Full retract in 3D contour tool path - didn't retract...

[walrys]

Hello,

I was working on a part yesterday and towards the end of the cycle (when the tool was at its lowest position in Z) I heard it load up and break.  (Bought new that very morning too...  😥)  and yes - I should've been paying closer attention to the software simulation but somehow managed to miss it.

I've since looked at the tool path and can rid myself of the issue by reducing the maximum stay-down to 0mm (which also means I get a less efficient tool-path in other areas).  Anyway, surely the tool shouldn't be staying down where there is stock present (the arrowed channels) - it even says so in the helper window "Note: Regardless of the stay-down distance, the tool retracts between passes if the rapid motion violates the workpiece."

Is this a bug related to how the tool moves at its lowest position in Z, or by 'workpiece' does it mean just the body/model that's being machined and that I should be quite prepared for the tool to want to go through solid stock?!

You'll also see on the 'nibbles.jpg' that the tool (again at maximum Z depth) is retracting, repositioning and then plunging through stock to get to the start of its lead-in.  This wasn't an issue as I had cleared a 4mm channel around the model to give the taper mill some clearance and I appreciate that I can reduce my lead-in distance to avoid this).

Have I overlooked something / am I thinking about the issue incorrectly - regarding the nibbles I realise that there are things I could be doing to minimise stock collision - like machining in clearance etc.  it's that ploughing through the stock that bothers me.

Many thanks for any advice.

Wal.

[evanp4509U4JZ]

Fusion changed the menus in the tool path settings. I would check your heights. They muddied it up with "highest of", "lowest of" allowing multiple things to accidently be selected leaving your offsets possibly from an unintended selection.

Chuckled at the Pink Floyd ref

[matthewrjacobs]

Looks to me like a model awareness problem.  Is "Include Setup Model" Checked? or possibly the body you chose is different than what is in the setup? 

[programming2C78B]

the height tab update didn't modify any existing programs. 

It has to do with how you selected your Stock in the contour toolpath. 

highly suggest using a 3D Adaptive before running this path. 

"Is this a bug related to how the tool moves at its lowest position in Z, or by 'workpiece' does it mean just the body/model that's being machined and that I should be quite prepared for the tool to want to go through solid stock?!"
The workpiece is your model, not the stock. You should have retract heights above your stock/fixtures. Stay down distance refers to transition moves between two cutting passes. 

[seth.madore]

Would you be able to share your Fusion file here?
File > Export > Save to local folder, return to thread and attach the .f3d/.f3z file in your reply.

[walrys]

Thanks all for getting back to me.  I think I've got to the bottom of this - and it's weird.

Whilst preparing the project for upload (I removed a large part of the model as I didn't want it public facing) I thought I'd tidy up the details I'd entered for the custom ball-nose taper tool I'm using.  My set-up is very 'touch-off on ciggy paper' manual and I tend to just input the strictly necessary details without worrying too much about the less relevant fields.

Look at pic_01 - that's how it was and the resulting tool-path shows the tool ploughing through stock.

In pic_02 you can see that I've altered the shoulder length so that the tool appears more like how it looks in real life and this time the tool-path has been generated without any collision taking place.

Weird huh?  Who'd have thought?

I'm also attaching the trimmed down project file so you can have a look at the issue first hand.

[programming2C78B]

What are the odds your stock is between 21 and 29mm thick at the retracting areas?

[walrys]

Stock is 5mm thick.  What were you thinking?

Do you not find it in the least bit strange that the tool-path would be generated either 'properly' (avoids ploughing through stock) or 'improperly' (traverses through an area at maximum z depth where the software knows there is stock) depending on how the (rather arbitrary) shoulder length is defined?

Even if we were to ignore the 'proper/improper' arguments (you might choose to define those terms differently to me!) in this particular case why would the tool-path be any different because of the shoulder length of the tool?  Of course, I understand that the shape of the tool influences how it interacts with the stock/workpiece, it's just that in this case it makes very little sense that one would be any different from the other.  I'd stop short of calling it a 'bug' - but it's certainly anomalous...

Wal.

[programming2C78B · Accepted solution]

"What were you thinking?"
That maybe in one case, your stock is between those two shoulder lengths at it would allow it to be a cutting move - after waking up I realize you aren't talking about the flute length changing.

I am able to replicate your issue when using a 30mm shoulder

1) no idea why the paths should change if the flute is the same 
2)This is something you shouldve noticed while simulating before running the machine
3) reducing your stay down distance removes that move (since it was a transitionary one) and replaces it with leads and a retract.

whats odd to me: despite the same settings, the two tools treat this corner differently

and fusion does think they are different, despite the numebrs looking identical to me.

 


Changing your surface tolerance will make it generate properly, even with the 50mm stay down distance and 30mm shoulder. (hi seth)

(it even alarms out where it rapids into the stock - a non issue if you have an adaptive before)

[walrys]

OK, so a number of ways to navigate around this being an issue.  Bumping up the surface tolerance won't make much of a difference for the machines I run, it might for some though.  For now I guess it's a case of paying more attention to the simulation (something I acknowledged in my OP) - but in my defence, I still reckon that F360 behaved in a very unexpected manner by riding that tool-path so rudely through the stock.

Ah well, ya live and learn (albeit remaining none the wiser..!)

Cheers.

Wal.

[programming2C78B · Accepted solution]

"Bumping up the surface tolerance won't make much of a difference for the machines I run"

This has ""nothing"" to do with your machined part, but moreso how Fusion itself is computing your toolpath. As I've shown, changing it is enough to actually fix your problem transition move. 

[walrys]

Many thanks for your time and help.

Wal.

[programming2C78B]

Please click "Accept Solution" if what I wrote solved your issue!

[walrys]

Solution accepted..!  *With the caveat that it's somewhat arbitrary - much like altering the shoulder length of the tool.  Still no real explanation as to why the problem is occurring in the first place (which in the interests of tool life it really ought not to be..!)

Don't take this as me being a miserable git - I totally appreciate your time and help, I just don't feel any the wiser as to why the problem occurs - but hey, double checking the preview and tweaking the surface triangulation tolerance if I see any potential ploughing is the way to go for now.  👍

Wal.