A resin-print failure does not begin when the machine reports completion and only a few supports remain on the platform. It is the result of a competition repeated at every layer: the newly cured cross-section must bond more strongly to the preceding layer or build platform than to the vat film, while motion must not tear the supports or distort the part. This guide does not provide universal exposure numbers. It builds a diagnostic order from symptom, location and repeatability so that one evidence-based variable is changed at a time.

The aim is not to answer every failure with more bottom layers, longer exposure and thicker supports. Those interventions can conceal the actual cause, reduce dimensional and surface quality, make removal difficult and load the release film unnecessarily. Useful diagnosis separates first-layer non-adherence, support-tip failure, delamination, position-specific optical defects, suction-cup geometry and damage created during washing or curing. The same cured sheet found in a vat can be the final consequence of several different failure paths.

Stop first and preserve the evidence

Do not immediately scrape the platform and launch the next job. Photograph the remains in their print position, the build platform, the vat bottom and the slicer's layer preview. Save the exact project, profile, resin name and batch, estimated room and material temperature, start time and printer firmware. Record whether the failure affected one object, every object or the same XY zone. The state before cleaning contains far more diagnostic information than fragments after washing.

Approach the machine with gloves and eye protection selected from the resin safety data sheet. Liquid photopolymer can sensitise skin, and contaminated wipes, gloves and wash liquid do not automatically belong in household waste or a drain. Ventilation, storage and disposal depend on the specific product information and local rules. Troubleshooting never justifies bare-hand contact, aggressive scraping in an open vat or an unknown solvent.

The first branch: what remains on the platform?

If the platform is empty or carries only a thin silhouette of the first layers, platform adhesion comes first. If the raft and supports remain but the model is missing, inspect support contacts, local minima and sudden growth in cross-section. If part of the object printed and then opened along a horizontal plane, consider interlayer adhesion, motion load, material condition or an interrupted exposure. When every object fails at exactly the same height, look for an event common to that layer or machine movement rather than treating each geometry independently.

A cured patch in the vat proves only that a layer detached and continued receiving exposure. Its shape and position provide evidence. A full raft outline suggests an initial-adhesion issue; a model silhouette points toward support or cross-section failure; a small mark recurring at one XY coordinate suggests film, display or optical trouble. Moving the same control model to another location helps distinguish a fault fixed in machine coordinates from one that follows geometry.

1. Nothing adheres to the build platform

Confirm that the platform is seated correctly, locked firmly, clean and calibrated by the manufacturer's procedure. Automatic levelling does not eliminate mechanical error: contamination at an interface, loose retention or a damaged surface can change real first-layer thickness. Do not sand a platform because a generic forum post recommends it. A coated or accurately machined surface can be damaged permanently. Begin with the maintenance and calibration method written for the exact printer and platform revision.

Bottom exposure must create a robust bond but cannot be treated as unlimited insurance. Too little energy can leave the first layer on the film. Excessive energy can produce an overgrown, over-cured base, elephant foot, difficult removal and inaccurate lower features. Verify the selected resin profile, layer thickness, light engine and printer variant. The resin supplier's range is a starting point; qualification should use a small measurable adhesion sample and incremental changes.

Temperature deserves its own check. Cold resin is more viscous, refills the layer gap more slowly and may resist release motion more strongly. Nominal room temperature is not the temperature of the full vat, especially after a cold overnight start. Use the resin maker's stated range and condition the material evenly. Uncontrolled direct heating, hot water or a heat gun can create local overheating and additional safety risks.

2. The raft remains but the model detached from supports

The first layer has now demonstrated attachment to the platform. Ask whether the earliest model layers had enough support and whether cross-section increased gradually. A support contact transfers release load; too small and it separates, too large and it creates a crater plus difficult finishing. Density alone is not the answer. Many thin contacts in poor positions may be weaker than a few correctly sized primary supports supplemented at the right local minima.

Search the sliced file for every island that starts independently. An unsupported minimum can form on the vat film for several layers, later attach to the part and introduce distortion or a delayed tear. Automatic support placement does not know which face is functional and may miss fine islands. Follow the layer preview from bottom to top and identify the first appearance of each feature, confirming that a continuous load path connects it to the platform.

Evaluate orientation by layer cross-section as well. A broad plate parallel to the film introduces a large area almost at once. Tilting it can make growth progressive, while excessive tilt needlessly increases height, time and support scarring. The target is not a universal 30- or 45-degree rule. It is a documented compromise among section change, visible surface, support access and dimension-critical direction.

3. Delamination or a horizontal crack

A horizontal opening in a partly completed object can mean that one layer bond became weaker than release load. Possible causes include insufficient normal exposure, cold or poorly mixed material, an aggressive lift profile, abrupt cross-section growth, contamination or a print interruption. If every model opens at the same Z height, inspect motion, source file, power and any environmental event. If only one geometry fails, its local load is the stronger hypothesis.

