A slit coater that runs perfectly in a supplier's demonstration and then misbehaves in your laboratory is a familiar story. The machine usually does exactly what it was asked to do; what was never specified is the set of measurements that would have proved it could do it on your substrate, with your slurry, at your line speed. The checks below are the ones worth agreeing before a purchase order is signed, and the ones to repeat once the equipment is installed and handed over.
1. Define the film you are trying to make before the machine
Start with a written film specification, because every downstream check refers back to it. State the target coating thickness and the tolerance around it, the coating width, the substrate and its thickness, the solids content and viscosity range of the coating liquid, and the drying conditions the material needs. For slit coating equipment on the PowerDe site, the published envelope is a coating thickness of 5-200 um and a coating width of 50-350 mm, with an equipment speed that can be set between 1 and 10 metres per minute. Compare your specification with that envelope before discussing options, and ask directly which part of the range the machine will be configured for. A unit set up for thin, fast coatings behaves differently from one built for thicker films at lower line speed, and the difference shows up later as a uniformity problem you cannot tune out.
Write the specification in the same units you will use to accept the machine. Vague phrases such as "uniform coating" cannot be signed off; "no more than the agreed thickness variation across the coated width, measured at the agreed number of points" can.
2. Fix the coating window for your own liquid
Slit coating is popular for perovskite, organic photovoltaic and battery work because it tolerates a wide range of process conditions while still producing an even film. That tolerance is not unlimited, and it has to be mapped for your formulation rather than assumed from a datasheet. Before commissioning, run a small trial matrix: vary the coating gap, the supply pressure and the line speed across a grid, and record where the film is clean, where it shows a meniscus line, and where the edge of the coating begins to bead or tail. The corners of that grid are your operating window.
Do this before the machine leaves the supplier if the laboratory there uses the same slurry and substrate you will run. If they will not let you test your own material, ask for a recorded trial on the closest formulation they have, with the parameters written down. A parameter sheet you inherit is worth far more than a demonstration you watched.
3. Specify how the web is held and guided
Most uniformity faults in roll-based coating trace back to the web, not the die. The slit coating machine listed by PowerDe uses an inflatable shaft of 3 inches for the unwind and an adjustable deviation-correcting device to keep the web tracking. Confirm three practical details for your substrate: whether the shaft holds your core size without slippage at full unwind length; how the correction device responds to a spliced roll, since a splice passing the sensor can shift the whole web; and how much tension the substrate can take before it wrinkles or stretches. Thin flexible substrates used in perovskite and organic work move easily, and a web that wanders by a couple of millimetres during a run produces a coating edge that wanders with it.
4. Agree the drying profile, not just the oven length
The oven is the part of the line most often treated as fixed. On the PowerDe listing the oven is given as 1 metre with electric heating, which suits a laboratory or pilot line; a production process may need a different length, more heating zones, or controlled extraction of solvent vapour. Ask what the oven can actually deliver at your line speed: the setpoint per zone, how the web surface temperature is measured, and how quickly the profile settles after a start. A wet film entering an oven that has not stabilised produces a film whose thickness and morphology change across the first metres of the roll, and operators usually blame the coating head for it.
For perovskite inks and other solvent-borne systems, ask separately how solvent vapour is removed and whether the oven atmosphere is monitored. Where the process is sensitive to humidity or oxygen, decide now whether the coating head sits in a glove box or in an open enclosure; PowerDe lists glove box spin coating equipment for work that must be isolated, and buyers moving from spin to slit coating on the same material should confirm the equivalent enclosure option before ordering.
5. Decide how the die is supplied, cleaned and purged
The die is where throughput and downtime are decided. Ask how the coating liquid is delivered — pumped or pressure-fed — and how stable that supply is across a run, since a drifting supply pressure shows up as a thickness gradient long before anyone notices a fault. Confirm the listed equipment pressure the machine is designed to work at, which the PowerDe specification gives as 6 kg/cm, and ask what happens at the start and end of a run: how much material is wasted purging, whether the die can be cleaned without removing it from the machine, and how long that cleaning takes. On an R&D line where the formulation changes weekly, thirty minutes of die cleaning between runs costs more than the machine's speed savings.
6. Set the uniformity test you will actually run
Thickness uniformity has to be measured in a defined way, or every supplier will report a different figure. Agree the measurement method before commissioning — the gauge, the calibration, the number of points across the width, the number of points along the roll, and how many consecutive metres must pass. Then define the acceptance level in the contract in the same terms. The PowerDe specification sheet for this series records a product pass rate of at least 99.5%; treat that as a claim to be reproduced on your own material, with your own measurement method, rather than a result your line is entitled to inherit.
Run the same test on a clean, known substrate before any new die or new ink enters the line. When a defect appears later, you will know whether the substrate, the ink or the die changed.
7. Check the coating on a functional device, not only on the film
A uniform film that produces poor devices means the coating process changed something the thickness gauge cannot see: wetting at the substrate interface, drying rate, or the surface energy of the layer beneath. For perovskite, OLED, quantum-dot and organic photovoltaic work, include a device-level step in commissioning. Complete a small number of full devices from the trial rolls, measure the parameter that matters in your application, and compare it with the same devices made on the reference process. If the film looks fine and the devices do not, the fault is in wetting or drying, and the fix belongs in the surface preparation or the oven profile.
8. Verify substrate handling and the room it sits in
Walk the route the substrate takes to the machine and back. Measure the door the equipment enters through, the lift, and the aisle to its final position; the machine outline on the published sheet is 2540 mm long, 800 mm wide and 1650 mm high, which fits most laboratories but not every doorway or goods lift. Then check what the room provides: the electrical supply and protection the heater needs, the compressed air the pressure system expects, a stable ambient temperature, and space for the operator to stand while loading a roll. Dust control matters more than most buyers expect; on thin-film work a room without filtered air produces pinholes that no parameter adjustment will remove.
9. Run the machine at its limits, not its middle
Commissioning normally happens at a comfortable setting, and the fault appears later at the extreme. Test at the thickest and thinnest film you intend to run, at the fastest and slowest line speed, and at the widest and narrowest coating width in your plan. Hold each condition long enough to see whether the process drifts as the roll unwinds, the ink warms, or the oven load changes. Confirm the fault indication behaves as documented: the PowerDe specification records a three-colour warning light, so test that a fault is actually announced when it occurs rather than assumed.
10. Agree the spare parts, service and training you will need
Ask which parts wear: seals and gaskets on the die, the shaft and its inflation fittings, heating elements, and the tracking sensor. Get prices and lead times in writing for the consumables and the first-replacement spares, together with a service response commitment and the language a support call will be conducted in. Then book operator training against the machine you actually own, not a general course: the trained people should know how to change a coating width, clean the die, correct web tracking and read the fault prompts. That training, combined with the parameter record from commissioning, is what lets a new shift repeat a good run.
11. Write down every setting the day the line is accepted
On the acceptance date, record the recipe that produced the passing result: gap, pressure, speed, oven setpoints, substrate and ink batch, and the measurement data that proved it. Store it where the operators can see it, not in the commissioning report. Six months later, when someone asks why the process is set that way, the answer should be a document with measurement data behind it rather than the memory of whoever was standing at the machine.
Shenzhen PowerDe Automation Precision Equipment builds coating equipment for perovskite solar, OLED, lithium battery, quantum-dot LED and OPV research and production, alongside the wider thin-film solar equipment line that covers sputtering, laser cutting and lamination. If the checks above do not match the specification you have been quoted, send your film specification, substrate and target throughput to the contact page and ask for a configuration built around your process.
