Adaptive Clearing Leaving Thin Walls - When HSM Gets Scary (And Unsafe)

[colin.bastien]

I absolute love Adaptive Toolpaths and Vortex in Powermill, I try to run them as much as possible, I push my machine to the limit every day and strive for lower cycle times and better production.

One of the things I love everything about fusion is adaptive clearing, however there is one important thing thats been overlooked here. I see this often, and I wish I had a setting in my adaptive options to avoid this, when it leaves this tall, long and unridgid wall when clearing out material.

Id love to see in Fusion and HSM works an option to avoid this, where the toolpath would (when clearing all the material anyways) leave a set thickness to this wall (Say 1inch, so its rigid enough) then start from the end, working its way in, using slot clearing or a similar strategy, thus avoiding breaking ends, shooting thin sections all over the machine, or worse, pulling the wall into the tool and snapping a 1 inch carbide endmill running at 12 grand, 800 inch/ min. 

I just took a photo, this is a very weak example of what is happening, but same idea. I often get it much worse, this is just a photo I took before writing this to illustrate the problem.

Thanks in advance for any input.

[kb9ydn]

There is a discussion about this going on literally right now, over on the HSM forum.

https://forums.autodesk.com/t5/hsm-ideas/adaptive-machining-improvement/idc-p/7236499#M2196

C|

[colin.bastien]

Thats good its being talked about, I see it on almost every single Adaptive tool path. That should be at the top of priorities for Autodesk, Adaptive is a huge selling feature of this product. Thanks for showing me that forum post! Was that idea accepted and if so when will it be fixed?

[trutech.eng]

Yes I would like to see this corrected as-well.  I recently had a problem with a deep adaptive cut and when the cutter got to the tall thin area it would grap and break the tool.

[kb9ydn]

It has been accepted but there doesn't seem to be a definite solution yet or ETA.  It's not a simple problem so I wouldn't expect a fix in the next release.  My guess is that they will have something fairly soon though.

C|

[colin.bastien]

That really does seem unacceptable though to have a problem like that. I wish this were a priority for them, this bug devalues the adaptive clearing tool path so much. 

[kb9ydn]

@colin.bastien

If you haven't already you should read through that thread I linked.  It's a pretty hot topic right now and it's getting a lot of attention.  I don't want to make any claims for Autodesk but it looks like it may be addressed sooner than I thought.

C|

[cj.abraham]

Unfortunately there is not an ideal general solution to just "avoid" this result (It's not a bug, just a result of the mathematics). There are some creative ways to manage the sliver result, but the fact of the matter is that the thin sliver is unavoidable when using a constant engagement strategy on particular types of open-sided pockets. Here is an example photo of a solution for a 3 sided, open, rectangular pocket:

The solution involves first avoiding the result of the tall, thin rest material by leaving a a drafted boss, then machining away the boss in a separate operation ensuring that the tallest parts of the boss are machined first as if it were a closed pocket. In this particular case, the passes can approach the wall because the material will be supported.

Another solution would be to use 2D contour with multiple stepovers.

In short, it is not likely this will be solved in adaptive clearing, but there are ways to avoid it.

[ColinBastien]

So whats the point of an idea station then if your customers all agree something needs to change in your product, and your answer is "it is not likely this will be solved in adaptive clearing" because "It's not a bug, just a result of the mathematics".

I happen to know you have some of the smartest people in this world working at Autodesk, if the algorithm in your software causes the adaptive toolpath to create these long thin walls, then that algorithm needs to be changed, not your customers approach or rather ability to bandaid the problem when arising. 

This is 2017 and Autodesk is the leading software provider in Manufacturing and a variety of other industries. Your customers are competing in a global marketplace and fighting for work against countries with no standardized wage, health safety or environmental policies. Our only chance of competing is for reduced manufacturing time and reduced overhead costs. That is where Fusion 360 Adaptive Clearing found its place. In today's computer age there is absolutely no excuse for an inability to manipulate code in such a way to yield desired results. I'm very disappointed reading that comment from you @cj.abraham and seeing it posted as "future consideration"

I deeply hope you take this problem up with your software and application engineers and brainstorm a legitimate solution for this problem that your paying customers are requesting here and on the forums.

[cj.abraham]

@ColinBastien I suggest you read this thread in depth to see how we are exploring ways to mitigate the issue:

https://forums.autodesk.com/t5/hsm-ideas/adaptive-machining-improvement/idi-p/7090365/page/2#comment...