machine or Gcode issue

[MrYuri]

Hello,
I have an issue that I need to solve.
When I do a relief job and export a toolpath it is machined as on the video attached.
I selected raster, but it's rastered 'in parts' causing the artifacts shown on the photos at the end of the video.
https://youtu.be/O7B-LraZ9rE

I need to have the surface uniform but it seems like the tool continued raised about 0,1mm above where it supposed to.
I tried to put material blocks in different heights in case of the issue with Z axis, but it appears at the same place.
Later I noticed it's because of how the toolpath is generated.
Tool will not do a full sweep of the model from bottom to top, but goes back and forth.

Is there any way I can force aspire to make the toolpath go from one side to the other with raster without lifting the tool?

Or how can I troubleshoot whether it's the machine issue or aspire/toolpath/Mach3?

I don't think it's the machine as tool was lifted more than once as seen on the video, but artifact happened only at that place.

Thank you for advice and help

[Leo]

Yeah, I get that also.

Be interesting to see some of the answers.

I do hope some Vectric Staff member sees this thread, and can make a suggestion.

My machine is VERY solid, so I doubt it is a machine, or tooling issue.

I get that even on HDU - which is stable.

I just stay away from that strategy as much as possible.

I KNOW - this is not the answer you want, but it's not just you.

[dwilli9013]

Hmm sounds like a pattern as I also will get the same thing happening. I actually had it happen in a larger panel of Corian. Almost ruined the whole job.

[mtylerfl]

Try making a boundary vector around the carve that is “offset way larger” than normal. Then select that vector, recalc your finish Toolpath (assuming no Roughing since you’re using foam material). Run a preview and see if that “fixes” the issue. If not, make the boundary offset even larger and try again.

[Leo]

I have seen it on a variety of materials and on 2 different, no make that 3 different machines.

One common thread is, mach3

Happens on raster, not on offset.

I think the OP is using wood.

I have 3 jobs on deck right now, so I cannot experiment for a while, but I will test what you suggest.

Perhaps it is something to do with mach3?

[mtylerfl]

Not exclusive to Mach 3. I’ve experienced it myself on ShopBot (carve an area, then jump to another area to partially carve, then return to finish the previous area, etc.)

However, I have not had any adverse visible issue when carved like that - no “depth mismatch” in the final result thus far for me.

I’ve seen others post about this in the past. Been awhile since I’ve seen it brought up. I THINK Fixitmike was the first to suggest the “jumbo” boundary vector to overcome the issue.

[ChrisInEstes]

Usually, when I post a response, it's info that's already known... So why stop now? Haha... What I post below is also posted on one of the Facebook CNC pages.

Yep... Mine (Aspire) does that too... I never thought much about it beyond figuring the lines were caused by backlash in my machine. The way the particular model you reference looks like it's done, it would have to pick up and move because some areas are in "shadows", or the tool would have to backtrack over previously machined areas if it didn't pick up off the material. Many years ago, when I used another CAD-CAM package with a non-Mach 3 MC3V controller, there was an option to keep the tool at depth, and it would backtrack over previously machine areas to get to the shadowed areas instead of picking up to move. That helped, but didn't make up for movement in the material.

Chris

[Leo]

I don't really care if the spindle picks up to go to another area than come back and resume.

The real issue is mismatch of the surfaces. That can really be an issue.

I guess it not just me and the OP, and I guess it's not just a Mach3 issue.

I am going to send a email to support to read this thread.

[kyeakel]

Did you check the Gcode to see if there are variations in the Z height?

Kipp

[ger21]

It's either lost position or mechanical. I'd lean towards mechanical, like a loose set screw on a coupler.

[Rcnewcomb]

but didn't make up for movement in the material

I believe this is the key issue. You can inspect the code to see that the Z heights are correct. You can reduce this by forcing the toolpath not to jump around, but that does mean a longer cutting time.

[MrYuri]

You can reduce this by forcing the toolpath not to jump around, but that does mean a longer cutting time.

how you can force it?

I found solution only by extending vector around model that is not weird too much and set is as boundary.

but this sometimes significantly increases cutting time.

[Rcnewcomb]

Yes. That is what I meant

[Leo]

I am 100% certain that on the three different machines I tried this on that the problem was not mechanical.

Checking "Z" heights in the G-Code on a 3D model carving is beyond my skill level. I have only been cnc programming for 30+ years. Granted 99.5% mechanical, not surfacing.

If the answer is to create a larger boundary - then, that is a Vectric issue and NOT a mechanical issue.

[MrYuri]

Agree with Leo on this.

Solution should not be to create bigger boundary (that is called a workaround).
It might significantly increase the cutting time.

I believe there should be option "don't lift the tool" or something like that to remedy the issue.

User than will have the option to wait for longer toolpath calculation, sure, or risk such artifacts on the product.