Contamination control: keeping the wrong material off and out of the part

Cleanliness has to be specified before it can be controlled

A drawing note saying the part must be free from contamination is unenforceable and is routinely argued about after a rejection. A usable specification names what matters: the type of contaminant of concern, the size above which particles are unacceptable, where on the part it applies, and the extraction and measurement method to be used, since results depend heavily on how hard you wash and what you filter. Deciding this requires the design function to say why cleanliness matters — blockage, wear, adhesion failure, electrical short, biological risk — because that determines which contaminant actually needs limiting.

Cross-contamination between products is a changeover problem

Wherever equipment is shared, residue of the previous run can carry into the next: material in a hopper or feed line, powder in a mixer, colourant in a barrel, an allergen or an active ingredient in a vessel, an oil in a wash bath. The controls are procedural and physical together — a defined line clearance with a second person verifying, dedicated tooling for high-risk materials, sequencing production so the sensitive product runs first after a clean, and verification of the clean itself rather than assumption. The verification is the step most often dropped when the schedule is tight.

Contamination generated by your own process

The largest sources are usually internal. Chips and burrs from machining migrate into blind holes and stay there. Blasting media lodges in threads. Wear debris comes off tooling, conveyors and gloves. Packaging sheds fibres and adhesive. Cleaning agents leave residue when concentration or rinse is wrong, and compressed air lines carry oil and water into parts that were clean until somebody blew them down. Mapping where the contaminant physically comes from usually points at a process change — deburring earlier, a different fixture orientation, filtration on the air line — rather than at more washing.

Cleaning is a process with parameters, not an activity

Washing that is not controlled produces inconsistent results and false confidence. The parameters that matter are the chemistry and its concentration, temperature, time, mechanical action, the condition of the bath, the rinse, and the drying. Baths degrade as they load with soil, so a wash that worked in the morning may not by the end of the shift, which makes bath monitoring and change frequency a real control rather than housekeeping. Where cleanliness is a designated characteristic, the wash process is a candidate for validation and its parameters belong in the control plan.

Handling, packing and everything after the clean

Parts are frequently cleaned properly and then recontaminated before dispatch. Bare hands transfer oils and salts, open containers collect airborne debris, dirty racks reintroduce what was removed, and internally generated dust settles on anything left uncovered overnight. Controlling the interval between cleaning and protection is often more effective than improving the clean itself: package immediately, use packaging chosen for shedding behaviour rather than for cost, keep the protected route separate from the general shop floor, and treat any resealed container as an event requiring a decision about its status. Where a customer receives parts that were clean when packed, the argument about who contaminated them is unwinnable without a record of the packing condition, which makes photographing a sealed container at dispatch a cheap piece of evidence.

Frequently asked questions

How do we test whether a part is clean enough?

By extracting the contaminant under a defined method and measuring what comes off — typically rinsing or ultrasonically agitating the part in a controlled fluid, filtering the result, then weighing the residue or counting and sizing particles under magnification. The method must be fixed and repeatable, because a more aggressive extraction always finds more. Blank runs matter too, since the fluid, the bench and the operator all contribute contamination that would otherwise be attributed to the part.

Does contamination control mean we need a cleanroom?

Not usually. Controlled environments are one tool among several and they are expensive to build and operate, so they suit products where airborne particulate or biological control genuinely drives the requirement. Many contamination problems are solved on a normal shop floor by changing where debris is generated, adding filtration, improving the wash process, and protecting parts immediately after cleaning. Treat the environment as a facility question only after the process sources have been addressed.

Who should own contamination limits — design or manufacturing?

Design owns the requirement, because only they know what the contaminant does to function, and manufacturing owns the method that achieves it. Problems arise when design writes an aspirational limit without a measurement method, leaving the plant to interpret it, or when manufacturing sets a limit based on what the current wash achieves, which makes the specification a description of present capability rather than of what the product needs.