyone to "be more careful".

The Three Fixes That Held

Corrective actions only work when they are physical or documented. Training alone decays within a quarter.

Action Standard implemented Verification
Eliminate part mixing Three color-coded containers at every compression press, with work-order scanning at demolding Zero mixed parts across 5 consecutive batches
Eliminate moisture flash Water-oil separators and automatic drain valves on all pneumatic lines; daily drain check added to the equipment checklist Clean self-tearing flash restored
Institutionalize the control P-040 quality risk procedure training for all workshop teams; segregation rules written into the Control Plan and SOP 100% compliance on poka-yoke audits

Three details matter more than the list itself.

The containers are color-coded and fixed to the press, not stored in a central rack — an operator cannot grab the wrong one without physically moving a labeled bin.

The drain check is on the daily equipment checklist, initialed per shift. A maintenance task that lives only in a training slide does not happen.

The segregation rule is in the Control Plan, which means it is audited during internal ISO 9001 reviews and referenced by ASTM D2000 material callouts on the drawing.

How We Verified the Fix

Verification is where most 8D reports stop short. "Retrained the operator" is not evidence.

We tracked five consecutive production batches of both part numbers after the corrective actions went live. Criteria: zero mixed parts at final inspection, zero abnormal flash at demolding, and a documented bin-color check at the start of every shift.

All five batches passed. The control has held since, including two mold changeovers — the exact event that produced the original mixing defect.

corrective actions for a mixing defect at the compression press

Six Questions to Ask a Supplier About Mixing Risk

If you buy similar part numbers from one factory, ask these before the first shipment. The answers tell you whether the process can physically produce a mixing defect.

  1. Do two similar part numbers ever share a press or a mold change window? They usually do. The follow-up is what matters.
  2. Where do parts go immediately after demolding? A shared bin is a mixing defect waiting to happen. Fixed, labeled, per-press containers are the control.
  3. How do you separate parts during trimming and deflashing? Batch-level separation by work order, not by operator memory.
  4. What is your containment procedure if a customer reports mixed parts? The answer should include a 100% warehouse audit, not only a replacement shipment.
  5. Do you have automatic drains on your pneumatic lines? If the answer is "we drain manually when we remember", expect flash variability.
  6. Is segregation written into your Control Plan? If it only exists in a training record, it will lapse at the next staff turnover.

Frequently Asked Questions

Why do similar cable seals get mixed so easily?

Because they are designed to look alike. 32A1PH and 32A3PH share an outer profile and differ by a few millimeters in body diameter, which is inside the range of normal molding variation. Sorting at line speed by eye is unreliable; physical container segregation is the only control that holds.

Can a mixing defect be caught by final inspection?

Rarely, and not reliably. Every individual part in a mixed batch passes dimensional inspection — the defect is in the batch composition, not the part. Catching it requires either 100% sorting against a reference sample or a process that never lets the two part numbers meet.

How does moisture in a blow gun cause thick flash?

Water droplets from a wet air line trap rubber debris in cavity corners and the parting line. The debris holds the mold faces apart, rubber seeps into the gap, and the flash becomes too thick to tear cleanly. The fix is a water-oil separator with an automatic drain, not more trimming labor.

What should a factory do when a customer reports mixed parts?

Four steps, in this order: freeze the affected lots, quarantine all related finished goods, audit 100% of warehouse stock, then sort and re-inspect before release. Replacing the shipment without auditing the warehouse only moves the problem to the next order.

Is a mixing defect a sign of a bad factory?

Not by itself. It is a sign that a process allowed two similar parts to occupy the same physical space — a design flaw in the workflow, not necessarily in the people. What separates suppliers is whether the corrective action is physical and documented, or a memo asking operators to pay more attention.

Conclusion

A mixing defect is a system failure that shows up as a single wrong part in a carton. It is cheap to prevent and expensive to discover, and the difference between those two states is almost always container discipline at the press.

The fix in this case was not a new machine. It was three colored bins, two automatic drains, one rewritten SOP, and a 100% audit that proved the warehouse was clean.

If you are sourcing similar part numbers — cable seals, grommets, plugs — and want to see how we control batch identity, send us your drawings for a free DFM review. Our custom rubber parts defect library covers the other failure modes we track, including uncured seals et rough edges on molded parts.


About the Author & Dibo Tech

My name is Kunfa Ye, but friends call me Jeff. I was born in 1986, the year of the Tiger. In 2011, my brother Autumn Ye and I co-founded Dibo Tech, rooting ourselves in the silicone rubber industry for fifteen years. Building from scratch, we navigated industry cycles and learned the true rewards of hard work. We have always built our business on technical expertise and integrity.

The silicone industry touches every corner of modern manufacturing. I deeply believe that behind every product lies trust, and the essence of business is mutual gain. Over these years, my proudest achievement is not personal financial independence, but helping our clients optimize supply chains and boost production efficiency. Their success is far more rewarding than personal milestones. Today, I aim to give back by sharing insights, answering engineering questions, and connecting resources.

Brand Name: Dibo Tech Slogan: Dibo Technology Silicone Manufacturing Specialists Website: https://samxine.com/ Our Mission: Based on strong technical capabilities and quality first, we are committed to innovative and eco-friendly silicone rubber production, providing reliable silicone solutions while fulfilling social responsibilities for a sustainable future.

Share This Article

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *

Chat avec nous sur WhatsApp