The quotation for a coating tool almost always says how wide the bead can be, how thick the layer can go and how fast the web or plate can move. It rarely says anything about the conditions that decide whether those numbers still hold on your floor: the ink in the tank, the substrate that arrives from your own upstream line, the room the machine will stand in, the air that touches the wet film. Those are the parts that fail in commissioning, and they are visible early if the review is ordered properly. A slit coating equipment enquiry is the right place to start that conversation, because the coating window is where most interface conflicts first appear.
Put the coating window in writing before anything else
The published envelope of the slit coating equipment listed under thin-film solar equipment states a coating width of 50 to 350 mm, a coating thickness of 5 to 200 micrometres and a coating speed of 1 to 10 m per minute, with the machine frame at 2540 by 800 by 1650 mm and a one-metre oven section. Treat that envelope as the supplier's claim about the head, not as a guarantee about your process. The first document to produce is your own target window: the viscosity band of the material you actually intend to run, the areal weight or dry thickness you need, the width you need on the substrate, and the throughput per hour that makes the line worth building. Then ask the supplier to state, for each of your three worst cases - thinnest layer, widest bead, fastest line - what they will demonstrate, on what material, and what they will do if the demonstration fails.
Ask specifically which variables the machine holds closed-loop and which are set once and left alone. Pump metering, web tension, head gap, substrate temperature and dryer zone setpoints are the usual candidates. Whether a control loop exists for a variable, and whether it can be logged, decides whether a process engineer can later repeat a good run or has to chase it.
Characterise the material, not just the product name
Coating suppliers argue from viscosity and solids; on the floor the failure modes are different. Supply the rheology you measured, including how it changes with temperature and shear, the solvent or carrier and its boiling behaviour, the solids content and how much it varies between batches, the particle or filler loading and whether it settles in a static tank, and the pot life from fill to reject. Then state what your receiving inspection will actually measure per batch, and what you will do with an out-of-spec tank: dilute, re-circulate, or dump.
Two questions settle most of the tank and pipework design. First, does the material tolerate sitting still, and for how long - which decides recirculation, agitation and the size of the buffer. Second, what has to be cleaned out between runs, and how often - which decides whether the head is removable, whether the pump is quick-release, and whether you will buy a spare head rather than schedule a clean-down around every product change.
Describe the substrate exactly as it arrives
Coating quality is usually limited by what came in, so document the incoming substrate instead of assuming a clean, flat, dimensionally known plate or web. Record dimensions and thickness tolerance, flatness or warp, surface energy as received and how quickly it decays, residual dust or handling marks, and whether the substrate has already been through a plasma, corona or wet clean upstream. If you are moving from lab plates to a production format, say so explicitly in the enquiry and ask what changes in the head, the web handling and the drying profile at the larger format.
For development work that still runs on small wafers or coupons, the machine envelope matters less than the chamber. The glove box spin coating equipment page, for example, states 300 to 8500 rpm for silicon wafers up to four inches with uniformity of plus or minus 3 per cent, a leak rate below 0.05 volume per cent per hour, a differential pressure of plus or minus 15 mbar, a water probe reading from 0 to 500 ppm and a working environment of 15 to 30 degrees Celsius. Ask the equivalent questions for your line: what atmosphere the coated area is held at, what the water and oxygen limits are, and whether the transfer between coating and drying breaks that atmosphere.
Define the environment and the interfaces the machine must connect to
Make the room a line item. Temperature and humidity band, whether the coating zone is enclosed or open, dust class, and where extraction is permitted. Ask for the heat the machine rejects, the air volume it needs, and whether the dryer vents inward, outward or through your duct. A machine that was accepted in a supplier's factory and rejected in your building usually loses on one of these.
List every connection in writing and confirm each in both directions: electrical supply and rating, compressed air quality and dew point, any process gas, water or chiller duty, drainage for solvent and cleaning waste, floor loading and access route, and the door and aisle sizes the machine has to pass through on delivery. Then the signals. State your control platform, the historian or MES you already run, the tags you expect to receive, whether recipe management lives in your system or theirs, and how interlocks are exchanged with the upstream and downstream steps. A coating tool that cannot hand a fault signal to the neighbouring station stops your line, not theirs.
Agree the trial plan and the acceptance test before the order
This is the part engineering teams skip and then pay for. Write the trial as three stages. A bench or sample stage where the supplier coats your material on your substrate and returns samples you measure with your own instruments. A witness stage at the supplier where the run is repeated at your target speed and width, with the parameters recorded, not just the film. A site stage where the machine runs your material, your substrate and your shift pattern after installation.
For each stage, name the measurements and the instruments up front, since both sides will otherwise argue about method after the fact: wet and dry thickness and where it is measured across the width, coating weight, visual defect classes and their size thresholds, adhesion or cross-cut result, pinhole and thin-spot count, thickness uniformity across the width, and the yield you will accept over a defined number of consecutive runs. Fix how samples are handled and by whom, and what happens if your lab and theirs disagree. Also fix the consumables and wear items with their expected replacement intervals and prices, so lifetime cost is on the table rather than in the purchase order that follows.
Close the loop on after-sale service
Ask, and get in writing, who attends when the coating starts drifting: response window, whether an engineer travels or works remotely first, what spares are held where, and what training comes with the machine for your process engineers as opposed to your operators. The commissioning plan should name the person who signs acceptance and the evidence they will see.
If your substrate format, atmosphere requirement or throughput sits outside the published envelope, that is an engineering question and not a catalogue one. The team behind the automated production line solutions and the wider thin-film solar equipment range handles non-standard configurations, so send them your substrate drawings, your material datasheet and your trial plan and ask them to mark up what they can and cannot commit to. A supplier who returns a marked-up trial plan has answered the manufacturability question. A supplier who returns a quotation has not.
The short version: manufacturability is assessed on paper before it is assessed on film. Material rheology, incoming substrate, environment, control signals and a signed acceptance test are the five places to spend the review time. Fix those and the coating window in the datasheet becomes something you can actually hold.
