Announcements
Attention for Customers without Multi-Factor Authentication or Single Sign-On - OTP Verification rolls out April 2025. Read all about it here.

Rapid collision with Stock

Anonymous

Rapid collision with Stock

Anonymous
Not applicable

I'd like to preface this post by stating that I've scoured the forums on this and can't find an answer that addresses my issue. I'd also like to say that before anyone asks, I cannot share the file that is giving me trouble. I'm trying to reproduce the issue in a test model to be able to share it, but I figured I'd see if anyone here knew right away that I was doing something dumb. I have taken some screenshots to try to help explain the problem. This is going to be a long post, so I'll get right to it.

 

I am running a 3D Adaptive Clearing toolpath which calculates in at around two hours machining time (down from over four and a half hours on a different CAM Package. Gotta love that Adaptive Clearing!). This is my first (of two) operation(s) in this file. Upon generation of the toolpath, fusion also generates somewhere between 160 and 900 collisions with the stock, despite the fact that my tool holder never comes in contact with the part and my maximum step down is set to 15mm (0.591") from a tool flute length of 19.05mm (0.750").

 

Many of these collisions were due to an apparent bug in Fusion's tool [library and/or tool simulation model]. When setting up my tool (a 1/4" dia., 3 flute carbide end mill with 3/4" flute length, no shoulder, 1/4" dia shank and 4" OAL with 2.15" body length), I set the shoulder to 1", because it didn't have one and 1" was longer than the flute length. I figured that because the diameters were the same, Fusion's kernel would ignore the shoulder completely. I was wrong, and when running the toolpath for the first time, fusion was stepping down into the second stepdown pass and trying to cut beneath material it hadn't cleared the first pass. Shouldn't have been possible to generate that and call it a valid toolpath, but it tried. Thankfully it was just around the machining boundary and the cut was in foam, so when the tool shank hit the material, it burnished/melted/ripped it off instead of breaking my tool and destroying my machine.

 

Checking the forums for that error, I found a post where someone stated that if you have no shoulder on your tool, then shoulder length should be set to the body length (for me, 2.15"). I did this, and inexplicably [to me], fusion immediately generates a toolpath minus over 500 "tool shaft collides with stock" errors. So despite the flute and shank diameters being equal, Fusion is somehow unable to take the tool shank into account if the shoulder isn't fully defined. Even if you don't have a shoulder on your tool...

 

Anyhow, on to the next issue. I still have between 30 and 500 collisions in my toolpath, depending on how I generate it. All of them now appear to be "rapid collision with stock" errors. As you can see in the pictures below, Fusion is generating rapid plunge moves. It's a bright yellow line indicating a rapid move straight down into the material, ignoring my plunge feedrate (600mm/min vs. a rapid feedrate of 6000mm/min). I haven't checked them all to see if they're all plunges, but I'll assume most of them are, because when I increased my stay down length and priority, I'm generating hundreds fewer collisions with my toolpath.

 

I could understand if there were a previous operation here which should have removed the material, but there's not. I'm machining from modeled stock the exact same size as what's on the bed of the machine. The only thing I can think of is increasing my vertical lead in as another post suggested, but one, that's a workaround, not a solution, and two, I'd presumably have to increase my vertical lead in to greater than my stock height of 2". Besides being a ridiculous idea, it would likely increase my machining time by several hours at the least.

 

Anyway, I know I haven't given you all a lot to go on, hopefully someone can immediately point at something I've said and tell me, "you're doing it wrong, dummy!" I will try to reproduce this in a file that I can share and post it here. Thanks in advance to a great community here!

0 Likes
Reply
2,464 Views
17 Replies
Replies (17)

LibertyMachine
Mentor
Mentor

Hmm.. The images aren't a lot to go by.

 

Totally understand not being able to share to the forums.

 

Would you entertain inviting someone to the project? I'm willing to sign an NDA and you can boot me from it as soon as the problem is resolved.


Seth Madore
Owner, Liberty Machine, Inc.
Good. Fast. Cheap. Pick two.
0 Likes

Anonymous
Not applicable

My current NDA with the customer doesn't give me that option, unfortunately. I will reach out to the customer and see if we can go that route. In the meantime, I'm trying to reproduce it in that other file. My largest question here is why doesn't Fusion see that plunge into the material as a feed move? Why a rapid? If I change my ramp option to "helical", it still generates a rapid into the material, the very first cut.

