Help with odd patterns in engraving

[Margaret Turco]

Hi all,

I am still working out my issues with engraving images. I realize I need to make changes to this one but my question is about the odd patterns in the larger areas of this engraving. I had the same thing happen to a photo I engraved in the same way about a month ago but now I want to solve this puzzle.

This is a 4.25" square ceramic tile I spray painted with black auto paint. Engraved grayscale in 3D mode using the Error Diffusion dithering in my print driver, power 15, speed 25, air assist on but without the cone.

I suspect that it has to do with using the spray paint rather than the laser but thought I would ask whether anyone else has run into this before. Thank you as always for your help!

[Mike Null]

Margaret

I can be of no help but why use the 3d mode?

[Margaret Turco]

3D seemed to give me better shades of gray.

[Dan Hintz]

Margaret,

I think you're trying to use competing algorithms. 3D mode changes the power of the laser at each spot it hits, which is great for materials that give you grayscale with changing power levels (marble is one example). Dither changes the location of the spots and simulates grayscale by changing the dot cluster density (less dots = lighter color, more dots = darker color), which is great for materials that only give one shade of color when hit with the laser (granite and glass are two such examples).

In the case of a painted ceramic tile, dither is the proper selection as changing the laser power will have no effect on perceived grayscale levels. Stay with the error diffusion as that will cut down on any patterns seen in large areas (it partially randomizes the dot pattern). If you still see this pattern, we need to look at something wrong hardware-wise.

[Tim Bateson]

Dan you explained it better then I ever have. I too use to make the mistake of using 3D on Ceramic because it would dig deeper in the black areas. However as Dan said this doesn't achieve gray scale on this material.

The best you can do - and it's pretty good is to simulate or create the allusion of gray scale by spreading the dots using dithering.

[Margaret Turco]

Thank you for the explanation guys! I'll give it a try, but I think I will have to reduce the power setting somewhat. Would you have a suggestion as to the PPI setting I should use? The picture was resampled at 300 DPI.

[Dan Hintz]

300 is good for this substrate, though make sure you are etching at the same resolution as your photo is resampled to. For example, don't resample the photo to 300 dpi if you're etching at 333 dpi (a common setting for ULS machines)... doing so will lead to patterns similar to what you're seeing (and this may very well be your issue).

When I resample, I always stick to what the Image Density settings offer me (75, 150, 250, 333, 500, or 1000 dpi). Anything else and I'm asking for trouble.

I also suggest running some sample grids to determine the min power necessary to break through the paint... tile and spray can paint is dirt cheap, so the $1 spent in test materials is worth it.

[Scott Shepherd]

Looks like a classic case of banding to me. A very common problem with lasers. Try cutting the speed down to 50-60% and see if the pattern goes away.

[Margaret Turco]

Ok, Dan, I understand what you are saying but I have never known specifically what the Image Density settings mean. I think a higher setting means more lines per inch but that's all I know.... I will play with this some more tomorrow.

Steve/Scott, I'm already running this at a verrrry slow speed, so I am thinking it's not the classic banding. The lower part of the tile looks checkered to me, and the upper right has sort of wavy diagonal banding.

Thank you again for your suggestions! I will let you know how it goes.

[Frank Corker]

That is a moire pattern. Have you tried putting the image through photograv.

[Margaret Turco]

I thought I would be able to get a good image without photograv, but maybe not in this case. I found a remove moire function in corel photo/paint, do you think that would help?

[Dan Hintz]

Steve, if her resampled image and settings match and she still has banding, then I would agree it's a hardware problem, as you've suggested.

Margaret, I'd suggest trying both (separately) and see if one works and the other doesn't (oops, just saw slowing it down doesn't resolve the issue... that points towards a resampling issue causing moiré patterns). The list of dpi's I gave in my last post correspond to Image Density settings 1-6 (write them down while you're thinking about it... I have no clue why ULS continues not to include this information on the driver screen, and there's simply no excuse for not including it in the manual!)

Even if you can get a better image in the short run using PhotoGrav, you should be able to resolve this problem without it... if you can't, you'll be constantly applying a band-aid rather than solving the original problem, and that always leads to a lack of quality in the final product. Any moiré pattern would be caused by the final application of error diffusion, so applying a filter in paint would be like adding more air to your car's tire before you run over the nail.