A Practical Walkthrough Of The Method
I first encountered this workflow around 2019 when a client sent me files that needed to go through a specific reproduction process. The instructions they referenced were vague and full of jargon, so I spent about two days reverse-engineering the steps from a combination of printer notes and test prints. What follows is what actually works in practice. At its core this is about managing halftone patterns and dot structure when reproducing artwork that relies on cross-hatched line work. The challenge is that cross-hatching already simulates tone through overlapping lines, and running it through a standard halftone screen creates a moiré mess or turns clean linework into a muddy gray blob. The trick is knowing when to treat it as line art versus tone art, and adjusting your dot geometry accordingly. The workflow I use starts with the source file. If you have a vector or high-resolution raster with clear black linework at 100% K, you do not run it through a halftone filter. You set it to line mode at 85 to 120 lpi depending on your substrate. Paper stock matters a lot here. Uncoated newsprint will spread dot gain significantly, so bump your screen to 100 lpi minimum and compensate the gamma curve by roughly 1.15. Coated stock lets you drop down to 85 lpi without losing detail.
When the artwork has actual tonal areas alongside the crosshatch lines, things get more complicated. I separate the image into two channels. The line layer stays in line mode. The tonal layer goes through halftone screening at whatever lpi makes sense for the piece. Then I composite them back together. This takes maybe ten minutes once you have the workflow dialed in, compared to the hour it used to take me doing it manually in Photoshop before I understood what was going wrong. One thing people routinely miss is the angle selection. Standard practice puts the line screen at 45 degrees and the tone screen at something like 15 or 75. But when you have crosshatching that is already at an angle, rotating the tone screen can still create interference. I had a job last year where a scanned drawing had crosshatching at roughly 30 degrees, and no matter what tone screen angle I chose, a visible moiré pattern appeared in the midtones. The fix was to slightly detune the line frequency from the tone frequency by about 3 percent. So instead of running lines at exactly 100 lpi and tones at 65 lpi, I went 100 and 62. The moiré disappeared and the detail stayed intact. Another counter-intuitive point: higher resolution input does not always equal better output here. If your source is 600 DPI and you blow it up to line mode at 120 lpi, you are just giving the RIP more data to interpolate. A 300 DPI source often produces cleaner results because there is less digital noise in the halftone conversion. I usually downscale to 300 or 350 DPI before running anything through the screening process.
If you are working with spot color separates rather than CMYK, skip the halftone approach entirely and go straight to vector-based line art wherever possible. Spot color reproduction with crosshatched originals tends to show every scanning artifact, and no amount of screening will hide that. In those cases, recreating the crosshatch geometry as vectors in Illustrator gives you infinitely cleaner results and cuts proofing time down to almost nothing. The whole process from file preparation to final proof runs about 20 to 30 minutes on a modern machine. If it is taking you longer, you are likely either overthinking the dot geometry or working from a source file that needs cleanup before you even start screening. Scan or digitize at the highest practical resolution, check the original for any smudges or artifacts, clean those up, then proceed with the separation method outlined above.
Get the Full Details
