Pinholing problem

I’d probably put down a layer of sacrificial material underneath to reduce flashback. And probably reduce both the power and speed.

1 Like

What @tim1724 said, especially about reducing the speed.

You get overburn because the speed changes a lot in the corners. If you’re moving slowly already, then there is less of an effect.

2 Likes

Okay. Thanks! We’re using some thin chipboard to prevent flashback on the thin stuff, and that’s definitely helping. Sounds like we need to slow down, decrease power, and try multiple passes.

1 Like

I never used more than 200 speed on anything thin. Even then I could see the corner issues but it was as slow as I was willing to go. Still only did single pass though

2 Likes

I was going to suggest around 150 speed and find the necessary power.

Going faster definitely won’t help, it only means it has decelerate more and depending upon how intricate the designs are, it can lead to different results in very intricate areas vs non-intricate/straightaway areas.

3 Likes

This whole thread will be of interest but at least read from here down:

Don’t know if this will help your problem or not, but since i’m a model railroader i make my own building from plans I purchase as well and turn into SVG’s. One thing I noticed is that when you zoom in on the corners and look at the paths there are two nodes on each of the corners. I delete one and then straighten the line between the two ends nodes. This stopped mine from creating the pinholes at the corners.

4 Likes

Ooh; that’s an interesting idea. I’ll take a look at the plans and see if we’ve got the same thing going on with our plans.

Ya know what would be great, an Illustrator plug-in or script that turns a rectangle into 4 line segments with overlapping corners.

Hard telling because the part is so small but it does look like the burn lines seem a bit wide for such thin material. Is the focus set correctly? Lens in correct. Just eliminating possibilities in addition to others.

Hm. We’re using some chipboard behind the 1/64 ply. We should… really adjust the “material thickness” setting accordingly, shouldn’t we? SMH… :slight_smile:

2 Likes

If you have the chipboard behind the 1/64" ply, then your material thickness should be the both of their heights added together. And both should be measured with a calipers to get actual dimensions rather than nominal. With such small pieces and thin materials, differences add up quickly.

Great advice: thanks! I knew those digital calipers would come in handy some day. Now, if I can just find them again…

Have you physically watched the cut? Another thing that can contribute to the observation @marmak3261 made is multiple passes. A lot of times, duplicate lines get stacked on top of one another and the Glowforge just thinks, “woohoo, another line to cut!”

Autofocus should be (unless you’re manually changing the focus height) taking a reading and setting itself accordingly (provided the measurement beam is hitting the material, etc.). It’s hard to get a sense of just how big the burn channel looks because it’s definitely a close-up image - but they do look very dark (which could be just the picture too).

2 Likes

First thing to do is put the wood on masking material. You will still get flash back but it will be on the masking material. If you are double burning any actual burning in the joint between two materials will maintain itself easier.

The second point is only partially useful but a straight line where the laser turns on and off does not slow down and make the pin hole. So if a series of straight lines are connected and then the space between opened the laser does them all as one line without stopping turning the laser off and on instead of continuous. You will still get pinholes where it stops but even then you can often have those places beyond the work . Doing this is a great deal of work but if the result is worth the effort…

1 Like

So how do you do that? Is it by path, with some nodes connected by line segments and others not? Sounds like a lot of work for a person but maybe easy for a program…

It is a lot of work for a person, but the question is if it is worth it. I had a situation that is ideal and so was planning on making a demonstration of it. Some moves can be ganged and others not, and some that look ganged only make a mess if you go that way.

1 Like

The problem…

we need to cut out all those blue squares as there would be a pinhole at every corner…

So you select every corner,and break the line at those corners …

Now with many bits you break them all apart…


Now set the top row parts together in a group they are one multiple object
by not combining, you keep the order of the segments
and then break out all the empty spaces. you can do every other pair but not every pair.

take measurements later we see they are not perfect but a start…

By duplicating one and moving it the measured distance away and then Grouping those two you have the top and bottom as one thing…

Then more duplications doubling the number each time and array again, but to have them hit perfectly takes a bit of playing with the numbers…

Then you group those, duplicate again and change the color and rotate 90-degrees even multiple times arrayed at the distance you got but sideways…

Then you group all the verticals, and separately all the horizontals and the by deleting points delete all the excess bits.

Turn what was hidden back on al all the lines should be connected to cut sequentially the rest is up to the GFUI. And by testing the first line went perfectly and none of the rest did :frowning: so combining apparently breaks that sequence,

3 Likes

If I were trying to cut all of those squares, I would just set up rows and columns.

1 Like

If I needed to cut all of those squares, I would just create one rectangle, split it into a grid, round the corners slightly using the Xtream Path plugin that I have. Done.

Split into grid:

Round them a bit:

:slightly_smiling_face: These will not create a pinhole:

1 Like