Four miniatures went on the plate. The three at the back are clean. The one at the front left is a stump with a raft and a forest of supports attached to nothing — and it was a different model last time, in the same place. When resin prints fail in the same spot on the plate, print after print, file after file, the slicer is not the thing that is wrong. A repeatable, position-locked failure is a measurement of your machine, and it is one of the few resin problems that will tell you its own cause if you ask it in the right order.
This post contains affiliate links. If you purchase through these links, 3D Prototyping Hub may earn a small commission at no extra cost to you.
This post also contains Amazon affiliate links. As an Amazon Associate we earn from qualifying purchases.
First, Establish What the "Same Spot" Is
Three different things get described with the same sentence, and they have nothing to do with each other.
| What repeats |
What it points at |
| The same X/Y area of the plate, whatever you print there |
Exposure uniformity: tram, panel, film |
| The same height in Z, wherever the part sits |
Lift and peel, resin level, a Z-axis fault |
| The same feature on the model, wherever you put it |
Supports and geometry — not a machine fault |
The third is the most common and the least interesting here: a thin blade or an unsupported island fails because of how it was prepared, and moving it changes nothing. That is the subject of why resin prints fail — if your failure follows the model rather than the plate, start there instead.
The rest of this is the first row, because a failure fixed to a region of the plate means the machine is delivering something different to that region than to the rest.
Print the Map Before You Change a Setting
Do not adjust anything yet. Make the machine draw you a picture of itself first.
Take one small part — a 20mm test object, a single mini, anything with fine detail and a couple of thin features. Duplicate it into a grid across the entire plate: three by three on a small machine, four by four or more on a large one, spaced out to the corners, same orientation, same height, identical supports. Every slicer has a duplicate-and-arrange tool for exactly this. Print it once at your normal settings.
Then score every copy on the same three things: did it adhere, are the thin features present, are the fine details sharp or soft. Write the scores on a sketch of the plate rather than trying to hold nine results in your head.
Two rules make the map worth printing:
- Change one variable only. Same resin, same bottle, same layer height, same room. Diagnosing on a single bottle of plain grey model resin is the cheapest control you will ever run, and opaque grey is far easier to read a half-cured surface on than a translucent colour.
- Shake the bottle properly first. Pigment and additives settle, so resin poured straight off the shelf is not the same at the top of the vat as at the bottom. Two minutes of agitation removes a variable that otherwise looks exactly like a machine fault.
Read the Pattern, Not the One Bad Tile
The shape of the weak region is most of the diagnosis.
- One corner weak, the opposite corner strong. A tram problem almost every time. The plate is not parallel to the panel, so the first layer is thicker on one side and thinner on the other.
- A whole edge or one half weak. Also usually tram, but if the plate checks out it points at the light source — the LED array and its diffuser are not delivering evenly across that half.
- Edges weak, centre strong, roughly symmetrical. Normal optical falloff, exaggerated. Every MSLA light engine is brightest in the middle; a tired panel widens the gap until the corners fall below the exposure your resin needs.
- A small, sharply bounded patch that does not move. That is the panel surface. A cured skin of resin, a spill under the protector or a fingerprint masks a region with hard edges, which no other cause produces.
- Centre weak, edges fine. Look at the film before the panel — this is more often peel than light. The middle of the vat is where a stretched film sags and where suction forces are highest.
Tram Is the Cheapest Thing It Could Be
Start here regardless, because it costs nothing and it invalidates every other test if it is wrong.
With the vat out and the plate clean, loosen the clamp, let the plate sit flat against the panel under firm, even pressure, and tighten in a cross pattern rather than one screw at a time. Use whatever paper or shim your manufacturer specifies and check that the resistance feels the same at all four corners — not similar, the same. Then check the two things people skip: that the plate is not bowed from being levered off with a scraper, and that there is no cured resin on its underside, on the panel protector or in the clamp threads. A flake of cured resin under one corner tilts the whole plate, and it looks exactly like a failing screen.
