Ask a machine shop how long it takes to clean a machine tool and you will get three different answers from three different people. That is not because the people are careless. It is because cleaning a machine is not one job. It is a sequence of jobs with different durations and different owners, and the part everyone pictures - the minutes spent spraying - is usually the smallest part of the work. A time estimate that starts from a machine's specification sheet will fail in the first week. An estimate that starts from your own machine, your own soil and your own crew can be defended in front of a production manager, and it is the only kind a supplier can answer usefully.
The method below builds the estimate the way a buyer actually needs it: segment by segment, from what is found on the machine, with the equipment numbers read as time inputs rather than as promises.
Write down what the job actually contains
Before any number is estimated, the job should be written out as the segments it really has. For a machine-tool exterior clean done inside a planned stop, those segments are almost always the same:
- Shutdown, cooling, and the lockout and tagout your own procedure requires
- Dry removal of loose chips and debris before any water or heat arrives
- Protection and masking of parts that must stay dry or must keep their oil film
- The cleaning passes themselves, with the dwell time the soil actually needs
- Drying, and directing the runoff to a place where it can leave
- Re-lubrication of every surface the cleaning stripped
- Inspection, including checks whose results only appear hours later
- Restart checks that return the machine to production
Each segment has its own duration driver, and none of them scale with the others. A quote built as spray time multiplied by two hides most of the work inside the multiplier. Write the segments separately and estimate each one separately; that is the whole trick.
Let the soil state set the number of passes
The largest single driver of cleaning time is the condition of the soil, not the capability of the machine. Three states are worth distinguishing, because each answers to a different amount of time.
A fresh oil film moves with a cloth without resistance. It responds to brief heat and a short dwell, usually one pass. A polymerised film feels tacky or looks dull, darkens at its edges, and resists wiping; it needs sustained heat and two or three short passes. A bound deposit that has collected fine grit feels gritty between the fingers and will not move under heat alone; it needs mechanical removal first, and that removal belongs on the estimate as its own line, not buried inside the cleaning line.
Walk the machine at end of shift while surfaces are still warm, use oblique light rather than overhead light, and wipe two or three representative surfaces with a cloth the whole crew has agreed on. Record one line per surface: location, state, and the pass count that state implies. Where heat and dwell are the mechanism, as they are in machine tool steam cleaning, the pass count is fixed by how fast the soil softens at the surface, which is why the soil survey changes the estimate more than the choice of machine does. An estimate built on the wrong soil state tests the wrong method and reports the wrong conclusion.
Count the protected parts, not just the surface
A machine tool is not a truck panel. Guideways, ball screws and racks carry a film of oil that protects them, and cleaning removes that film along with the dirt. Re-lubrication the same day, per your maintenance schedule, is therefore part of the job's duration and not an optional extra. Electrical cabinets and cable glands need masking. Nameplates and decals need a test corner before anyone commits. Precision tables and gauges should be out of the spray zone entirely.
The honest arithmetic is simple: the cleanable area is the visible area minus the protected area, and the protected area takes time to cover and to uncover. Condensation also travels. It collects under labels, beneath seams, at the bottom of cabinets and behind the machine, whether or not anyone sprayed there, and checking those places is time as well. An estimate that counts spray minutes but not masking and checking minutes will make every cleaning method look unreliable, because the uncounted minutes are exactly where the day goes.
Read the specification sheet as time inputs
Once the job is broken into segments, a specification sheet stops being marketing and becomes a set of time inputs. Take the published data for the PowerDe HW02 fixed steam cleaning machine as a reading example. The 160 degrees Celsius steam temperature and the 0.8 MPa steam pressure are two readings of one saturated state, not two separate capabilities. They describe how quickly soil softens where steam condenses on a cooler surface. They do not describe how many square meters an hour the machine covers. Ask whether the figures are measured at the boiler or at the nozzle, because a long hose changes what actually arrives at your surface.
The 6 kW heating power and the 80 L stainless tank describe a work cycle and a refill walk, not an area rate. They tell you how long the machine runs before someone walks to a tap, and that walk belongs in your estimate whether you count it or not. The mains water inlet means water must reach the working point. The 220 V supply needs your own electrician to confirm a dedicated circuit. The built-in protections - electrical leakage, short circuit, over-temperature, over-pressure and low water - protect the machine. A low-water stop is a scheduling event, not a fault, and it should be planned like one.
Two numbers that matter most to a time estimate are not published on that page: how long the machine takes to reach working pressure from cold, and how long one tank lasts in real work. Hose and lance length, which set the working radius from wherever the machine stands, are also absent. Request them in writing before the estimate is finalised, or measure them on the first day and write the measurements into the procedure.
Add the site conditions that quietly consume minutes
Several of the largest time losses are properties of your shop rather than of the machine, and they can be counted before anything is bought. From the position the machine will actually occupy, measure the distance to the nearest water point, to the power outlet, and to the path the runoff will take. Decide who will drag the hose and how often.
Then look at access. Reaching over the top of a machining centre, working at its rear, or standing beside a neighbour machine all take longer than the same square meter on an open floor, and the floor plan will not tell you this; a tape measure and your own feet will. Note whether you have low-angle lighting for the inspection checks, because a check done badly is a check repeated. The destination of oily runoff is a question for your site and your local drainage authority, and their answer may add containment steps to the job. Every one of these is a number you can write down today.
Measure one real machine before you buy
Everything above narrows the range of possible answers; only a measurement closes it. The estimate worth paying for comes from one machine you already own, cleaned by the crew who will do the work, inside a window that already exists in your plan.
Write the acceptance language before starting: what a surface must look like under oblique light, what the agreed cloth must show, what must be absent from seams the next morning. Tape half of one representative panel and leave the other half as the comparison. Change one variable at a time - distance, pass speed, dwell, steam mode - and photograph fixed reference points so the second attempt can be compared honestly with the first.
Then time everything: preheat, each pass, the refill walks, the reheating, drying, re-lubrication and inspection. Check the machine again the next morning for flash rust and for water that migrated into seams overnight. If your machines carry two different soil profiles, measure two machines. Send the supplier the record rather than the summary. A supplier who answers a record with numbers is one worth planning with; a supplier who answers with a brochure is not.
What the finished estimate should say
The finished estimate is a one-page document. It contains the segments with their measured durations, the soil states found on each surface, the protected areas and their masking time, the refill and reheat events written into the timeline, and the questions still owed by the vendor:
- The nozzle-side values behind the published temperature and pressure
- Hose and lance length
- Preheat time and runtime per tank
- Written confirmation for your own surfaces and soil types
If the measured job does not fit the window you have, the honest options are to shrink the scope to one surface, to split the work into repeats that return to the same surface at the same stage, or to move the job out of the shift calendar and give it the downtime it needs. Asking the crew to go faster is not one of the options; it is how good estimates die.
If you want the numbers the page cannot give you, describe your machine, its soil, and the window you can offer through our contact page, and ask for them in writing. An estimate built this way will not be a small number, but it will be a true one, and a true number is the only kind that survives its first week in the shop.
