Acrylic Inlay Knotwork Butterflies

[evermorian]

This is an acrylic inlay experiment with one of my knot work designs. Bottle cap for scale in one shot. They are about 2.5" by 2.25". These are intended to be pins but, my backs disappeared. I am waiting on more to arrive. They are possibly too plastic-y for anyone to want to actually wear. Maybe beveling the edges would help?

The outer frames are frosted acrylic in “cherry blossom” and “kiwi,” while the inlay are just translucent (all from Inventables).

I mirrored the inlay pieces and, cut them from the back in order to try to get a good fit (adjusting for kerf and all). The difference between a tight fit and cracking the frame seems to be about .001”. There are some gaps that show up in the pictures. They are not particularly noticeable in person (being quite small).

It all friction fit, except for three tiny pieces total (some of the smallest). So, I doused them with acrylic cement – the thin stuff so it would wick into the seams. That firmed it all up nicely.

I haven’t annealed these but, still might.

Suggestions for improving the fit, removing the gaps or, anything else about the process are welcome!

[dklgood]

Beautiful.

[evansd2]

Other than inlay/kerf correction tricks like mirroring one piece to improve fit and removing all sharp points from the design (replace with small radii or curves), you could also back the entire piece with black cardstock or something akin to it. This would stop light from coming through the gaps and reduce their appearance.

What kerf correction method did you choose?

[Jules]

Great design!

[rbtdanforth]

Very beautiful pieces that work very well, there is a basic problem in making small pieces that the parts keep getting smaller but the kerf does not and so stands out more the smaller you get. Even variations in speed coming around corners can be significant in kerf width as the scale gets smaller and I have not been able to match Evansd2 long before that.

I am wondering if a filler material might fill the difference, or for that matter a pourable acrylic that would be a perfect match no matter how fine the detail. It would not solve the issue with wood but might in the case with acrylic. I did make some earrings where the interior was spiderweb thickness in much of it but it was engraved and left open, Perhaps a pourable acrylic would work in that case.

[pubultrastar]

Those look great. I’d love to hear what people have to say for suggestions on Kerf corrections, as this is something I’d love to get better at. The designs are pretty awesome.

[evermorian]

What kerf correction method did you choose?

I made the inlay pieces half a kerf larger (.003") with an outer contour (in Corel Draw) and, made the openings half a kerf smaller. .007" was too tight. It seems like .0065" would have been about perfect but, that gets rounded.

The inlay pieces but, not the frame were mirrored. My expectation was that this would compensate somewhat for the expected slight angling of the cuts. During assembly, the pieces will pretty much only go in from one side.

I didn’t think about the dwell time at the sharp corners you and @rbtdanforth mention. So, that might be worth another pass.

you could also back the entire piece with black cardstock

I would hate to lose the translucence.

[evermorian]

I am wondering if a filler material might fill the difference, or for that matter a pourable acrylic

That sounds like a promising approach!

[evansd2]

I made the inlay pieces half a kerf larger

If you mean you scaled the piece to be 1/2 kerf wider, that’s going to be an issue. You need to offset your path by a kerf amount. Scaling and offesetting don’t mean the same thing when operating at this level of detail and precision.

[evansd2]

The inlay pieces but, not the frame were mirrored. My expectation was that this would compensate somewhat for the expected slight angling of the cuts. During assembly, the pieces will pretty much only go in from one side.

This is exactly correct.

[evermorian]

I think I did what you are describing. e.g.:

Edit: I know you are an Inkscape user and, this is Corel Draw. So, might have different terminology.

[evansd2]

Yeah, that’s why I couched it in a big if statement That seems like offset to me.

It’s possible that the strain you’re putting on with the tight friction fit is causing flex that makes the fit go a bit wonky. Try backing off a bit (a thousandth or three?) for a very loose friction fit as a test maybe? I find that if I go too aggressive with my kerf settings on inlaid tray edges the corners pull away from each other as the wood strains to fit together.

[evermorian]

Ah! Thanks! I will give that a try.

[GrooveStranger]

Those look good but perfection should be possible. But I am stumped so far. How you adjusted the path does look correct to me.

It is also possible, I guess, that Corel’s function just isn’t very good but I would be surprised if that were the case. The answer must be out there somewhere.

This is how I do “Offset Path” in Illustrator. I assume the “Corner” you have selected is the same as “Miter join” in Illustrator and that is one of the most important settings.

Maybe the stairstep is the “Miter limit” and increasing it will help?

Hmm. When you do this and zoom in, does the new path look right to you? It is possible this is some kind of cut problem, not a file prep problem.

[evermorian]

Maybe the stairstep is the “Miter limit” and increasing it will help?

The stair icon is for the number of contours. It’s not an exact match for Illustrator’s Offset Path. Corners can be mitered, rounded or beveled. For Corel Draw, I think using mitered corners is equivalent to automatically setting Miter Limit to the maximum in Illustrator.

This does seem to be the way to do it in Corel Draw, according to the tutorial Jules shared some time ago ( CorelDRAW - Adjusting for Kerf in your Design ).

If I am understanding Miter Limit correctly, I don’t think it would relevant here. The obvious gaps are mostly along (relatively) long curves.

[GrooveStranger]

The obvious gaps are mostly along (relatively) long curves.

Yah, good point. No idea what the issue is.

[evansd2]

Ah now that I’m reminded of this thread I bet this would apply:

Improving kerf adjustment on complex shapes

This is an inkscape-specific thing, but may apply elsewhere. Kerf adjustment is well-trod territory at this point, but I recently decided to see if there was anything to improve performance. The issue: when doing a “stroke-to-path” action on an irregular shape, you can sometimes see deformation of the resulting paths. The solution: add more nodes. The software is making a best-guess estimate, which you can improve with extra nodes. Illustrated: On the left, the original shape. On the righ…

[evermorian]

Thanks!

I did try that approach with some spiral inlays but, never quite got a result with which I was happy before I got interrupted with something else. I never did go back and try it with the irregular shapes in the butterflies. I should do that.