Lead-out make stock collision undetected in simulator

[skylord73]

Hi all,

I create a multiple depths pocket in my stock with Lead-In and Lead-Out options.

The Lead-In is not a problem because it's made for every depths, but when the tool arrives to the end of the toolpath, at the bottom of the pocket, for leading-out it moves throught the stock.

During simulation not collision is detected and this is a problem because I was going to break the tools...

I think the Lead-out option should be disable in multiple depths.

You can see the problem in this file http://a360.co/1SfvaMt, simulating the 2d pocket, and looking the end of the path, when it leads-out.

[HughesTooling]

I downloaded your file and there are no CAM setups or oops so can't help. Next time select export from the file menu and save as a F3D file and attach to this thread.

Thanks Mark

[skylord73]

You are right…
The cloud version misses the CAM paths.
This night I’ll check why the local version, with the paths, was not uploaded, and I’ll post the file.

The problem is not specific to my drawing, you can check with every 2d Pocket with multiple depths and Lead-out strategy

Regards

[skylord73]

My fault...

I forgot to save the last version with CAM paths.

Now should be ok.

http://a360.co/1SfvaMt

[HughesTooling]

Try unchecking Lead-Out (Exit). 

Mark

[skylord73]

Unchecking the lead-out is of corse The solution.
My point was that is a non sense to have a lead-out strategy if you enable a multiple depts, and the system should remove the dialog to avoid errors.
The second point is that the simulator has to consider a collision every interaction with the stock deeper than the max depth configured in multiple steps.

[HughesTooling]

A few thing for you to consider. First HSM was designed for Mould\Die work, it has some features added for working on a router but not a lot. For me as a mould maker the open pocket works perfectly because I'd be working with a block not in the middle of a piece of stock like you are, although the pocket's open on your part it's not really open as the part is within the stock. The 2d ops don't know there's material there unless you select the stock. If you enable stock it will machine to the edge of the stock, probably not what you want.

The best option in this situation is to use a 3d pocket as it takes the stock into consideration so it knows the pocket is not really open and produces a closed pocket type toolpath.

Mark

[Anonymous]

Unless the simulator works on the actual post-processed G-code, you cant trust it, especially for rapids. Use something like the program cutviewer to simulate the g-code once it leaves fusion 360 and before it goes to the machine. Takes a little setup but will save you from many crashes that CAM software doesn't detect.

[jeff.walters]

Even back plotting software would have a hard time showing rapids like your machine works. Simply because there are so many different machines out there and they can and do work differently.

[Anonymous]

---

@jeff.walters wrote:

Even back plotting software would have a hard time showing rapids like your machine works. Simply because there are so many different machines out there and they can and do work differently.

---

Can you elaborate? How is it that it makes a difference that the machine is a router as far as detecting rapids in a g-code simulator?

[jeff.walters]

It has nothing to do with it being a router. Even on mills different machine controls move the machine differently when rapiding. Some simply go point to point, some move at a 45° till the short leg is finished then finish the lone leg as a straight line, while others do the long move till it’s a 45° left to the end, others move X first then Y, and so on and so on. Point being its basically logistically imposable for us to know who a particular machine moves when its rapiding.

[HughesTooling]

In this case it's debatable whether it a collision anyway. The lead out is at the same feed as the cutting feed it just 3mm deep where the rest of the pocket is done in 1mm steps this screen grab shows the offending lead out and the feed rate, how is the simulator going to judge this a collision?

Mark

[Anonymous]

---

@jeff.walters wrote:

It has nothing to do with it being a router. Even on mills different machine controls move the machine differently when rapiding. Some simply go point to point, some move at a 45° till the short leg is finished then finish the lone leg as a straight line, while others do the long move till it’s a 45° left to the end, others move X first then Y, and so on and so on. Point being its basically logistically imposable for us to know who a particular machine moves when its rapiding.

---

Nobodys asking you (the CAM) to know this stuff. Thats the point of using a g-code simulator to see what actually happens. G-code simulators can be configured to take control differences into account. Cutviewer does and its probably the most basic one out there.

Of course, real CAM softwares do have NC simulators built in or as options. But we'll pretend those dont exist.

[Anonymous]

---

@HughesTooling wrote:

In this case it's debatable whether it a collision anyway. The lead out is at the same feed as the cutting feed it just 3mm deep where the rest of the pocket is done in 1mm steps this screen grab shows the offending lead out and the feed rate, how is the simulator going to judge this a collision?

Mark

---

If its at feedrate then its a gouge not a crash. So thats where the CAM softwares material removal heatmap comes into place so you can quickly see where the over cuts and undercuts are, even if they are very small.  Of course, fusion 360 doesn't have that either. But Cutviewer, a 12 year old program, does.