3D adaptive clearing makes a lot of retract moves

[Muzzerboy]

I'm trying to create a simple 3D adaptive clearing operation to rough out the 2 cylindrical bores here. The path starts out OK, spiralling down, then clearing out the material in 2 passes. Then it starts endlessly jerking in and out of the holes. I've tried changing all the parameters relating to smoothing, tolerance, tool containment, staydown etc (and many more besides) but I can't get these endless retract move (yellow) to go away. I started with pretty much standard settings. For the geometry I used "machining boundary" and selected the 2 circular outlines at the top of the holes and chose "tool centre on boundary". For my heights, I have the bottom set 1mm below stock bottom and top is 1mm above model top.

Can anyone suggest what is going wrong here? This is one of those conFusion problems that comes up with the CAM settings. I expected a simple(?) 3D clearing op be fairly straightforward but I'm up against a brick wall here.

Thanks!

[Muzzerboy]

Even with a very simple model (round bar with cylindrical centre hole), I get the same issue - once the hole intersects with a cavity. 

I changed to 2D adaptive and the problem went away. But it's not at all clear what the problem is. Or that there won't be situations where I would need a solution. I'd love to know what the recommended solution is - use 2D adaptive?

[Anonymous]

Have you tried the changing the linking policy? Also, changing the maximum stay down distance has helped me.

[Muzzerboy]

I tried the whole range of stay down distances and most of the other settings in linking. And all the other tabs. I think it's getting upset when it encounters the cavity in the model, even though I haven't actually got around to machining that yet.

The lesson seems to be to avoid 3D adaptive for this kind of profile when 2D adaptive will do the trick. All rather confusing....

[Anonymous]

I have had to make different versions of parts in order to get the operations sorted out. 😞 For example having a version with holes and one without so the operations sort out.

[martin.dunschen]

Hi @Muzzerboy

I'm sorry to hear that you have these problems with the retract moves you describe. Would you be able to share the part here, for example by exporting it to a *.f3d file and attach here? 

One thing you could try is adjust the optimal load on the passes tab. If the optimal load is very small in relation to the cutter size it can happen that particularly when reaching curved contours the machining passes fragment, because the maximum load (based on the optimal load) would be exceeded if it machined all the way to the contour, so it might need several passes, and the transition moves might go wrong, resulting in excessive retracts.

Something else to try is change the tolerance (also on the machining tab), make it slightly smaller, that should also influence the number of retracts.

But I would still be very keen to get your part to work out if this behavior could be improved. I will update you as soon as I have a chance to take a look at it.

Martin

[Anonymous]

What everyone has said so far, is correct. But, I do think  you have the wrong operation given the geometry, If the features is indeed round as it appears, I would drill it first then use bore. I know a 3d adaptive is so tempting as far as ease of use, but Bore can kick a lot of butt on holes

[Steinwerks]

In conjunction with @martin.dunschen's great advice, I also recommend setting your Fine Stepdown to the same value as you Maximum Roughing Stepdown as this will eliminate the small vertical moves. Playing with Minimum Axial Engagement can also help all these little cutting moves. As I understand it, this is a stock triangulation issue, so tightening the tolerance will also help as it will generate a finer stock body to work with.

[Muzzerboy]

Hi Martin - thanks for the offer of help. As pointed out by Thomas, this is probably not the right operation for the task but I can't help thinking there may be occasions where a similar task will be required and 3D Adaptive is the necessary solution. Perhaps I'm overthinking it - or possibly an issue with the settings.

I've attached the bracket, although you can get a very similar effect with a trivial model such as the second part I've attached. It seems to be the addition of the rectangular cavity in that example that causes the issue. If you delete the cavity, the problem doesn't arise.

I fixed the problem in this instance by changing to 2D adaptive, although perhaps the bore operation would be even better. I suspect it's often a question of having the experience to know which operation to choose. I hope I'm getting there...

Thanks for the great support here!

Murray

[wheels600]

Not too fast with the bore operation........I will admit to being a new user but my experience with bore operation is holes that are not to spec....I  readily admit that perhaps i am missing something but beware.

[Anonymous]

It should be faster than adaptive and leave a better finish if see
similarly set . Both would have the same plunge depth,depth of cut and step
over if set the same with out the rapid ups. Up load your setting as an
adaptive and I'll make a bore to match. Adaptive is an easy pass to set
but not an optimal in all setttings

[Muzzerboy · Accepted solution]

John Saunders summarised the problem and the solution recently.

.