Solved: Polygonal profile instead of round

DAVID.LEWIS149NS4G · ‎01-21-2019

I thought I'd try HSMWorks within Solidworks for 3D toolpaths, since I already have and use Fusion, and I don't think I want to export solids to F360 anymore if I don't have to. I'm machining molds in 4140 alloy steel.

The problem is I am getting facets in a surface that should be a smooth arc. The mold should make a round tube, but instead it will make a tube with a profile of a polygon with a whole bunch of sides. I have my doubts that they are toolpath artifacts though - Solidworks displays similar anomalies in the graphics view, and I am starting to wonder if HSM is just following along for the ride. The part comes out of the mill looking exactly like the graphic below, and the bottom is especially apparent - IT'S FLAT! No bueno.

The part isn't built anything like this though - it is totally round. I am beginning to think the problem lies in Solidworks building the model incorrectly. How would this happen though?

DAVID.LEWIS149NS4G · ‎01-21-2019

Not sure why I can't include an image in a reply, so attached is another view:

Marco.Takx · ‎01-23-2019

Hi @DAVID.LEWIS149NS4G,

From only a screenshot it’s hard to say.

When it is possible please upload your model so one of us can take a look at it.

What kind of tolerance/Smooting settings are you using.

These settings can create flat spots by the STL this is created in that operation in the background.

If my post answers your question Please use .Accept as Solution & Kudos This helps everyone find answers more quickly!

Met vriendelijke groet | Kind regards | Mit freundlichem Gruß

Marco Takx
CAM Programmer & CAM Consultant

DAVID.LEWIS149NS4G · ‎08-17-2019

I know this is old, but I'm chiming in just so I can set the record straight.

This is 3D adaptive clearing at a very high resolution. It produces a many-sided polygon shape, but since this is a roughing strategy it doesn't really matter. When set to fine toothpaths it will never make a round shape.

Nobody at Autodesk even knew this.

johnswetz1982 · ‎09-01-2019

What do you mean by "when set to fine toolpaths"?

lenny_1962 · ‎09-02-2019

We are still waiting for you to up load the SolidWorks part.

I design and machine in SW and HSMWorks every day and never have what your are talking about.

DAVID.LEWIS149NS4G · ‎09-02-2019

@lenny_1962 wrote:
We are still waiting for you to up load the SolidWorks part.
I design and machine in SW and HSMWorks every day and never have what your are talking about.

Who is this "we" business? I figured out what the problem was some time ago. Maybe you didn't read the whole thread?

3D Adaptive delivers polygonal toolpaths that do not resolve away as the stepover is made finer. For this reason, it is unreliable as a finishing toolpath - and I learned the hard way. I'll say it again: even Autodesk was stumped.

The way Solidworks resolves onscreen was a red herring.

Not sure why the post from @Marco.Takx was made the solution, but I'm removing it in favor of this post.

johnswetz1982 · ‎09-02-2019

I doubt Autodesk was stumped about this because it is a roughing toolpath. It is not meant to produce a finished surface. It would need to be followed up with a finishing strategy.

lenny_1962 · ‎09-03-2019

@DAVID.LEWIS149NS4G wrote:
@lenny_1962 wrote:
We are still waiting for you to up load the SolidWorks part.
I design and machine in SW and HSMWorks every day and never have what your are talking about.

Who is this "we" business? I figured out what the problem was some time ago. Maybe you didn't read the whole thread?

3D Adaptive delivers polygonal toolpaths that do not resolve away as the stepover is made finer. For this reason, it is unreliable as a finishing toolpath - and I learned the hard way. I'll say it again: even Autodesk was stumped.

The way Solidworks resolves onscreen was a red herring.

Not sure why the post from @Marco.Takx was made the solution, but I'm removing it in favor of this post.

The "WE" was myself and Marco.

Asking to see the file not only helps you but also others to see what your settings are that cause said outcome.

You stated you thought HSMWorks follows what SW showed on the screen and I haven't had that happen, just curios to see.

I cut an entire 4th axis part and finished it with a 3D Adaptive and had no faceting what so ever on the screw, not a straight wall on it.

DAVID.LEWIS149NS4G · ‎09-03-2019

Maybe this helps explain it a little better.

3D Adaptive

lenny_1962 · ‎09-03-2019

That tells us nothing.

What are your settings?

Take a snapshot of them and show us.

DAVID.LEWIS149NS4G · ‎09-03-2019

@lenny_1962 wrote:
That tells us nothing.
What are your settings?
Take a snapshot of them and show us.

I deleted the toolpaths a long time ago - since they didn't work - and I learned my lesson not to use 3D Adaptive as a finishing toolpath. I have moved on.

Clearly this phenomenon is new to you.

It shouldn't be hard to recreate - just resolve the stepover to something absurd, and use a ball end mill. The photo I posted is external, but the results are identical on internal cavities.

johnswetz1982 · ‎09-03-2019

@lenny_1962 This is nothing more than mis-use of an operation. He was using Adaptive as a finishing strategy even though it is not meant for that. He was expecting to only have to run Adaptive and not follow up with a finish op such as parallel, etc. What I dont get is the "even Autodesk was stumped" comment because they know Adaptive is only a roughing strategy.

DAVID.LEWIS149NS4G · ‎09-03-2019

@johnswetz1982 wrote:
@lenny_1962 This is nothing more than mis-use of an operation. He was using Adaptive as a finishing strategy even though it is not meant for that. He was expecting to only have to run Adaptive and not follow up with a finish op such as parallel, etc. What I dont get is the "even Autodesk was stumped" comment because they know Adaptive is only a roughing strategy.

All things I have previously admitted to.

I can't help your not understanding why Autodesk was unaware of this phenomenon, except to say that they weren't, and neither was @lenny_1962. @Marco.Takx had a hypothesis, however by the time this conversation went anywhere in this discussion (not the first I've had on this issue) I had moved on.

If you're curious about this phenomenon, I suggest you do what I suggested above and see if the same thing happens as in the photo I posted.

lenny_1962 · ‎09-03-2019

@DAVID.LEWIS149NS4G

here is an airfoil i just did with a 3D Adaptive, see pics, can be done without facets.

this has no stock to leave.

@johnswetz1982

Yes I agree not what Adaptive is for, but I do small intricate parts that I leave .008 to .005 stock to clean up so I have to make 3D Adaptive follow without facets.

DAVID.LEWIS149NS4G · ‎09-03-2019

BTW @lenny_1962 @johnswetz1982 @Marco.Takx this is the original part in question. It's still in a semi-finish state in this photo, and there is enough stock left over for a smooth finishing pass. I asked originally because I was concerned that if I used 3D Adaptive again it would be at depth like this, which is no bueno.

johnswetz1982 · ‎09-03-2019

I have made very similar molds. But knowing that Adaptive is a roughing operation, I left .010 stock to leave then Finished it with a Parallel operation. Run a Parallel with small step over and that will be a smooth part. You can even leave a courser finish in the adaptive just so long as its not enough to overload the finish tool.

lenny_1962 · ‎09-03-2019

@johnswetz1982 wrote:
I have made very similar molds. But knowing that Adaptive is a roughing operation, I left .010 stock to leave then Finished it with a Parallel operation. Run a Parallel with small step over and that will be a smooth part. You can even leave a courser finish in the adaptive just so long as its not enough to overload the finish tool.

If you go to Beta Mode there is a Blend toolpath that is way better than Parallel.

Parallel isn't good on near vertical walls.