What You Must Know About Bidirectional Printing — Mechanical Issues (EP.3) ARTJET 2026

The quality difference between bidirectional and unidirectional printing will be explained at the end of this article.

First, let’s describe the symptom.

• Even accounting for the fact that bidirectional printing produces lower quality than unidirectional,
• the print quality shown in the photo below looked noticeably rougher than usual.
• It’s not immediately obvious from a distance,
• but anyone experienced in print production would take one look and think: “That looks rough.”
• It’s difficult to show in a photo, so we’ve zoomed in on the right side. The background behind the bear character has a slightly rough texture.

A brief note on where bidirectional Offset values are configured by carriage speed in the printer software.

• Head carriage speed and flatbed feed speed each have three settings: Slow, Normal, and Fast.
• Our printer uses Normal as the default.
• 

The print conditions for the image above were as follows.

• Head carriage X-axis speed: Normal
• Flatbed Y-axis feed speed: Normal
• Print direction: Bidirectional (Double direction scan)
• G-mask mode: Quality mode — 3rd level G-mask used as default (out of 4 levels) to reduce banding
• Front-side printing: White printed first, then Color printed on top

Theory 1. Different conditions between front-side and back-side printing?

• Could the difference in Mask logic (the pattern in which dots are laid down) between front-side and back-side printing cause the rough output?
• We prepared a front-side sample (White first, then Color on top) and a back-side sample (Color first, then White on top) using the same image.
• Both samples printed rough — no significant quality difference between them.
• Front-side vs. back-side Mask was not the cause.

Theory 2. White ink density causing the quality issue?

• Our printer can print White at up to 600% density relative to Color.
• Up to 300%, the White layer forms evenly. At 600%, too much ink is applied and the White surface becomes uneven and rough.
• We reduced White density to 200% — the result was the same.
• White ink density was not the cause.

Theory 3. Acrylic thickness variation causing the issue?

• We measured the thickness at all four corners of the acrylic: 1.51mm, 1.55mm, and 1.58mm — some variation existed.
• Bidirectional print roughness typically occurs when the standoff differs by 1–2mm from the calibration height, so this variation was within acceptable range.
• Acrylic thickness variation was not the cause.

Theory 4. Would lowering the head height help?

• With acrylic thickness at 1.5mm, the Z-axis head height was set to 1.5mm for printing.
• We tried 1.4mm (0.1mm lower) — the problem remained.
• We tried 1.6mm (0.1mm higher) — still no improvement.
• Fine-tuning the head height was not the cause either.

A puzzling observation

• “It prints fine on pet film… so why does the image come out rough on acrylic?”
• Printing on pet film resolved the roughness issue immediately.
• We started thinking about what was different between printing on acrylic vs. pet film.
• Going back to basics, we examined the acrylic prints under a 10x magnifier — and found that at all three Z-axis heights (1.4mm, 1.5mm, 1.6mm), the text appeared doubled, to varying degrees.
• Normally, a smaller head-to-substrate gap produces better bidirectional quality — but in this case, 1.6mm actually looked cleaner than 1.4mm.
• On a hunch, we tried 1.7mm — it looked even cleaner. At 1.8mm, print quality improved as shown in the photo below.
• Important: the improvement was not because the head height was raised. The real cause is explained below.

How bidirectional Offset calibration works

• The bidirectional Offset value must be recalibrated whenever either of two variables changes.
• The first is X-axis carriage speed. An Offset calibrated at Normal speed cannot be applied at Fast speed — because the ink landing point shifts with speed.
• The second is Standoff (the gap between the substrate and the head). As shown in the diagram below, the ink landing point differs between head height 1 and head height 2.
• For example: if the bidirectional Offset was calibrated at a Standoff of 1.2mm, but the actual print is done at 1.5mm, the landing points won’t align — resulting in doubled text or rough output.

Confirming whether the Standoff matches the calibration height

• Bidirectional Offset is typically calibrated at a Standoff of 1.2mm.
• If the actual print Standoff is 1.2mm, the image renders cleanly.
• If the Standoff differs from 1.2mm, the ink landing point shifts — causing rough output.

Why the 1.5mm acrylic produced rough output:

• The original bidirectional Offset was calibrated at a Standoff of 1.5mm — not the standard 1.2mm.
• When printing on the acrylic, the head was set 1.2mm above the substrate surface — which did not match the calibration Standoff, causing the Offset to be applied incorrectly.
• The reason quality improved at 1.8mm was that it brought the total Standoff close to 1.5mm (1.5mm substrate + ~0.3mm gap) — matching the original calibration condition.

Field Tip

• Diagnosing the issue by adjusting head height in small increments takes a lot of time.
• A much faster solution: recalibrate the bidirectional Offset directly on the actual substrate you intend to print on.
• If text appears doubled under a 10x magnifier or the image looks rough during bidirectional printing, recalibrate the bidirectional Offset on that substrate before starting the actual job.

Quality limitations of bidirectional printing

• Bidirectional printing is 70–80% faster than unidirectional, but always operate with the understanding that it has inherent quality limitations.
• Flatbed flatness variation can cause print quality to differ from zone to zone.
• A Standoff variation of up to 0.1mm makes little difference — but beyond 0.1mm, the Offset value becomes misaligned, causing rough output or doubled text under magnification.