A steam cleaning trial usually answers the question it was designed for, and then the record stops. Someone notes that the soil came off faster, the machine goes back to the store, and six weeks later nobody can say which nozzle made the difference, whether the extraction brush was fitted, or whether the result came from a hotter pass or from the second cleaning cycle. Write the record so it carries the machine's configuration as well as its outcome, and one trial becomes a baseline that later trials can be measured against.
Why accessory configuration decides the result
On a steam machine, the accessory set is part of the process, not an afterthought. The same cabinet produces different cleaning behaviour depending on the nozzle on the lance, whether the high-pressure water function is in use alongside the steam, whether a vacuum attachment is fitted and connected to a recovery tank, and which power source the unit is drawing from. A PowerDe PWD02 steam cleaner, for example, is documented with hot water, dry steam and wet steam modes, a high-pressure water gun with nozzle, and a vacuum cleaner with a flat nozzle served by a sewage tank. Changing any one of those changes what the trial proves, so the accessory list belongs in the record with the same weight as the surface and the soil.
Configuration also changes what you are allowed to compare. The same 4.5 kW, 60 L specification that appears on the PowerDe HMW2 steam cleaner page is supplied in a variant that runs from a 48 V mobile battery provided by the customer, and that variant's page does not list the vacuum method used by the wheeled version. A trial run on one variant is therefore not evidence about the other, and the record has to say which variant was on the floor.
1. Fix the trial question before the first pass
Write one sentence that states what the trial is meant to decide. Typical questions look like this: can the machine remove the soil to an acceptable standard on this surface; is one accessory configuration clearly faster than another; can the cycle be repeated by the staff who will actually run it. A trial without that sentence tends to collect impressions, and impressions cannot be compared across machines or across days.
Name the surface and the soil as specifically as you can. A floor finish, a painted panel, a gasket, a workbench or a vehicle body each respond differently to heat, to moisture and to mechanical action. A trial that mixes several surfaces in one session produces a record that supports no conclusion, because the surface variable moved with the accessory variable.
2. Record the machine as delivered, not as advertised
Before the first pass, write down the identifying data: manufacturer and model, serial number if one is fitted, and the power and supply arrangement actually used on site. For a fixed-supply trial that means the socket, cable and protection arrangement; for a battery trial it means the battery itself, where it came from, and how it was charged. Where a machine is offered with more than one power option, the record should state which one was connected, since run time and recovery between passes follow the supply.
Add the physical footprint and handling observations. A unit listed at L1275 × W570 × H1185 mm needs a doorway, a ramp and a working radius that match those dimensions, and the trial is the moment to find out whether the access route assumed in the plan exists. That information is worthless if it is not written down next to the machine that produced it. If published documents for the same model give different figures, record the figure you measured and name the document it disagrees with.
3. Photograph the accessory set as a configuration
Lay the accessories out and photograph them before the trial starts, then photograph them fitted to the machine. A photograph settles arguments that a written list does not: which nozzle was on the lance, whether a brush was added, how the hose was routed, whether the extraction tool was in use. Attach both images to the record with the date written on them.
Then give each accessory a line of its own in the table, with three columns: what it was used for, what was observed with it, and what the cleaning operator would change next time. Lance and nozzle, gun, brush or pad, extraction tool and its recovery route, and tank arrangement are the items that normally carry real differences in outcome. Where the machine documentation lists a mode the trial did not use, say so explicitly; an unused mode is a fact about the trial, not an omission.
4. Log the consumables and the support services
Water is a consumable with consequences. Record the source, the fill volume and how often the tank was refilled during the session. A PowerDe steam cleaning machine in the cleaning equipment series is documented with a 60 L clean-water tank and a tap-water inlet with 4-point internal thread, so the fill and inlet arrangement belong in the same row as the cleaning result, not in a separate note.
Record recovery as a separate line as well: how much soiled water and residue the machine collected, where it went, and how long that step took. Trials are often won or lost on the recovery step rather than on the cleaning pass, and a record that measures only the pass hides the reason a process does not scale from a demonstration to a routine job.
5. Score the result against criteria agreed in advance
Agree the criteria with the people who will live with the result, and write them before the trial. Useful criteria name something checkable: the soil removed from a defined area, the number of passes needed, the time per area, the residue left behind, the drying time before the area can be used again. Score the trial against those criteria for each accessory configuration, not against a general impression of cleanliness.
Where the comparison spans machines of different sizes, keep the comparison honest. A bench-marked comparison between a mobile unit and a fixed unit such as the PowerDe HW02 steam cleaner, listed at 160 °C, 0.8 MPa, 6 kW and an 80 L tank, only makes sense if the same surface, the same soil and the same operator are carried across both runs. Note the differences you could not control, because a reader of the record needs to know how much weight the conclusion can carry.
6. Record safety observations and limits, not just performance
Every trial touches safety, and it is the part most often left out of the file. Record the protections the machine is documented as having, and what the operators actually did with them: the driving style steam cleaner page lists leakage, short-circuit, high-temperature, high-voltage, water-shortage and high-voltage explosion protection, and a trial is where you verify that the workflow respects them. Note hot-surface handling, the routing of hoses in a walking area, and how the area was protected while it was wet.
Also record what the machine did not do, or what the trial did not test. If a surface type was excluded, if a mode was not exercised, or if the run time on battery was not measured to depletion, say so. Buyers who read a record that states its own limits trust the rest of it; a record that stays silent invites them to assume the trial covered everything.
7. Compare configurations, then keep the winner
When the trial is over, produce one summary table with one row per configuration and the criteria as columns. Read down the column that matters most to you. The comparison usually shows that the difference between configurations is larger than the difference between machines, which is exactly the finding a written record can prove and a conversation cannot.
Turn the best configuration into a written instruction: the accessory set, the mode selection, the sequence of steps, the consumables needed and the time expected. Archive the photographs, the raw scores and the parameter notes with it. On a multi-machine programme, keep one template and let each machine add rows to it, so the next trial inherits the comparison instead of starting one.
If your cleaning process is already partly documented, it is worth checking the written steps against what the machine can actually do before you standardise them; kitchen cleaning process review covers how a cleaning plan is normally laid out and where such plans go wrong.
Before you file the record
Three checks decide whether the file will still be useful in a year. Does every result name the accessory configuration that produced it? Does every claim about the machine name the document or the measurement behind it? And does every untested item appear as untested? A record that passes those three checks supports a purchase decision, a second trial and a training document at the same time. If you want help designing the trial and the record around a specific machine and site, the team at Shenzhen PowerDe Automation Precision Equipment can go through your surfaces, soils and access constraints and tell you which configuration is worth testing first.
