Millipore Cleanliness Testing: What the Numbers Actually Mean

What is measured

Residue mass on a membrane filter

Reported as

mg per part or per m²

Often specified with

Particle-size class limits

Common failure cause

Recontamination after washing

Process requirement

Filtration + DI rinse + clean handling

Automotive, bearing, hydraulic and injector customers increasingly ship against a cleanliness specification rather than a visual standard. The test is usually called Millipore, after the membrane filter it uses, and it produces a number that is unfamiliar enough to be misread in both directions.

How the test works

1. 1The component is washed under controlled, standardised conditions — a defined fluid, volume, time and agitation method — to extract whatever contamination is on it.
2. 2The extraction fluid is drawn through a membrane filter of known, pre-weighed mass.
3. 3The filter is dried and re-weighed. The difference is the gravimetric residue.
4. 4The result is reported as milligrams per part, or normalised to milligrams per square metre of surface.
5. 5For most automotive specifications the membrane is then examined optically and particles are binned by size class.

The extraction step is where comparability lives or dies. Two labs using different agitation methods on the same part will get different numbers, which is why the specification names the method as tightly as it names the limit.

Why mass alone is not enough

A gram of fine dust and a single 500 µm hard steel particle can weigh the same and mean completely different things. The dust washes through a hydraulic circuit harmlessly; the particle scores a bore, blocks an orifice or destroys a bearing raceway.

That is why modern specifications add size-class limits — for example, no metallic particles above 600 µm — and often require classification into metallic, non-metallic and fibre. The classification is diagnostically useful: metallic swarf points at machining and deburring, fibres point at wipes, gloves and packaging.

Why good baths still fail the test

This is the part that surprises people. A part can leave the wash genuinely clean and fail the test, because most cleanliness failures happen downstream of the bath.

Unfiltered rinse water puts particles back on. Dirty baskets transfer contamination from the previous load. Packaging sheds fibres. Bare-hand handling adds skin oil and lint. Compressed air used for drying carries oil and pipe scale unless it is properly filtered. Every one of these is invisible on a shop floor and obvious on a membrane.

What the wash process has to provide

Chemistry is necessary but not sufficient. A cleanliness line needs a low-foam concentrate — because the geometry that holds contamination is exactly the geometry only ultrasonic or agitated immersion reaches, and neither tolerates foam. It needs continuous filtration on the wash so removed particles do not come back. It needs a counter-flow rinse with a deionised final stage below about 10 µS/cm, so nothing is left behind as the part dries.

And it needs drying that actually reaches blind features, because residual moisture both carries particles and, on steel, starts corrosion that will itself appear as contamination.

A practical route to passing consistently

1. 1Establish a baseline: test parts straight from the wash, and again after normal handling and packing. The gap between those two numbers tells you whether your problem is cleaning or recontamination.
2. 2Fix the larger gap first. In most plants it is the second one.
3. 3Add filtration progressively — wash first, then rinse — and re-test after each change rather than all at once.
4. 4Control drag-out, because dragged solution carries suspended particles into the rinse.
5. 5Only then adjust chemistry, concentration or cycle time.

Frequently asked questions