Unusual problem with gouge in verify, but how?

[lowcountrycamo]

Tool path is new swarf, only one pass.  Collision checking on.  You can see in the picture  the gouge in the wall corner.  When stepping through verify I can see the gouge happening but the tool is where is should be, in other words, the tool is cutting beyond its boundary.   I am certain the gouge is happening in this path.   In back plot the tool is not gouging.  Have you ever seen this?  I could post the file on Monday when I get back to work.  I took these screen shots yesterday for this post.  No collision warnings are called. 

Thanks,

Steve Austin

[lowcountrycamo]

I just noticed the way the tool axis vectors are leaning forward.  This may not be the problem but how to fix this?

[lowcountrycamo]

I will set up a test file to show if I can.

 

[motor-vater]

I see this alot in verify, not so much at the machine. I have learned to just hope/assume its the verification itself. So I am watching hoping the greater minds weigh in on this

[Colin Gilchrist]

You fix it by building 'good' chains of geometry, and then using "sync" mode during chaining.

Syncing a chain, tells the software to treat each 'section' as being directly linked to the section below.

I try to make sure there are an equal number of chain elements, top and bottom. It just makes life easier. But there are something like 5 or 6 different chaining Synchronization methods. Each has pluses and minuses, based on what you have for existing geometry. 

[huskermcdoogle]

Syncing is a topic I would like to learn a lot more about.  I have had to figure it out a few times, and have had mixed results.  There probably isn't a magic explanation for it, but I am sure something would help lots of us.

[Leon82]

I usually use sync by branch or entity I think. Also what Colin said about number of entity in the chain

[Colin Gilchrist]

Without "Syncing" the chains, Mastercam will "step" along each chain, starting at the "start" of each chain. Each of these "step" distances is based on your cut tolerance. Mastercam will step "equally" along each chain. When the "chain length" is different between the upper and lower chains, you will see the tool begin to lead or lag (tilt forwards or backwards, relative to the "cut direction"). As the tool starts to lead or lag on the path, the algorithm is trying to "adjust" the tool vector, to keep the bottom of the tool tangent to your drive geometry. What you will begin to see is the tool "wiggle around", as it is trying to keep the outside flank of the tool tangent to your walls, while also keeping the tip of the tool tangent to your floor geometry. The situation degrades to the point where gouges occur.

The "Sync" modes tell the software to "break up the path into discrete sections", and then "Fan" the tool (adjust the path), so that the tool starts and stops at the "branches" evenly, between each section.

Instead of the path being "from start to end" of the entire chain, the path only considers the "start and end" positions of each section, and then smoothly interpolates the tool axis, maintain that flank tangency.

I will always attempt to "Sync" my chains for Swarf tool paths. Not so much for Curve 5 Ax Paths.

When in doubt, I use the "Point" method, and I create Points on the Upper and Lower Chains, where I want the tool to "line up" from the upper to the lower chain.

--------------------

Another way to think of it, is this is "Tool Axis Control" for Swarf. Since the Swarf path is "automatic" in how it creates the Tool Vectors, you can get weird stuff like the tool leading/lagging on the chains. The "Sync" methods tell the Swart path where to align the tool axis vectors, between each segment, and this has the result of smoothing out the transitions between sections, as well as "matched motion" across a segment.

 

[Leon82]

I have also corrected some funky tool motion by breaking an entity on the top and bottom of a set of chains and adding a vertical sync line.

Also on a radial pocket I was able to get my tool to gradually enter against a wall by breaking the one large ark into may with sync lines leaning more and more towards the end tool position. This left less of a "snap" as the tool positioned against the vertical wall paralell to y axis.

 

 

[lowcountrycamo]

I have used sync in old swarf as you select both chains out of one dialogue box. Is it possible to sync in new swarf where we select chains separately?

[crazy^millman]

4 hours ago, lowcountrycamo said:

I have used sync in old swarf as you select both chains out of one dialogue box. Is it possible to sync in new swarf where we select chains separately?

The old school swarf is a the only one that allows syncing. The new one does not allow this. It assumes and does what it thinks is best.

[Leon82]

Under tool axis control there is a box to limit rotation axis movement.

I was playing with an oring groove and checking this box got rid of weird motion along what should have been a straight cut along y axis . Before checking it was moving a axis in a weird dip in the middle of the cut.

 

[jlw™]

I prefer the old but the new has easier to use lead in/out.  Old swarf definitely give cleaner motion IMO.