Ask three maintenance supervisors how heavy the oil is on the same machine and you will get three different words for it. Each of them is right about what they can see. None of them is answering the question that matters, which is not how heavy the deposit is but whether it has reached the point where someone should act on it today.
That distinction is the whole difficulty of industrial equipment oil removal on metalworking equipment. A dull film on a guard panel that has been there since the last planned shutdown is a different situation from a bright wet patch under a fitting that appeared between two inspections, even though the darker one looks worse. A plant that assesses buildup by appearance will spend its cleaning hours on the surfaces that photograph badly and will keep missing the small, recent, still-growing deposits that are actually changing how the equipment behaves.
An assessment that is worth doing produces two things: a written statement of the condition at which a named person acts, and a check frequency that follows how fast the condition returns. Everything below is in service of those two outputs.
Why "heavy" is a judgement rather than an observation
The word heavy carries no shared meaning across a plant, because it quietly combines three separate facts that belong in different hands. The first is what the deposit is made of, which decides what will remove it. The second is where it sits, which decides whether anyone can reach it and what it is standing in front of. The third is how fast it came back after the last time it was removed, which decides when anyone needs to look again.
A single word collapses all three, so the answer it produces is a cleaning job rather than a decision. Write the three facts separately and the work usually gets smaller. One deposit turns out to be a seal that is still leaking and belongs on a maintenance work order. Another turns out to be a routine film that belongs in the end-of-shift wipe-down. A third turns out to be old and hardened and needs a mechanical step before any cleaning method is worth discussing.
Fact one: what the deposit actually is
On metalworking equipment a deposit is rarely only oil. The oil is the carrier, and what it carries changes what you are dealing with. Grinding and honing oils hold fine abrasive. Sump and conveyor deposits hold swarf. A water-soluble fluid that dries on a guard leaves wax-like additive solids that feel oily to a thumb and are not. Press and drawing compounds harden with heat and light until they no longer move. Each of these responds to a different first action, so identifying the material comes before choosing the method.
Three checks take about a minute each and need no chemicals:
- Shine a work light across the surface at a low angle rather than looking down into it under the ceiling lights. A fresh film moves with the light and lifts clean on a cloth. An oxidised film looks dull, darkens at its edges, and smears instead of lifting.
- Press a white cloth that the whole group has agreed on, and note whether it slides or drags. Sliding with no resistance is a surface film. Drag, or a cloth that comes back grey on one pass and clean on the next, is something bonded to the surface.
- Lift an edge with a plastic scraper and rub a little of what comes off between your fingers. Grit means the deposit is carrying solids, and heat alone will not remove it: the carrier softens and the solids stay, to be redeposited downstream. Those deposits need a dry mechanical step first.
Name the source while you are looking. A film that returns within days of a cleaning has a source that is still running, and cleaning it resets a clock without changing the rate. A hardened coat that has clearly been building for months is a backlog, not a leak, and belongs to a different kind of work order. Mixing the two is how a plant ends up cleaning the same floor corner every week and never fixing the fitting above it.
Fact two: where it sits and what it is standing in front of
Severity is not the thickness of a deposit. It is what the deposit prevents. The aisle-facing panel is usually the cleanest surface on the machine and the least informative, because it is the one everyone sees and therefore the one most often wiped. The deposits that matter hide where nothing is designed to be looked at: under the drip edge of a guard, below fastener heads, behind the chip conveyor, under the lid of a coolant tank and along its overflow path, on the outside of a sight glass or oil-level window, on a nameplate, on the top face of a control cabinet where mist settles, and on the floor beneath the machine where migration ends.
For each location you record, write what it blocks. The list of things oil blocks is short and concrete:
- A reading: a sight glass, a level gauge, an oiler window, a grease nipple, a fill port cap.
- An access: a drain plug, a fill cap, an inspection cover, a bolt head that a planned service will need.
- A check: a hairline crack behind an old film, a weeping fitting, a loose bracket, a worn hose cover.
- A function: a motor housing that runs hotter under an insulating coat, a hot surface with a fuel load near it, a walkway where film becomes a slip, a downstream station where oil migrates onto packaging or onto a finished part.
A deposit that blocks a reading or an access is a maintenance problem wearing cleaning clothes. A deposit that only looks bad on an outer panel is housekeeping. Sorting the two is the most valuable thing an assessment does, and it is invisible in a photograph.
Fact three: how fast it comes back
This is the step almost every assessment skips, and it is the only one of the three that tells you when to act. Choose one reference area on the machine, photograph it at a fixed angle with a fixed light, and clean it. Then watch it. Record the first day the film is visible again, and the day it reaches the state you defined in the section above. Express the result in days.
A rate expressed in days is the only form that can be compared across machines, across shifts, and across months. Two machines with the same appearance can have rates that differ by a wide margin, and the one returning faster is the one with an active source. The rate is also how you find out whether a repair worked: replace a seal, clean the area, and measure the rate again. If it has not changed, the repair did not reach the cause, and the plant has bought itself a clean machine and the same schedule.
Check the reference area on a short fixed cadence at first and a longer one afterwards, and keep the same person doing the check, or write the check so precisely that anyone can do it. Different observers produce different rates, and that difference looks like a change in the machine.
