Can't get a decent toolpath for rest roughing

[SlaveCam]

I'm trying to machine a large planar surface to the side of a cylinder. The cylinder has been defined as stock and I use area rest roughing but keep getting 5-10 times slower toolpaths than if using the "conventional" method: a simple contour for each Z-depth and adding multiple radial passes as Z becomes deeper. The tool is a D80mm high feed mill.

Is it just that there is no proper toolpath for this kind of machining in Mcam?

REST ROUGHING.mcam

[Manofwar]

I took a quick look at this, it appears that your are trying to use a 2D dynamic path. This works great for flat surfaces, however it does not detect 3D chains or stock. It simply takes the outer most boundary at maximum z height and works down from there. Do you happen to have a MC seat for 3D? There are ways to get around this, but a 3D path would be much simpler and faster.

[SlaveCam]

Hello,

It's a 3D toolpath, under the 3D tab. Area roughing was the only one that even recognized the stock the way it should - optirough just plunged (ramped) from the top of the stock into the final depth.

I asked this same question maybe 2 years ago, but have forgotten everything. There was a thread here about roughing a ball into a half-sphere but I can't find it anywhere and all my examples and ops are gone due to employer change. Lesson learnt: ALWAYS put your examples in the cloud!

[Manofwar]

I dug into this a bit more. Sorry for the confusion. I reopened this and realized you were indeed using a 3D path. However, there are some problems with it. 

 

First off: the problem with the path ramping into the part. The problem here is when using stock the containment boundary locks to "Stay Inside" this button is poorly named, and confused me for a long time. What you need to do is set the "compensation to:" to outside and type in an additional offset distance. Typically I type something fairly excessive into this field as it is only looking at stock, it won't actually make a path that is cutting air out in the middle of nowhere. This should fix all of the ramping as it can now start outside the part.

 

Second: Conventional will probably be faster for one reason. It is not running constant stepover like the 3D paths are. You said you were just milling right down the center of the part until you got to a depth where you needed to move over. So, radially your stepover was increasing with each z depth step. I'm not saying this method is bad. It just can't really be an apples to apples comparison.

 

Third: I don't believe any of the 3D paths are going to work the way you want them to. This is for one reason. They aren't really designed for variable radial loads, they're designed to maximize tool life. You could possibly 'break' one of the paths to get it to work. Or there might be in option in a legacy path that I don't know about.

 

Sorry this isn't that helpful.

 

Caleb

[JTaylor]

Is this more what you are looking for? I don't have any experience with feed mills so I can't really say if it's better than a straight contour.

REST_ROUGHING-JT.mcam

[SlaveCam]

5 hours ago, JTaylor said:

Is this more what you are looking for? I don't have any experience with feed mills so I can't really say if it's better than a straight contour.

REST_ROUGHING-JT.mcam

That's exactly how it was done two years ago in a similar example, it seems I just forgot to use Z-limits. Thank you.