Ask ten people how they chose the wiper in their cleanroom and nine will describe a sample someone left on a bench. It worked, nobody complained, and it has been re-ordered ever since. That is not a specification; it is a habit with a part number.
The habit becomes expensive at the worst possible moment — an audit, a deviation, or a customer complaint that traces back to a fibre on a critical surface. The alternative is not complicated. Four parameters, decided in order, will narrow almost any catalogue to two references.
1. Start with the surface, not the product
Write down what is being wiped. Not "equipment" — the actual material. Anodised aluminium, borosilicate glass, polycarbonate, 316L stainless, a coated optic. Each has a different tolerance for abrasion, and abrasion is the failure mode that no certificate will save you from.
Soft coated optics and polished decorative surfaces rule out mesh textures and high-friction blends. Welded stainless and painted steel can take a more aggressive cloth, and often need one: a soft wipe will skate over carbon without removing it.
2. Name the solvent and its concentration
Isopropyl alcohol at 70% is not the same cleaning problem as 99% IPA, and neither behaves like acetone, MEK or a quaternary ammonium disinfectant. The wiper has to hold the liquid without swelling, dissolving or shedding colour.
Two practical rules. First, cellulose-containing cloths are poor partners for strong oxidisers and some solvents; they lose wet strength and leave residue. Second, foam swabs and some polyurethane materials swell in ketones. If the solvent list includes acetone or MEK, specify a polyester or polypropylene reference and ask for compatibility data in writing.
A one-page compatibility table listing the wiper against the solvents you actually use, at the concentration you use them. If the supplier cannot produce it, that is information too.
3. Fix the cleanroom class
ISO 14644-1 gives you the particle concentration limit, but it does not tell you what the wiper sheds. Those are different numbers. A cloth can be packed in an ISO 8 room and still release thousands of fibres per square metre when rubbed.
In practice, the class sets the minimum evidence package. For ISO 8, a cut-edge pulp cloth with basic particle data is usually defensible. For ISO 6 and 7, move to sealed edges and asked-for extractables data. For ISO 5 and cleaner, the wiper should be laundered or manufactured for the class, individually bagged, and supported by lot documentation that names the test method.
| Class | Typical reference | Edge | Evidence to hold |
|---|---|---|---|
| ISO 8 | Wood pulp / PP cloth | Cut | Particle count, material declaration |
| ISO 7 | Polyester or sealed non-woven | Sealed or laser cut | Particle + fibre release, extractables |
| ISO 6 | Knitted polyester, laundered | Sealed | Full IEST-RP-CC004 suite |
| ISO 5 / Grade A | Sterile, individually bagged | Sealed | Sterility assurance, lot traceability, extractables |
4. Decide the disposal route before the pack format
This is the parameter everyone skips. If the wipe leaves the room as hazardous waste, a heavier sheet may reduce the number of pieces per task and therefore the waste volume. If it goes into general waste, the calculation flips: lighter sheets mean less mass to handle.
Pack format follows. A pop-up box is a dispensing convenience, not a transfer pack. Triple-bagged sterile wipes exist because the bag, not the wipe, is the thing being decontaminated at the airlock.
The shortlist test
With four answers written down, the shortlist should be two references: a primary and a fallback. Then run the same three checks on both.
- Wet strength. Soak a sheet in the actual solvent for five minutes and wring it. If it tears at the fold, it will tear on the line.
- Residue. Wipe a clean mirror with the solvent-wet sheet and let it dry under a lamp. A haze ring means extractables you will be cleaning off later.
- Edge behaviour. Rub the edge across a dark glass slide ten times. Visible fibre lines mean the edge is wrong for the class you wrote down.
A worked example
A medical device assembler wipes an anodised housing with 70% IPA in an ISO 7 room, and disposes of wipes as non-hazardous waste. The surface rules out high-friction blends; the solvent permits pulp or polyester; the class requires a sealed edge; the disposal route favours a lighter sheet.
That points to a sealed-edge polyester wipe, pre-wetted if the dose needs to be consistent, dry if the operator is already controlling solvent volume. The fallback is a lighter sealed non-woven for touch-up work. Two references, one decision record, no samples left on a bench.
Where the framework breaks
It breaks in two places. First, when the surface is unknown — a new coating, an unlabelled composite, a supplier change. Second, when the process is regulated in a way that prescribes the consumable, as in aseptic filling.
In both cases the answer is the same: test, do not infer. A five-minute wipe test on a representative coupon costs less than a deviation investigation, and it produces a record you can attach to the specification.
Should the wiper be pre-wetted or dry?
Pre-wetted if the solvent dose must be repeatable or if the operator cannot carry a spray bottle into the area. Dry if the process already controls solvent application and you want to avoid a wet pack with a short open-life.
How many wiper references should a site hold?
Two per cleaning task: a primary and a qualified fallback. More than that and operators choose by habit; fewer and a supply interruption becomes a change control.
Does a higher GSM always mean a better wipe?
No. Heavier sheets hold more liquid and resist tearing, but they also take longer to dry, cost more per piece, and can leave more residue if the fibre blend is wrong. Match the weight to the task, not to the price list.
The framework is deliberately boring. That is the point: a specification should be reproducible by someone who was not in the room when it was written.
