feed rate optimization - what am I missing

[PeterBelfanti]

Why am I not getting a feed rate reduction in the tightest corner of this pocket?  it reduces on larger radius but not the tightest ones?

Tool dia .125

Corner radius of part .081

[Steinwerks]

Deselect 'Only Inner Corners' as I believe this has been bugged for years now.

[PeterBelfanti]

thank you for the tip..  I tried that just now and no difference in the path

[Steinwerks]

Can you export as a .F3D and share the file here? That or make a quick dummy file with the same geometry that exhibits the same issue, it would be most helpful.

[PeterBelfanti]

part attached... one of the last paths in the current setup.

[Steinwerks · Accepted solution]

I had to drop the Maximal Directional Change all the way down to 5° which I believe is a function of the tool radius and the corner radius, but then it worked properly:

Edit: since there's quite a bit of material left after the previous roughing (you may want to consider reducing the minimum radius on that toolpath) if the Maximal Directional Change is reduced to 1° it will turn the full corner at the reduced feedrate.

[PeterBelfanti]

Thank you.. not greatly intuitive but if it works..

[Steinwerks]

I can't be sure without some developer insight (want to chime in here, @fonsecr?) but I believe this has to do with the way that the directional change is calculated in conjunction with the tool radius.

I wouldn't be surprised in the slightest if the directional change distance calculated is a percentage of the tool radius interpolated as a straight line (think unwrapping the circumference), and so in your case where the corner geometry is significantly larger than the tool's radius, the calculated change in direction is quite small as the tool travels that radius.

I could be wrong, though, and now I'm curious and want answers 

[fonsecr]

Hmm, that doesn't look right.

I'll check it next week.

[fonsecr · Accepted solution]

I see the problem now.

The inner corners is actually a lot of small line segments. Not an arc. The feed optimization currently support finding a global curvature across multiple segments.

You can turn on Points in the Simulate command to see this.

I dont have a solutions for this problem for now. I can bring it to the attention of the team - but I dont see this getting high priority at this time. But at least we now understand why it happened 🙂

[fonsecr]

Just one additional comment.

You can turn on smoothing which will fit arcs. And then these will get reduced feed due to the small radii. So an option when it is ok to use the Smoothing feature.

[Steinwerks]

Hah, I thought it was on. Oops. Setting it to .001" makes generation correct, as well as reducing the toolpath from 99.6kb to ~44kb.

[PeterBelfanti]

My fadal only has 128k. That matters! And it makes me sad.. back to dncx for me

[PeterBelfanti]

Fonsecr

For your reference the model was 100% native to fusion 360. No import anomalies to create all those points.

[fonsecr]

The toolpath strategies are using the triangulated surface to generate the toolpath which is why you get these points by default.

[PeterBelfanti]

The dirty underlying tessellated model.. needs a pretty dress.. kidding.. thanks for your help!!

[joshCPH3D]

I have searched the forums for a couple of hours now. I've tried everything to get feed rate optimization to work. All I want is the tool to slow down in the inner corners of a square pocket. I've got my max directional change set to 1 deg, reduced feed radius at .001, distance at .5" and a large reduction in feed. My corners were drawn using the sketch fillet tool. I've tried everything. What am i missing here? I have attached a fusion file so anyone who wants can check out whats going on. Thanks in advance for any help!

[seth.madore]

It's a deceptive field, that "Minimum Radius". One would think that it infers any radius larger than the entered field would be subject to a reduction in feedrate. That is not the case. Rather it's stating that anything SMALLER than the entered value would be subject to reduction

So, set it to a larger value. Simple math: radius of feature minus radius of tool. In this case, setting it to .064" gives me this:

[joshCPH3D]

Thanks so much! I've been using Fusion for years now and I've never taken the time to ask.