Step 1: List the Enclosure Surfaces and Name the Soil on Each
A press exterior is not one surface. Around the same machine you have a painted steel cabinet, an operator control panel with a display, cast or machined frames where cutting oil and graphite dust bake onto the metal, guarding panels that are handled constantly by gloved hands, and floors immediately around the machine where dropped oil collects dust into a film. Each carries a different soil, and each reacts differently to heat and pressure.
Walk the machine with a checklist and record, for each surface, what the soil actually is: cutting fluid residue, hydraulic oil, graphite or abrasive dust, sugar and flour dust, food residue, or simply fingerprints and grime on controls. The soil decides whether you need a cleaning equipment series that runs dry steam, wet steam or hot water, and at what pressure. Grease and cutting-oil film need heat and dwell time; loose dust needs volume and extraction; fingerprints on a panel need neither, only a method that will not damage the display.
Write this list down in the machine's own language — panel, guard, frame, base, feed table — so that operators and the equipment supplier are reading the same item numbers later.
Step 2: Decide, in Advance, What a Pass Looks Like
A criterion you can sign off on is one a person can judge without a laboratory. For exterior cleaning, practical pass tests are visual and physical: no visible oil sheen on a light-coloured panel; no slippery film on the floor around the machine; no dust packed into the seam between guard and frame; no residue on a wipe drawn across a horizontal ledge; controls and labels legible with no film on the display.
For shops running CNC and machining equipment, CNC machine cleaning equipment is usually specified around one harder test: can a hand run across the machine base and come away clean, and does the floor stay dry enough that nobody slips before the next shift starts. Set that as the standard for those surfaces and photograph the accepted condition, so a new shift worker judges against a picture rather than against memory.
Also write the negative criteria — what a pass must not include. A surface cannot count as clean if water has entered an electrical enclosure, if a label has lifted, if a bearing or guideway has been sprayed directly, or if the machine has been left damp where corrosion starts overnight. Acceptance criteria that ignore these will quietly reward over-spraying.
Step 3: Fix the Boundaries Before the First Cleaning Trial
The fastest way to lose an internal argument about cleaning equipment is to let a trial spray something it should not. Before any trial, mark the no-touch list on the machine itself: open cable entries, connector housings, air intakes and filters, bearings and guideways that rely on their film of grease, optical and measurement surfaces, and any controller that is not rated for the humidity a steam process creates.
Then choose a machine for the trial that is representative: one that is genuinely dirty, one operator, one shift, and the surrounding floor included in the scope. Cleaning the easy machine first produces a flattering result that sets an expectation you cannot repeat on the hard ones. Record the time taken, the water used, and what the operator had to stop and re-do.
A mobile, battery-capable unit lets the operator work around the machine without dragging a cable across the shop floor; a fixed unit with its own tank and heating power suits a set bay where machines are brought to the cleaning station. Which one fits is decided by your layout, not by the machine's brochure.
Step 4: Put the Numbers on One Sheet and Sign It
The finished criteria sheet should be short enough to sit on a clipboard: the surface list, the soil on each surface, the pass test for each, the no-touch boundary, who inspects, and the time the process is allowed to take. Add the equipment settings actually used, so the result is repeatable when the experienced operator is on leave.
Review the sheet after a month, not after a year. Ask three questions: which surfaces still fail, which settings were changed informally by operators, and whether the time on the sheet matches the time the work takes. Criteria that nobody updates drift into fiction, and the first audit that catches the gap is usually a customer's.
If your surfaces and soils are mixed across several machine types, send your surface list and the soils you recorded to PowerDe, and ask which cleaning method and which machine class matches each group before you standardise on one unit for the whole shop.
