Ink Friendly

Printer tests · Print-true

Printer alignment test page

Vernier offset ladders for both axes, a registration grid at a 20 mm pitch, and a duplex front-and-back pair for checking that the second side lands where the first one did. Nine rungs, a symmetric series centred on a labelled zero, drawn at true millimeter size in your browser. No watermark, no sign-up, and nothing you do here leaves your device.

SheetLetter 8.5×11 in Patternladders + registration Step0.20 mm · ±0.80 mm Est. ink0.5% Pages1 sheet

Print at 100% or "Actual size" — at any other scale the rung labels are wrong by the same percentage. Steps below 0.1 mm are refused rather than drawn, because the ticks themselves are 0.08 mm wide.

How this alignment test page works

The measurement is a vernier. Each ladder draws nine rungs; every rung is two rows of ticks, the first at the nominal position and the second displaced by that rung's offset, and the offsets form a symmetric arithmetic series centred on zero — at the default 0.2 mm step that is −0.80, −0.60, −0.40, −0.20, 0.00, +0.20, +0.40, +0.60, +0.80 mm, each printed beside its rung. You find the rung where the two rows read as one continuous set of ticks, and its label is the displacement between those two rows as they landed on your paper — which is a smaller claim than it sounds, and the limits paragraph below spends most of its length on the difference. There is always a zero rung and it is always labelled, because a ladder with no zero cannot tell you that nothing is wrong. The left ladder displaces horizontally along the carriage axis; the right one displaces vertically along the paper feed. The registration grid underneath is a lattice of crosses at exactly 20 mm, with corner marks at the grid's own corners, for judging skew and drift against the paper's edges.

Everything is SVG geometry in real millimeters and the page declares your paper size to the browser at print time, so nothing in the pipeline has a reason to rescale. That matters more here than anywhere else on this site: the sheet's output is a number you read off a label, so a dialog silently printing at 94% would hand you a number that is wrong by 6% while looking perfectly correct. Two reference rules print near the foot of every sheet — one exactly 100 mm, one exactly 4 inches — so a ruler settles it before you read anything else. The marks are drawn in sRGB values, not CMYK ink: this sheet is deliberately almost all black, because alignment is a geometry question and adding color to it would only add variables. The "Est. ink" figure is measured from the geometry as it is drawn, which is why the ladder-only mode costs a fraction of the full sheet.

The honest limits, and the largest one first: this ladder cannot separate a correctly aligned head from a misaligned one. Your printer's own alignment utility reads a head offset because it lays the two halves of each rung down in different carriage passes, or from different heads — the thing you judge there is the offset between the passes. This page cannot do that. It hands your driver a single document, both rows of every rung are drawn in the same ink at the same moment, and which pass or head puts them on the paper is firmware's decision, invisible from a browser. So on a sheet laid down in one pass the 0.00 rung is the one that reads continuous whatever the head is doing, and a rung label is what it says it is — the displacement between two printed rows — not a number to enter in a driver. That number comes from the utility that can control the passes, which is your printer's, not this page's. What this sheet is good for is the part it does control: a fixed set of known displacements at true size, so you can see the finest one your printer resolves at all, so a rung other than 0.00 reading continuous tells you something is moving the marks even though the sheet cannot say what, and so the sheet you print after an adjustment is genuinely comparable with the one you printed before it. It also refuses rather than pretending: an offset step below 0.1 mm would displace a 0.08 mm tick by less than its own width, so the page declines and draws nothing instead of an unreadable ladder, and a step above 2 mm is clamped with a note. On color the limit is absolute — the page cannot measure what reached the paper, because a browser has no instrument on the output side and reading a printed mark means putting a colorimeter or a spectrophotometer on it. And all of it rests on the actual-size precondition: at anything other than 100% ("Actual size") the printed offsets are not the labelled offsets.

Frequently asked questions

How do I read the vernier alignment ladders?
Each ladder has nine rungs and each rung is a pair of tick rows: the first row sits at the nominal position, the second is displaced by that rung's labelled offset. Look for the rung whose two rows read as one continuous set of ticks with no step between them — its label is the displacement between those two rows as they landed on the paper. Read that literally, because it is not your printer's head offset: both rows are drawn in the same ink in one document, this page cannot choose which carriage pass or which head prints each of them, and on a sheet laid down in a single pass the 0.00 rung is the one that reads continuous whether the head is aligned or not. So take the ladder as a fixed reference rather than a verdict — it shows the finest displacement your printer resolves, and if some rung other than 0.00 reads continuous then something is moving the marks, though the sheet cannot tell you what. The left ladder displaces side to side along the carriage axis; the right one displaces along the paper feed. Reading an alignment rather than measuring a gap is deliberate: people judge "is that one line or two" far more reliably than they judge a fraction of a millimeter.
What is the duplex front and back option for?
It prints two pages carrying the same registration grid. Run them through as a double-sided job, then hold the sheet up to a light: if the crosses on the back sit exactly over the crosses on the front, the two passes agree. If they are consistently displaced in one direction, the offset you can measure against the 20 mm grid pitch is your duplex registration error, which is a different fault from head alignment and is usually adjusted somewhere else in the driver. This is the pattern that needs two pages, which is why the sheet count in the readout changes when you choose it.
Why does the page refuse a very fine offset step?
Because below about 0.1 mm the rungs stop being distinguishable on paper. The ticks themselves are drawn 0.08 mm wide, so an offset step finer than that would displace a line by less than its own width, and every rung would look identical — a ladder you cannot read is worse than no ladder, because it still looks like a measurement. Ask for a step under 0.1 mm and the page refuses in words and draws nothing rather than printing something unreadable. Above 2 mm it clamps instead, and says so: a displacement that large is visible without a test sheet.
How do I print the alignment sheet at the right size?
Set the print dialog's scale to 100% (or "Actual size"). This sheet is more sensitive to scaling than most: at 94% a rung labelled 0.20 mm is printing a 0.19 mm offset, so the number you read off would be wrong by the same few percent the dialog silently applied. Every sheet prints two reference rules near the foot — one exactly 100 mm, one exactly 4 inches — so check those with a ruler first. "Save as PDF" as the destination writes the file with your browser, on your device, at whatever scale the dialog is set to.
Will this fix my alignment, or tell me my colors are right?
Neither, and it does not hand you a number to type into a driver either. Head alignment is measured and adjusted by your printer's own utility, which can lay the two halves of a pattern down in different carriage passes; this page hands your driver one document and cannot, so a rung label here is the displacement between two rows on paper rather than a head offset. On color it is even more limited: the marks are sRGB values handed to your browser, this page never sees the paper, and it cannot measure printed color at all — that needs a colorimeter or a spectrophotometer held against the sheet. What it can do is give you a fixed, repeatable reference at true size, so the sheet you print after an adjustment is genuinely comparable with the one you printed before it.

Related tools