Double sided cut

Yeah, sorry. I know it was discussed for testing by marmak3261. Honestly, I couldn’t tell you the answer. I have read every single post but that discussion made my eyes gloss over. Too many people with very specialized needs that didn’t interest me enough to burn brain cells.

An aside: marmak3261 already does way too much. The forum is relentless in it’s need for individual attention. I probably have dozens of questions left to be answered. But I’m going to let him enjoy the unit and eagerly await any new information as he discovers it.

7 Likes

The answer was something like, “No, you will not be able to enter where you want to place something numerically, but we’ll put it in the hopper.”

I am also having trouble wading through the 340 posts in that topic, so I can’t link it for you.

I think someone is supposed to report back a different way of making jigs reusable. Using the edges of the machine or adding some physical guides. I have a feeling that it’ll have to wait 'till some of the software accuracy issues are waded through.

2 Likes

As noted, the only challenge is if you’re trying to engrave or cut both sides of a circle. Engraving or cutting a circle from rectangular material is fine.

4 Likes

Presumably to do a double sided cut that meets perfectly in the middle the edges of the blank material would need to be absolutely vertical, so that when flipped the corners are exactly the same viewed from the other side. Since laser cutting leaves edges not quite square one might need to square up the blanks on a bench sander first if they have been precut.

Do the proofgrade sheets have perfect edges?

1 Like

I don’t think you need “perfect edges”. I think you just need a defining characteristic. As long as the software can positively identify a part and able to determine placement and rotation off of that part, you should be fine.

It won’t matter how straight they are, they can be irregular and curved, but if they slant vertically then any features on the other side won’t be laterally aligned with those on the first side. They will be offset, so the cut from the other side will be offset by the same amount unless Glowforge can build a 3D model of the work piece, not just a 2D silhouette.

2 Likes

Maybe I’m misunderstanding what you’re saying, but I’m thinking… Sure they will. You’re just going to slant your backside image in relation to the medium’s slant (if any). As such, everything will be aligned.

  • Tom
1 Like

We are talking about a double sided cut that meets in the middle to cut through material too thick to cut in one pass. It won’t meet if the edges slant. Yes a double sided engraving will work as you want that aligned with the edges of the surface it is on, rather than the edges underneath.

1 Like

Yeah… Not seeing a difference. The optical alignment will mean that when you flip the item over the software will know how to cut appropriately with the flipped item.

But, hey, I’m probably missing the point entirely. I’ll bow out now.

  • Tom
1 Like

It would if it could see in 3D but if it looks at the 2D outline of the top surface, and that doesn’t match the bottom surface due to slanted edges, then it will be offset relative to the cut it did from the bottom.

1 Like

I see what you are saying, so the real question is whether the overhead camera will view the pieces as a 2D or a 3D structure.

I would suspect that the answer will be 3D given all the other features presented.

I don’t think the lid camera will be able to see the bottom edge because the top is closer, so in general hides all of it.

The head camera can move, so could get a view of the edge from several angles and build up a 3D representation. It is all looking very complicated though. I hope all this was taken into consideration when they advertised it. To double side cut without an ugly seam requires alignment to a precision less than the kerf.

1 Like

What @palmercr has brought up is something I’ve been wondering/worrying about as well.

Here’s a bad drawing of the potential issue…

7 Likes

It seems that the parallelogram case is least likely, because lasers generally cut a V. More worrisome for me would be any residual errors in the optical system when you’re cutting things that are off center. Alignment will be pretty good if your errors are equal (and opposite) on both sides, but lousy otherwise. (I’m also assuming that the machine doesn’t rescale in this case, but maybe it does.)

2 Likes

Agreed. Once @marmak3261 lets the Forge succeed, let’s hope he puts it to the test!

3 Likes

Hopefully the head camera will move over the edge to get an accurate fix as long as the edges are in the cutting area. I don’t think the lid camera will be anywhere near accurate enough for double sided cuts.

2 Likes

I don’t think that would address the trapezoid case, though, unless it did a very thorough analysis of the edge conditions on each side.

1 Like

Indeed. I was replying to [quote=“paulw, post:74, topic:4075”]
More worrisome for me would be any residual errors in the optical system when you’re cutting things that are off center.
[/quote]

I.e. off centre shouldn’t matter to the head cam.

1 Like

Hold on a second. Just realized that there are two separate cases here. For double-sided engraves you’ll (possibly/probably) be dealing with a laser-cut outer edge, but for double-sided cuts you’re much more likely to have a sawn/milled/cast/stamped edge, because the stock can’t easily be cut with the laser in the first place. Which might make the parallelogram case more likely, but the amount of misalignment smaller.

1 Like

Not with our laser, but the stock might have come from somebody with a bigger laser.

1 Like