Then print the map again before concluding anything.
The Panel Is a Consumable Light Source
If the map still shows a region, look at the light.
Run the firmware's screen or pixel test with the vat removed, and photograph the panel with your phone rather than looking at it directly — a 405nm panel at full output is not something to put your eyes on, and the camera reads dim patches better anyway. Dead rows, blocked zones and a cured film of resin all show clearly in the photograph.
Panels age. Manufacturers quote monochrome LCDs in the low thousands of hours, and the output falls gradually rather than failing outright, so the honest symptom is "it needs more exposure than it used to" long before anything looks broken. Because most plates do most of their work in the middle, that ageing is not even across the area.
Two habits are worth starting the day you own the machine rather than after the panel has gone. Fit a screen protector film cut for your model — a spill that cures onto the panel is permanent local under-exposure, while one that cures onto a sacrificial film costs a few dollars. And clean spills while they are still liquid, printer off and gloves on: isopropyl on a lens cloth, never a scraper or a blade on the panel surface.
The Film and the Vat Add a Gradient of Their Own
A film that has done a few hundred hours is not flat. It stretches, sits lower in the middle under the weight of the resin, and changes both the first-layer gap and the peel force depending on where you are on the plate. That reads as a weak or lost centre rather than a weak edge, and it is often misread as a dying panel.
Check it by pressing a fingertip against the film from below, with the vat out and empty: it should feel like a drum across the whole area, with no slack pocket in the middle. Cloudiness, dimples and a permanent dish are all replacement signals. Change the film as a set with its gasket and screws rather than re-tensioning one that has taken a set — why resin vat film fails covers the fitting and the habits that double its life. A spare FEP film on the shelf stops a diagnosis stalling for a week.
Exposure Is the Last Number You Touch, Not the First
The tempting fix for a weak corner is more exposure, and it is the one move that makes the problem permanent.
Raise exposure globally and the weak region comes up to barely acceptable while everything else is over-cured: fine detail bloats, gaps between features close, the first layers spread into an elephant foot, and supports fuse hard enough that removing them chips the part. The plate is now uniformly mediocre rather than locally bad, and the evidence you needed is gone.
Tune exposure only after the map prints evenly — then run an exposure ladder for the resin you actually use, remembering that a cold room slows the cure. Most resins specify a working range in the low twenties of degrees Celsius, and a vat beside a draughty window in October can sit below it on one side. If the gradient tracks the room rather than the plate, heat the room. Running the ladder off a single bulk bottle such as Chitu Systems' Conjure standard resin keeps every tile on the same batch, which is the whole point of a ladder.
One thing exposure cannot change: the pixel pitch of the panel, which sets the smallest feature the machine can resolve anywhere on the plate. If detail is soft everywhere and always has been, that is resolution rather than uniformity — why resin prints come out the wrong size is the better post for it.
Replace the Panel, or Replace the Printer?
When the map says panel and the panel is old, price the repair honestly: the panel for your exact model, plus a film, plus a gasket, plus an evening of re-tramming. On a current mid-range machine that is a routine repair and clearly worth doing.
On an older budget machine the same total often lands within sight of a new printer. ELEGOO's US store and Anycubic's US store both publish panel size, resolution and plate dimensions per model, and both list what is sold as a spare for their own machines — check that before assuming a panel exists for a model three generations old. The current range is in our resin printer buyer guide. Buy on pixel pitch and plate area rather than quoted speed: speed is a setting, resolution is not.
When the Deadline Decides Instead
Everything above is an evening per step, and the steps have to run in order — fine for a hobby plate, expensive when the part is promised to somebody. The honest cut-off is the third failed plate on a deadline. A bureau running calibrated exposure on a maintained machine is selling the thing you are trying to rebuild, which is repeatability, and SLA and resin printing services quote single parts as readily as batches. The providers in our directory are listed by location and process, so the part gets made while you fix the machine properly rather than under pressure.
Hero photograph by Azeez Maree via Unsplash.
