A slit coating equipment purchase is usually decided in the laboratory, where a uniform film on a small coupon looks like proof that the process works. The machine you receive, however, is a roll-to-roll production asset with a web path, a dosing system, a drying section and a control platform attached to that coating head. The factory acceptance test is the last moment when most integration problems can be corrected without dismantling a line that has already been shipped. A useful factory acceptance plan does more than confirm that the machine powers up and moves web. It states which results will be measured, how they will be measured, and what both sides agree counts as a pass.
Start with the process, not the machine
The core of the test should be a wet coating run using your own ink or slurry and your own substrate. A trial with the builder's reference fluid on a spare PET web proves that the hardware functions, but it does not tell you whether your formulation still levels correctly, whether the drying profile still suits your solvent, or whether your substrate feeds reliably. Agree the trial material in writing before the machine is built, and arrange for the material to arrive early enough that several set-up attempts are possible.
Record the accepted recipe during that run: solids content and viscosity of the ink, substrate type and thickness, target wet and dry film thickness, web speed, pump settings, header pressure, and drying settings. That record becomes the reference you will compare against when the line is reassembled at your site, and it is the document your process engineer needs when the same product is reproduced months later.
Define how coating quality will be judged
"Uniform coating" is not an acceptance criterion. Specify the measurement method, the sampling positions and the tolerance. In practice this means:
- Thickness measured at several positions across the coated width, and repeated at intervals along the length, so that variation across the web and variation over time are separated from each other.
- The measurement instrument named in advance — offline gauge, spectroscopic film-thickness meter, or sheet-resistance mapping — together with the instrument that will be used at your site, so that the two data sets remain comparable.
- Visual criteria for defects such as streaks, comet lines, stripes, pinholes, dewetting, dry ink at the die lip, and edge beading, with sample panels kept as physical references.
- A repeatability check: the same recipe run three times, with the results compared against each other rather than against a single successful trial.
Also test the coating head and dosing system across their working range rather than at one comfortable point. Confirm that the pumps hold the set flow, that start and stop transitions do not leave a heavy or thin band on the web, that the purge sequence limits how much ink is wasted at the head, and that the coating window — the combination of speed and flow that produces an acceptable film — is wide enough to be useful in daily operation.
Test the web path and motion systems under load
Coating defects often originate upstream of the die. Check that unwind and rewind run smoothly from full roll to core, that the dancer or tension control responds without overshoot, and that web guiding keeps the substrate tracking straight into the coating unit. Verify speed stability at the rated line speed and inspect the acceleration and deceleration ramps, because film thickness follows the ratio between flow and speed. Confirm that the load and unload path handles your substrate format without marking, creasing or scratching it, and run the machine continuously at speed for the agreed duration so that thermal drift and any mechanical vibration have time to appear.
Measure the vacuum and thermal performance you paid for
If the coating chamber operates under vacuum, record the pump-down time, the ultimate pressure and the leak rate after the chamber has been closed and baked out. For the drying or heating section, ask for a temperature survey across the working width and down the web path, taken with independent sensors rather than only the machine's own control readings, and check how tightly each zone holds its set point once it is reached. These two items are frequently left out of acceptance plans, and both are difficult to correct after installation.
Check the control system, data and safety functions
Work through the machine as an operator would. Save a recipe, recall it, and confirm that every parameter returns to the stored value. Trigger the interlocks and emergency stops and confirm that the machine responds as the safety documentation describes. Force a few alarms — a broken web, an empty supply tank, a door opened during a run — and check that the message shown matches the actual fault. Confirm that the trend and batch data the system logs are the parameters your quality records require, and that units and limits in the interface match the conventions used in your laboratory.
Confirm the utilities and the site requirements
The acceptance report should state the machine's actual measured requirements, not its nameplate values: supply power and measured current draw, compressed air pressure and consumption, exhaust and ventilation demand, and any nitrogen or other gas supply. Add the floor space and service access needed for maintenance, the weight and door dimensions relevant to delivery, and the consumables and spare parts that will be required in the first year. Buyers integrating the equipment into a wider perovskite production line equipment plan should check these figures against the utilities available at the installation site before signing, since a mismatch found after shipment is an expensive change to the building rather than to the machine.
Agree the documentation, training and punch-list process
Documentation is part of the deliverable. The package normally includes general arrangement and utility drawings, electrical and pneumatic schematics, a parts list with the consumables and recommended spares identified, operator and maintenance manuals in the agreed language, and the declarations required for your region. Confirm which items are supplied at shipment and which follow later.
Training should cover both operation and troubleshooting for your own staff, and the plan should say how remote support is provided afterwards and what the warranty period covers. Then handle the punch list properly: every deviation found during the test is written down with the agreed corrective action, the responsible party, and the criteria that will close the item. Sign-off follows completion of those actions, not a promise to complete them.
Decide how you will witness the test
If your team cannot travel, a witnessed remote test can still be reliable provided the plan fixes the format in advance: live video rather than recorded clips, your own measurement samples tested on the builder's instruments, independent measurement records shared as they are produced, and a named contact on each side with an agreed procedure if a test fails. Agree in the same document which results carry over to site acceptance testing and which must be repeated after shipping and reinstallation, since alignment, web handling and drying behaviour can all shift in transit.
Before you sign: a factory acceptance plan is complete when every line in it can be answered with a measurement or a physical document. If a section can only be answered with "normal" or "as agreed", rewrite it now, while the machine is still in the builder's hall. PowerDe's thin-film solar equipment range covers slit coating and related thin-film R&D and production equipment, and acceptance terms can be agreed with its engineering team before the build begins.
