How to Preserve Transparency for Printing on Silver Vinyl (and Avoid Flattening Artifacts)

Question

I have an Illustrator file that contains a number of icons that are "Placed" transparent Photoshop files (i.e. they were never flattened so there is no white background in any of these placed icon files). The Illustrator file is being printed by a commercial printer on silver vinyl and then cut on a cutting machine. Before giving the file to the printer I have been instructed to do the Object => Flatten Transparency operation to make sure that no artifacts are created during this process. (The file must be flattened because the cutting machine requires that this be done.) My problem is that when I do this a white rectangle artifact is created around each of the icons--and obviously this is not what I want to see when I am printing on silver vinyl! (I have read that one way to ensure that these artifacts are not created would be to select the layer containing all the icons, and then select the layer underneath it, so that the icons are flattened "onto" something. But in this case, there is no "layer beneath!") Does anyone know of a type of "fake" layer that I could put under the icons that would make the Flattening process work--but not impact the look of the silver vinyl? Does anyone have any other type of solution? (I am new to Illustrator and am basically desperate to get this problem solved!) Thank you very much in advance.

Answer 1

They are 720 dpi images being printed on the silver vinyl by a Roland XJ640 inkjet printer
Knowing now that the images in question are, in fact, full-color raster images, the first thing I would ask the printer is this: "Since you are printing onto silver vinyl, how are you going to "back up" the color images with areas of white so that the process color renders correctly?" The references I looked at suggest this is a six-head inkjet printer using 4-color process (CMYK). That would make it a "process plus 2 (spot color)" device.
The lists of compatible inks I saw included a metallic silver and an opaque white. So that suggests two ways this might be being printed:
Depending on the resulting area of "background" silver in the finished pieces, it could be practical for the project to actually be printed on white vinyl, not silver, and the silver "background" would actually be one of the printed inks. Typically, to accomplish that in a program like Illustrator, you would enclose your full-color raster images in a clipping path of the desired shape, draw a background shape behind the clipped raster images which will bleed beyond the eventual cut paths, and fill that background shape with a spot color Swatch which is named so as to be recognized by the software driving the device as its silver ink.
If the vinyl is, in fact, silver, then a white object needs to be printed first, and the CMYK inks printed on top of it. Typically, to accomplish that in a program like Illustrator, you would draw shapes surrounding your images in the desired shapes, and fill those shapes with a spot color Swatch which is named so as to be correctly recognized by the software driving the device. It is usually more convenient to position these objects in front of the CMYK images and set them to overprint. Or, if they are positioned behind the images, then the images themselves need to be set to overprint or the CMYK inks could be set to overprint in the Separations section of the Print dialog of whatever software is actually doing the printing.
The icons on the globe pattern are very small 1440 dpi placed Photoshop files....printing at 1440 dpi with bicubic interpolation...
This is a common misconception about printer resolution. Knowing that a given inkjet printer has a hardware resolution of 1440 dpi does not mean that maximum detail is going to be achieved if the full-color raster images are 1440 PPI.
1440 DPI is not the "extremely high resolution" it sounds like to some. To understand, consider normal commerical offset printing: A typical PostScript imagesetter has a hardware resolution of 3000 SPI (printer Spots Per Inch). And compared to the relatively "fuzzy" dots that are created by an inkjet device effectively "spraying" droplets of liquid ink on a sheet, the printer spots of an imagesetter are sharp-edged dots made by exposing a photo emulsion on film with a laser. And yet, designers wanting maximum fidelity and detail when printing full-color raster images do not go around creating their raster images at 3000 PPI, do they? Why not?
Here's the deal: Whether you are talking about inkjet composite printing or imagesetting, printing full-color images involves process color. "Process color" means you use little tiny dots of (usually) just four primary colors of inks to simulate all the colors in the image. In order for those little tiny dots to simulate other colors, they must do one of two things: They have to either vary in size (halftone screening), or they have to vary in number (stochastic screening) across a given area.
Common commercial offset printing employs the first method, halftone screening. Remember: the resolution of the imagesetter is 3000 SPI. In other words, that imagesetter just does one thing: It creates spots which measure 1/3000 inch. That's a fixed function of the mechanical hardware; the imagesetter cannot vary the size of the spots it creates. So how does it print the variable-size dots required by halftone screening? Here's how: Each of the variable-size halftone dots (the dots of ink that will actually be printed) are built up out of the much smaller printer spots.
This raises the question, how many printer spots are available to build up the shape of each halftone dot? That's where the halftone ruling (expressed in LPI; Lines Per Inch) comes in. And that's the value (halftone LPI) that you consider when determining best PPI for your images. The rule-of-thumb is one-to-two times the halftone ruling. For example, 175 LPI is considered above-average printing in offset lithography. It's one step above the most commonly used "magazine quality" of 150 LPI. So assuming a project being printed on quality coated stock at a halftone ruling of 175 LPI, appropriate raster image PPI would be 175-to-350. Using more is just needlessly wasting data that is going to effectively be "averaged away" anyway ("oversampling"). It can, in fact, degrade the printed results. Note that that's a far cry from 3000, the hardware resolution of the imaging device.
Composite inkjet printing nowadays commonly employs the second method, stochastic screening (also called FM screening, diffusion dither, etc.). Nowadays, 1440 is a common hardware resolution for even cheap desktop inkjet printers. Again, similar to the imagesetter described above, this means that the inkjet printer does just one thing: it bursts onto the substrate little fuzzy dots of ink which measure (roughly) 1/1440 inch. That's fixed; the inkjet printer cannot vary the size of the spots it creates. Note that this is actually less than one fourth the number of printer spots created by a typical imagesetter. So how does an inkjet render all the colors required to reasonably reproduce a full-color image? It prints fewer printer spots of each of its primary colors within a given area. And to avoid a discernable (and distracting) pattern, it distributes those printer spots in what looks like, to a human, a random pattern.
So once again, there is no 1:1 correspondence between each printer spot produced by an inkjet printer and each pixel in a full-color raster image. Trying to achieve that correspondence by making your image PPI equal to the SPI of the printer is just as silly in inkjet printing as it is in imagesetting. It's still just needlessly gross oversampling with absolutely no advantage.
In fact, one of the much-hoped-for promises of stochastic screening in the early days of desktop prepress was that it reduces the PPI needed for apparent sharpness in printed results. (In those days of low memory and slow processing, file size was of paramount concern.) That is, for "comparable detail and sharpness" stochastic screening required lower PPI.
What all this means is: You are not going to achieve any visible detail advantage by sampling your full-color raster images at 1440 PPI instead of at a much lower PPI. Ridiculously high PPI will not make a screened (halftone or stochastic) image include more detail when viewed under magnification. Increasing the PPI of your raster images to the moon does not overcome the rendering capability of the printing process itself.
JET