Only one stepdown generated for "in control" toolpath

[ltomuta]

In this project the last operation is supposed to enlarge the existing hole from 3.5 to 4mm. If the compensation type is "in computer", all looks good (a). However, if compensation type is "in control" only the first stepdown is generated (b).

a)b)

I don't understand why. I know that sometimes in control toolpaths can't be generated for a given hole/tool diameter ratio. But why would it now compute for one stepdown and not for the rest?

This also brings a related question: a warning is generated saying that "a contour was not machined because the given lead parameters would cause a collision". This is not very useful because:
a) it does now tell where the affected area is, nor what may have caused it

b) if there is a collision, you can show it to me in simulation

c) I've seen this warning for toolpaths which, as far as I can tell, are generated as expected, so I tend to ignore the message anyhow as random noise.

[DarthBane55]

Very strange for the Z not generating issue... I don't have explanation for it, someone else will hopefully. 

I can clarify your 3 points at the bottom though:

A-It doesn't need to be more clear I think, the toolpath that does not generated is the one with the issue.  Look at your selection, look at the result, it will be missing at least 1 toolpath.  It does say what the cause is, it is actually pretty much all it says...  The lead parameters would cause a collision.  Usually, the lead-in/out is too big to fit in a pocket, and would make the tool collide with material.  In these cases, it does not generate the path.

B-It cannot show you in simulation, because with this alert, the path causing the issue is (usually) not generated.

C-Sometimes it seems like it generates as expected, but the lead parameter might have been skipped.  Look carefully at the result, sometimes there is no lead-in/out on the path that generates this alarm.  I would not ignore it, specially if you are doing a finish pass (could create a line on the wall or something like that).  Adjust your lead parameters instead.

[seth.madore]

Remove the Vertical Lead-in Radius. You're getting too close to the wall and it detects a gouge. Either that, or put your Sweep Angle to 45 degrees. Shown with no vertical lead radius:

[ltomuta]

The problem is that what I see (in both toolpath preview and simulation) is
a toolpath that does not match my parameters. Even if I make small changes
and I gradually improve it the right direction, the improvement will not be
reflected by the toolpath and I will have no idea that I made it better or
worse. Unless I fix the problem right on, all my changes are dropped just
like the original input.

That is why I would rather prefer the toolpath to generate "as designed,
with problems" (with collisions signaled clearly in simulation) and the
"improve, with warnings" to be an option _after_ I've seen my path.

[ltomuta]

Reducing the lead-in distance worked, thanks. But how does that value's impact change with Z?

Changing the angle did not help and, as mentioned above, there was no indication of any progress/improvement with the change. Did the problem improve with 45 or was made worse? Is it worth trying a 20 degrees angle or a 120 one?

[ltomuta]

Another operation, another error. No lead-in/out related root cause hint.

Does this look like an incomplete toolpath to you?

Which "contour" of the two selected edges would you say was not machined?

The only real hint that something is wrong was on machine when the feature ended up with about 1mm smaller than expected. So here I am at 21PM debugging a thing that's blocking the machining of the piece. Not fun.

Yes, I could have inspected the warning. But virtually any multi-step contour/pocket operation will have ignorable warnings so inspecting every new one is largely a waste of time. Plus there's no useful info in it and I cannot see the collision either.