Lift speed and distance are machine-specific. Excessive speed can increase dynamic load on supports; insufficient distance may not complete separation from the film. Needlessly long and slow motion adds time and cannot rescue poor orientation. Tilt vats, conventional vertical lift systems and different release films require different profiles. Use the verified factory profile as a baseline, then change one parameter with the same control model and multiple repetitions.

4. Suction cups, sealed cavities and blowouts

A downward-facing cavity that closes can create a pressure differential and resist resin flow. It is not a perfect vacuum, yet it can place substantial additional load on the film, walls and supports. In a hollow model, the interior must communicate with the outside from the layer at which it begins. A hole appearing hundreds of layers later does not correct the earlier sealed cup. Find the first layer of the cavity and design pressure equalisation there.

An opening must be large and well placed enough for air and resin during printing, wash fluid during cleaning, and trapped liquid during drying and cure. One tiny hole may clog or fail to permit real circulation. Multiple openings should be positioned for both print orientation and washing. Internal lattice or supports can trap resin. A sealed, incompletely washed cavity can crack or leak uncured material months later, creating both product and skin-exposure risk.

Formlabs uses the term cupping blowout for a rupture in the side of a concave or hollow part. That is a symptom category rather than one universal force model for every printer. Practical prevention is to remove sealed sections, validate wall thickness, provide effective vents and orient the part for progressive growth. Increasing exposure alone does not remove hydraulic loading; it may merely make the affected wall stiffer or more brittle.

5. Failure that repeats at the same XY location

When different objects fail over the same platform area, rotate or move a control model. A machine-coordinate pattern may indicate damaged or contaminated film, a display defect, light non-uniformity, platform-plane error or debris beneath the vat. Inspect the release film under grazing light for scratches, cloudy stress lines, dents and loss of tension. Run display checks only by the manufacturer's method, with suitable eye protection and without looking directly into an active UV source.

Cured resin on a display protector can distort the image, yet removal is risky. Do not use a metal blade or aggressive solvent unless specifically permitted. A single droplet below the vat can change spacing and optical behaviour. After cleaning the vat exterior and display, run a suitable exposure or uniformity pattern. If the failure moves with the vat, film or vat becomes more likely; if it remains at the same machine coordinate, inspect optics and the platform.

6. Blurred, enlarged or missing details

Normal exposure controls more than whether a layer exists. It affects lateral cure spread, closure of holes, pin thickness and support-contact size. Too little can produce weak walls, incomplete features and poor layer bonding; too much can close negative text and small bores, enlarge external dimensions and fuse fine gaps. The useful setting is not the sample that feels hardest. It is the condition that reproduces measured positive and negative features together.

Display pixel size is not guaranteed part resolution. Light spread, vat film, pigment, layer thickness, exposure, orientation, washing and post-cure all contribute. A resin marketed as 8K or 16K cannot guarantee higher accuracy on every machine. Calibration samples must use the same mixed batch, temperature, cleaning and curing conditions, and the measurements should represent the feature that matters in the real application.

7. Wavy, rough or ragged surfaces

Partially cured flaps and rough patches—covered by defect categories such as ragging or rashing in Formlabs documentation—may relate to debris, unsupported material, a compromised optical path or an unsuitable profile. Determine whether the defect appears on every side, only on the film-facing region or in one XY zone. Photographs taken before washing help distinguish a printing defect from damage caused by excessive solvent time, curing while wet or rough support removal.

Mix resin thoroughly without introducing a large volume of bubbles. Pigment and filler can settle, leaving different composition near the top and bottom. Use a soft tool approved for the vat and do not scratch the film. If a new bottle or lot coincides with the problem, do not assume machine failure. Compare with a known, correctly stored resin, but do not blend unrelated products in unknown proportions.

8. Dimensional error, warping and post-process distortion

Print completion is not the final geometric state. A thin wall may flex under release force, a green part may soften during washing, support removal can release stress, and post-curing adds shrinkage and heat. Record when a measurement was taken: immediately after printing, washed, fully dried or fully post-cured. The same part can measure differently in each state. Functional tolerances must be qualified in a defined final condition.

Orientation also affects the measurement direction. A support-facing surface can be indented, while a long plane may bow under layer forces and polymer shrinkage. Compensation belongs only after a stable repeatable bias has been established. A global XY scale derived from one coupon may damage another geometry. Use several sizes of positive and negative features, at least three repetitions, and separate systematic machine behaviour from model-, support- and cure-induced distortion.

Vat film, vat condition and release mechanics

Release film is a consumable but not the automatic cause of every failure. Deep scratches, whitening stress lines, permanent dents, leaks or lost tension can justify replacement. Film type, thickness, orientation and retention are machine-specific; FEP, PFA/nFEP and ACF cannot automatically share one movement profile. A phone microphone frequency is not a universal tension standard. Follow the vat maker's screw sequence and parts list, then test for leakage away from the printer.

A hard fragment left after a failed print can puncture the film or break a display during the next homing cycle. Use the printer maker's tank-clean or filtering procedure. Do not press the membrane onto sharp debris or lever fragments with a metal tool. Inspect filtered material before reuse. Resin of uncertain contamination does not become safe merely because it has returned to a bottle; filter, funnel and storage compatibility matter too.