0 Likes

LibertyMachine
Mentor
Mentor

Can you open up the toolpath in question and take a screenshot of the Heights page?

Can you also take a screenshot of your Setup definition, where your WCS zeros are coming from and what size stock?


Seth Madore
Owner, Liberty Machine, Inc.
Good. Fast. Cheap. Pick two.
1 Like

Anonymous
Not applicable

I was able to reproduce the problem in a test file. Same CAM settings IIRC, stock height is the same, and the contours of the model represent some of the challenges I'm facing in this model, including undercuts which will need to be machined from the other side. I haven't added any of the work holding bits to this model since they are irrelevant for these purposes and they don't seem to be causing me any grief in the actual model.

 

The stock and WCS setups are pretty much identical except in length and width of the stock. Also you'll note that I'm not including the setup model in the Adaptive Clearing, but picking the model again just to be sure it has the correct geometry. Hopefully this gives you more to go on.

0 Likes

Anonymous
Not applicable

Also it is worth mentioning on the heights tab, the heights are so close to stock surface because with 2.15" of tool coming out of the collet, my X-Carve machine barely clears the top of the 1.915" stock (less than 12mm to homing switch).

0 Likes

LibertyMachine
Mentor
Mentor

Well, I got good news and bad news.

 

Good news: It's not you, and it' not something simple you are missing.

 

Bad news: I don't get frustrated with CAM problems. This file changed that. I've not much clue as to why it's going what it is. The best I can reason out is that there are spots were it simply won't perform a lead-in/out. I've tried multiple combinations of settings and no luck. I can only try to send it up the chain and see if I can get some feedback from Autodesk

 

@jeff.pek @cj.abraham @angelo.juras  if one of you have a moment (haha) could you take a look at this file? I thought it was being caused by the "Both Ways" and then perhaps it was the stepover, or the tolerance (which is yuuge) or the lead-in/out being too large, but nothing I've changed seems to have an entire effect on the rapid collisions. I've reduced them somewhat, but they still persist.


Seth Madore
Owner, Liberty Machine, Inc.
Good. Fast. Cheap. Pick two.
2 Likes

Anonymous
Not applicable

LOL. The tolerance being so large is because of trying to replicate the original settings. The original model is three feet wide, and close tolerance is not needed for this particular model. The stepover is large because I'm cutting in foam at 27,000 RPM. My three flute tool doesn't even notice that load.

 

Also, FWIW, the "both ways" machining is new (as of the 17th?) and was not present the one time I cut the toolpath. It did cut another 30 minutes off of the machine time though. Nice!

 

I'm stumped on this, and so is my coworker, who's been using Fusion a lot longer than I have. I appreciate you looking at it, and hopefully something shakes out at Autodesk. To be honest, I ran this the first time because the material isn't going to care that this is a rapid. I can still make the part. What bit me last time was that tool shoulder bug. Maybe I'll post more on that when I have time. My problem now is, I know this rapid plunge is wrong and I can't rest until it's fixed.

0 Likes

angelo.juras
Alumni
Alumni

@LibertyMachine@Anonymous - I will take a look at it and get back to you.

1 Like

angelo.juras
Alumni
Alumni

@LibertyMachine and @Anonymous, I took a quick look and I see what you are experiencing. I need to dig more into this one.

0 Likes

Anonymous
Not applicable

Again, glad to see it's not just me. In the original file, there are exponentially more of these. I believe it to be a combination of using the silhouette, which forces the plunges, and not wanting to use any ramping methods. The plunges should rapid to stock top plus lead-in height, switch to lead-in feed rate (preferably a separate ramp/plunge lead-in feed rate) lead in, then switch to plunge feed rate until plunge is complete.

 

Possibly there is a confusion in Fusion's kernel on what to do when the horizontal move at the bottom of the plunge doesn't have room for a lead-in? I notice that all of those rapids appear to be ignoring the plunge completely and simply skipping to the lead-in for the first 'pocket' move.

 

Just my two cents.

0 Likes

HughesTooling
Consultant
Consultant

You are getting the collisions for 2 reasons. One you have not enabled stock so the op doesn't know it needs to pocket rather than clear leaving a boss.