Writing the trigger so it survives a shift change
A trigger is a sentence that a person who was not present at the assessment can apply. It has three parts: how to look, what state counts, and what to do. "Clean it when it gets heavy" has none of them. A trigger that works reads, in effect: shine the light across the guard at a low angle; if the film drags a white cloth and darkens at its edges rather than lifting, it has reached the condition; raise the task and tag the zone.
Write a separate trigger for each zone type rather than one standard for the plant. The trigger for a sight glass is whether the reading can be taken. The trigger for a walkway is whether the surface is underfoot and whether the film is fresh enough to transfer to a shoe. The trigger for an electrical enclosure is whether anything has reached the cable entries or the label. The trigger for a hot surface is how thick the coat is, because that is a fuel and heat-load question rather than a cleanliness one.
Name the person who acts and the person who signs the area back over. An assessment with no names attached produces findings that travel around the plant as opinions.
Setting the check frequency from the measured rate
Once you have a rate expressed in days, the frequency follows from it rather than from a calendar habit. Check faster than the deposit returns and the team stops believing the check. Check slower than it returns and the condition runs ahead of you, which is how a plant discovers a leaking fitting as a floor hazard instead of as a seal.
Two rules keep the schedule honest. First, a zone whose rate is short belongs to maintenance until the source is closed; no cleaning frequency makes an active source manageable, and every interval you set for it will be overtaken. Second, frequency changes on findings, not on dates. If the same zone appears twice in a row at the trigger condition, the correct response is to change the interval or the method and to record why, not to remind the crew. If a zone keeps coming back clean, lengthen its interval and spend the difference somewhere that needs it.
What heat contributes to the assessment, not only to the cleaning
Heat is also a measuring instrument. On a hidden lower edge, mask off a small patch and leave half of it untouched as a reference. Apply a hot, damp cloth for a minute, then compare the treated half with the reference under the same low-angle light. If the film lifts, heat plus time plus a cloth is the right family of method for that deposit, and the assessment can proceed. If it smears darker and stays put, the deposit is beyond heat alone and needs a mechanical step first, and any method you test on it will look worse than it is.
When you read a machine's data sheet, read it as a set of conditions rather than as a promise about your surfaces. On the machine tool steam cleaning listings, a published steam temperature and a published vapour pressure are two readings of one saturated state, not two separate advantages to compare between suppliers, and neither states what arrives at the surface after a long hose and a cold wand. Ask in writing whether the figures were measured at the boiler end or at the nozzle end. Heating power and tank capacity describe the working cycle and the refill distance, not how much area a shift can finish. Preheat time, how long a tank delivers steam, and how quickly a machine recovers after a refill are usually not published, and should be asked for or measured on your own equipment before anyone builds a schedule around them.
The PowerDe HW02 listing, often used as a worked example, publishes 160 degrees Celsius steam temperature, 0.8 MPa vapour pressure, 6 kW heating power, an 80 L SUS304 tank, a tap-water inlet, 220 V operation, and a selection between dry and wet steam. It also lists the machine's own protections against leakage, short circuit, high temperature, high pressure, water shortage and high-pressure explosion. Those protections guard the machine, not your circuit, your floor, or your schedule. The listing documents vehicle exterior cleaning, interior vacuuming and disinfection, and engine-compartment cleaning; it does not document metalworking equipment, so applicability to machine surfaces, together with hose and wand lengths, preheat time and per-tank runtime, should be confirmed with the manufacturer in writing before they are relied on.
What to record, and what to send the supplier
One page per machine group is enough, and it should be short enough that it actually gets filled in. For each zone: the machine and zone identifier, the date, the state of the deposit as you found it, the source if you can name it, the measured return rate, the trigger sentence, who acts, how the area is verified, and the date the entry was last reviewed. Verification is deliberately physical: the low-angle light, the agreed white cloth, a photograph from the same fixed point, and a low-angle check of seams and drip edges where water and oil collect after any cleaning.
Send that record to your equipment supplier rather than a summary of it. A summary invites a brochure. A record with zone names, deposit states, measured rates and your own acceptance wording gets a specific answer, and the difference between those two replies is the difference between a machine that fits your plant and one that fits a demonstration.
Before you order anything, close the open items in writing: which surfaces the manufacturer considers in scope for your deposits, what the published figures measure and where they were measured, the hose and wand lengths, the preheat and per-tank times, and the site conditions that your own electrician, water supply and drainage responsibilities require. You can start that conversation through contact PowerDe.
The output is a rule, not a report
A plant that assesses heavy oil buildup well does not end up with better photographs of dirty machines. It ends up with a short list of zones, each carrying a description of the deposit, a named source, a measured return rate, a trigger anyone can apply, and a check interval that was derived rather than inherited. That list makes the next decision easy: which zones are maintenance work, which are routine cleaning, which need a mechanical step before any method is discussed, and which can be left alone and checked later.
Start with one machine group and one zone on it. Write the trigger before you clean it, measure the rate after you clean it, and let those two facts set the interval. An assessment that produces a rule will still be working long after the people who wrote it have moved on to the next job.