Profile changes: one variable and repeated trials

Choose a short representative control model containing fine supports, progressively growing cross-sections and both positive and negative detail. After the baseline, change only one variable—normal exposure, lift speed or contact diameter, for example. Keep resin lot, temperature, layer height, platform location and post-processing fixed. One success can be chance; at least three repetitions begin to show stability, while production requires a recorded success rate.

Do not optimise time first. A faster profile has no value if it causes more scrap, film damage or uncertain dimensions. Establish a stable baseline before reducing movement time or increasing speed in steps. Give every profile version a name and date. When firmware or slicer software changes, preserve the proven profile, clone it and run the same regression coupon rather than silently overwriting production evidence.

When the file or data path is the likely cause

If geometrically independent objects shift or lose material at the same height, inspect the sliced file and transfer path. Corrupt export, failing removable media, interrupted network transfer and firmware compatibility can create a shared defect. Open the exact file sent to the machine, not only the design project, and inspect the affected layers. When possible, reslice with a factory profile and a different verified medium. Demonstrating a data fault is stronger than replacing hardware at random.

The source mesh can also create vanishing walls, internal shells and self-intersections. Absence of a slicer warning is not full qualification. Inspect layer preview, body count and surface normals. Preserve CAD source geometry and export a mesh with only the tolerance required. A coarse STL can lose curvature; an unnecessarily dense file can slow processing without repairing a flawed design.

When to stop and contact service

Stop immediately if resin leaks outside the vat, the display or wiring is damaged, the Z mechanism binds with abnormal noise, or smoke, overheating or an electrical smell appears. Do not open safety-critical or warranty-sealed assemblies casually. Photographs, logs and exact resin and profile information help the maker's support team. Another trial is not justified when it risks additional screen, film, electrical or health damage.

A persistent XY defect with a clean vat and known-good film, the same Z shift across every object, recurring calibration failure or visible play in the platform and axis may require service-level inspection. Replace components only from evidence. Changing display, film and resin at once may hide the symptom, but it destroys the root-cause trail and introduces several new variables.

Washing and post-curing are separate failure sources

A part that looks sound on the platform can still be damaged during washing. Excessive solvent exposure, dirty wash liquid, unapproved ultrasonic treatment and rough handling of thin green walls may cause whitening, softening or cracks. External rinsing does not automatically flush a sealed interior. Follow the resin maker's solvent and time instructions, then let the part dry fully. Alcohol or water left on the surface can create staining and uneven cure under UV light.

Post-cure time, temperature and wavelength are material-specific. More light is not automatically better: excessive treatment may increase brittleness, colour shift and shrinkage, while inadequate cure may leave final properties below specification. Critical parts need the manufacturer's post-process validated as a complete system and should be measured only in the defined finished state. If a crack appears only after wash or cure, raising print exposure is not an evidence-based correction.

Material changes and a minimum workshop record

Resins with similar names or colours can differ in viscosity, pigment, shrinkage and required dose. Do not carry a profile across brands or colours without qualification. Record machine and firmware, vat and film, platform, resin lot and opening date, temperature, slicer and profile version, layer height, exposure, motion values, model position, first failed layer, photographs, the single change made and the result. Describe the observation rather than embedding a diagnosis: 'raft remains, model missing' preserves evidence better than 'bad exposure.' Repeated runs and a retained physical reference turn a lucky success into a controlled process.

Verified facts, manufacturer guidance and editorial conclusion

It is an established process principle that every release cycle applies mechanical load and that support and layer strength depend on geometry, material, temperature and exposure. Formlabs documentation distinguishes non-adherence, delamination, undeveloped features, cupping blowout and surface defects. Prusa guidance connects platform adhesion, orientation and hollowing to machine-specific procedures. These are strong primary sources for their systems, but they do not prove that one numerical value transfers across all MSLA machines.

Every precise 'ideal' exposure, speed increase, film lifetime or automatic-support guarantee remains a manufacturer claim unless resin, machine, temperature, geometry and test method are stated. LAYERTIMES concludes that most resin failures are solved faster by preserving evidence and following a branching symptom tree than by generally increasing settings. A successful print is not the same as a validated process; validation is repeatable, documented and safely reversible.

Practical sequence

For an empty platform, check retention, cleanliness, official calibration, material temperature, the correct profile and bottom exposure. When the raft remains, inspect first model layers, islands, contact tips and cross-section growth. For a horizontal crack, inspect normal exposure, lift behaviour, a common Z event and resin condition. For a fixed XY defect, inspect film, vat, display, optics and platform plane. For hollow models, ensure ventilation and drainage from the first sealed layer. For dimensional error, control orientation, exposure and the complete post-cure state.

Make one controlled change at each branch and record the outcome. Save a reference file and accepted sample with the final profile. Before printing, confirm that the vat is free of debris, its outside and the display are clean, the platform is secure, and resin is mixed and within its specified temperature range. Dry parts fully before post-curing and document flushing and drainage of cavities. Troubleshooting then becomes traceable quality assurance rather than a sequence of workshop superstitions.