 

The second reason has come up before but was not taken seriously in the replies and is caused by smoothing. It looks like all the leads are added before smoothing is applied so after smoothing the endpoints shift a bit and what where vertical leads end up angled. See attached file, I've enabled stock and disabled smoothing and there are no collisions, in the second op I enabled smoothing gererates collisions!

 

Mark

Mark Hughes
Owner, Hughes Tooling
Did you find this post helpful? Feel free to Like this post.
Did your question get successfully answered? Then click on the ACCEPT SOLUTION button.

EESignature


2 Likes

Anonymous
Not applicable

Thanks for the assistance, @HughesTooling.

 

A couple questions for you. When you enabled 'stock contours', you didn't make any contour selection in either operation, so according to the tool tip, the stock defined in the setup is used, as shown by the yellow contour following the setup stock. My question is, what's the point? Shouldn't the setup stock be the "remaining stock" for the first operation regardless of settings? The tool tip also says that this is to define the area of the stock surface that needs to be faced. Perhaps the 'stock contours' option needs a better tool tip, because I don't see how it would apply here. I'm not performing any facing operations. Does Fusion consider a 3D Adaptive clearing toolpath a facing operation? Also, 'stock contours' are enabled in both operations, yet there are still collisions in the first. How many collisions were eliminated by enabling 'stock contours'?

 

Addressing the smoothing option, I do see that it helps, but this seems more like a workaround than proper behavior. If the lead moves are generated before smoothing takes place and result in collisions which ruin stock and possibly break tooling, isn't this unwanted behavior a bug in Fusion which needs to be fixed? I'd really like to be able to use the smoothing option, because smoothing shaves approximately 20 minutes off of my original machine time all by itself.

0 Likes

HughesTooling
Consultant
Consultant

When enabling stock contours, leaving it blank just automatically selects the stock from the setup, you have the option to select a sketch contour for more flexibility it's up to you to decide what's useful. Instead you could enable rest machining and leave it as from setup and get the same results.

 

Using Rest gives you the option to machine from previous ops but you can also select a body in the setup to give you more options like machining from a casting\forging or pre machined stock.

 

And yes i think the collisions created from using smoothing are a bug. I use much tighter tolerances so any deviation from vertical can usually be ignored as they are less than 0.02mm but for your application it is a problem.

 

Mark

Mark Hughes
Owner, Hughes Tooling
Did you find this post helpful? Feel free to Like this post.
Did your question get successfully answered? Then click on the ACCEPT SOLUTION button.

EESignature


0 Likes

cj.abraham
Alumni
Alumni

Enabling "Rest Machining: From Setup Stock" in the test file eliminated all collisions in the test file for me. The tool is plunging because the ramp mode is set to plunge. Setting the ramp mode to helix will always force a helix, and will never plunge (it will instead skip the area entirely).Capture.PNG

 Also, there is something to be said about having a tolerance that is nearly 20% of the tool diameter and bigger than the stock to leave value. That is definitely going to cause problems with gouging and causing undesired collisions. I realize you're working with a large file, but I still recommend using the default tolerance settings.

0 Likes

HughesTooling
Consultant
Consultant

@Anonymous Looking through your original file again and I noticed you have model enabled and selected in your op, you don't need model enabled in this case as the model is selected in the setup. Generally you only use model in an op to add patches to cover features to improve machining\finish. For example if you have engraving you can add a patch so you have a continuous surface while finishing, other ops without the patch selected will ignore it.

Generally leave model unchecked.

Capture2.PNG

 

@cj.abraham did you remove an attached file? I tried your suggestions in the original file and after waiting a few seconds the simulation still shows collisions. The file I think I downloaded from your post showed a load of collisions because the flute length on the cutter was too short.

Here's what I see with the original file with rest enabled and helical ramps, file's attached. Just reducing the smoothing tolerance to 0.1mm will fix those collisions.

Capture2.PNG

Mark

Mark Hughes
Owner, Hughes Tooling
Did you find this post helpful? Feel free to Like this post.
Did your question get successfully answered? Then click on the ACCEPT SOLUTION button.

EESignature


1 Like

Anonymous
Not applicable
Any ideas from autodesk? I've got a lot of good workarounds, but no real fix.
0 Likes

Anonymous
Not applicable

It seems that there is no good answer to this, but I appreciate the help and workarounds I received here. I will leave this thread open hoping for a better answer from Autodesk, unless a moderator sees the need to close it.

0 Likes