Cut & Proof  /  Notes

Paper grain direction, and why register drifts

If your registration is good on the first pass and drifting by the fourth, the paper is probably the culprit — and it is drifting far more in one direction than the other.

Paper is made of fibres, and in machine-made paper those fibres mostly point the same way. That direction is the grain, and it is the reason a sheet does not change size evenly when it takes on or loses moisture.

The physics, briefly

A paper fibre swells in width, not in length. Take on moisture and each fibre gets fatter without getting appreciably longer. So a sheet grows mostly in the direction that runs across the fibres, and barely at all along them.

The size of the difference is the useful part. Trade figures for a swing from 50% to 80% relative humidity put the dimensional change at roughly 0.25% across the grain and 0.06% along it.

Read this figure carefully

Those numbers are for machine-made offset stock, not damped washi, and a printmaker soaking a sheet is doing something far more drastic than a humidity swing. Do not take them as a number to work to.

What they do establish is the ratio: expansion is roughly four times greater across the grain than along it. That is the part that matters, and it is why applying one correction figure in both directions is worse than useless.

Put concretely: on a 400 mm dimension, a quarter of a percent is a millimetre. On a four-block print that is the difference between clean register and a visible colour fringe along one edge — and the fringe will be on two edges, not four, which is often the clue that grain is what you are looking at.

Which way should the grain point?

Long grain — grain running along the longer dimension of the sheet — is generally preferred for multi-colour work. The reasoning is direct: the greater expansion happens across the grain, so with long grain that larger movement falls in the shorter dimension of the sheet, where the same percentage produces a smaller absolute error.

There is a second, subtler point worth having. Register does not always matter equally in both directions. If your image has a strong horizontal structure — a horizon, a band of colour running across — a small vertical shift may be far more visible than a horizontal one. If so, orient the grain so the larger movement falls in the direction where a shift shows least.

Finding the grain in a sheet you already have

Good paper is often marked, and mould-made and handmade sheets have much less pronounced grain than machine-made ones. But when nothing is marked, there are four field tests. Use a scrap, and use more than one test.

  1. Tear. The most reliable. A tear runs relatively straight along the grain and goes jagged and wandering across it.
  2. Fold. A fold along the grain is cleaner and crisper. Folding across the grain cracks and resists.
  3. Bend. Bend the sheet gently in both directions without creasing. It gives more easily — offers less resistance to the curve — along the grain.
  4. Moisture. Damp one side of a small square and it will curl, with the axis of the curl running along the grain. Conclusive but destroys the sample.

Once you have found it, mark it on the wrapper of the whole pack. You only want to do this once.

What this does and does not fix

Grain direction is one variable among several. Damping technique, how long the sheets sit, the room's humidity, and the pressure of the press all move paper about too. Getting the grain right does not remove drift; it makes the drift smaller and, more usefully, makes it predictable enough to plan around.

The practical version: damp your paper consistently, keep the whole edition together so every sheet has had the same treatment, and orient the grain so the larger movement lands where it will be least visible.

The tool for this

Kento includes a paper calculator that takes grain direction into account, and deliberately refuses to apply the same expansion figure in both directions — because the paper does not.

Open Kento →

Where this comes from

  1. PrintWiki: Grain — the physics: fibres expand in width rather than length, and the preference for long grain in multi-colour work.
  2. British Letterpress, "Machine or grain direction in paper" — confirms it matters most for close register and colour work, and makes the point about one direction mattering less than the other.
  3. Welbound, "Know your grains" — the four field tests, and confirmation that a tear runs straight along the grain and jagged across it.
  4. The 0.25% / 0.06% figures are trade figures for machine-made offset stock. Treat as an order-of-magnitude sanity check, not a working number.