[magicniner]

I spent the best part of a day tearing my hair out assuming there was a problem with the model but then tried 3-Axis options on the problem surfaces and they all worked, potentially proving the model was OK so I started looking at the posibility that it was a software problem.
Has anyone else encountered a problem in 4- Axis Standard (and possibly found a solution) where BC works around any areas of a part below the centre line, leaving large chunks of stock untouched?
With some parts it's simply a case of using larger stock and moving the part to avoid any surfaces crossing the A-Axis centre of rotation and just making more chips. I'm currently working on something nominally S-shaped making it impossible to move both low sections above centre.

Paths generated -

I've experimented with moving the part around and verified that I can move it up or down far enough to get one or other face right. It's possible to vary the ignored areas slightly by choosing cutting around or along the A-Axis, but there are no feature settings that make any difference to the avoidance of any areas below the rotational axis.

Completed simulation -


Any help would be appreciated, even a confirmation that I need to get my cheque book out again ;-)
Regards,
Nick

[magicniner]

I forgot to include V25. Build 996

[RAF.]

So what is going on is that it would not be possible to mill a single throw crank shaft with counter weights?

[RAF.]

Yes I see what you are saying the tool path wants to omit anything below rotation axis

[magicniner]

RAF,
Yes, I just went and constructed a basic crank as a test, just in case it was compound curved surfaces below rotational axis causing the software to fail.



Sadly it seems to fail with anything below the rotational axis, even very simple jobs.
I wonder if there's a chance of a fix now they've moved on to V26, I only bought V25 this year and there was no mention of any serious limitations in the published info on 4-Axis Standard :-(
With any luck there'll be an easy fix,
Regards,
Nick

[ShortTrack]

Magiciner;

I have had similar issues in SIMULATION In V25-B996. As a workaround, I have been able to get it to simulate correctly if I break the machining along the X axis into segments. If I remember correctly, it simulated OK on V25-B769 - so I keep that version installed for this exact reason.

Mag, near as I could figure out - this was months ago, so I might remember poorly - BBCD has issues in 4th Axis when the part went from "thick" to "thin" to "thick" (highly technical terms here!) and I needed multiple passes to remove material.

It's worth a try & I hope this helps,
STrack.

[BurrMan]

Sadly it seems to fail with anything below the rotational axis, even very simple jobs.

Simple is simple.



If you want to post your model I would see if I can break it down for you.

[BurrMan]

BBCD has issues in 4th Axis when the part went from "thick" to "thin" to "thick" (highly technical terms here!) and I needed multiple passes to remove material.
.

What you remember is that the rotary toolpath is a finishing strategy, which has a step down, although, because it's a finish strategy, when you add steps, it adds steps to the entire toolpath, so it will add steps "up" from zero plane, equal to the down steps, so you cant really rough out these types of parts with it.

[BurrMan]

I have been able to get it to simulate correctly if I break the machining along the X axis into segments. If I remember correctly, it simulated OK on V25-B769 - so I keep that version installed for this exact reason.

Bingo. Although this is not a simulation thing, it's a toolpath thing.

[magicniner]

Simple is simple.



If you want to post your model I would see if I can break it down for you.

That's a very kind offer, this one fails too and is far simpler -





.bbcd file attached in .zip file,
Regards & Thanks,
Nick

- I've just tried this on an alternate HDD running V25 Build 996 in demo mode with the attached screenshot showing results.
If the demo gives 4-Axis Pro then this might be the answer.

[BurrMan]

Well, it's definitley not a "pro" thing, as I have full levels and produce what you see with your settings. I am taking a bit longer with responses because I can cut and sim the part, but
got worried about what I was seeing in the sim vs. what would actually happen at the machine with regards to the features settings, so I went and ran some of the output in an actual 4th axis and got my expected results.
However, I want to test a few other scenarios and some more complex shapes (It would be nice to have your part to address) before I can speak to it..

The other short answer to one of the questions is that the multi axis toolpaths can do this in different ways too.

"Why did it work on the demo mode part?"

It appears to me you have that part set at the rotation axis' zero point, instead of out and away from it.

[BurrMan]

I did some more testing with something more than the "simple" file. The toolpath will definitely not cut below the zero point of the rotation axis.

I am unsure if this is a bug in the behavior or intended/needed for some reason. Maybe Al can look at it and talk with someone in dev before a report is made. Or just make the report.
O wont report it until I can confirm it. It made sense to my brother at the machine. He said, "of course" it does that.

Anyway, here is a link to a file that will handle the "double undercut" (it's what I call it). I made a shape that will emulate yours. You can view the settings in it.

http://sdrv.ms/1gDFwp7

Essentially, you need to set a value in the rotary features "rotary axis base point" on the patterns page. Then, the model needs to have features added for each section that would require
a new base point to avoid the "undercut".

Attachment 204378

In this file it required 3 sections. I layed out some lines to give me the dims to use in the 3 features. The 2 vertical lines are my start and end points for each feature (at or just before the undercut).

The 2 horizontal lines are the to extra base point measurments for the first and last features.

Anyway, maybe BobCad will chime in and say it should do the undercuts, but here is what you need to get your file done.

BTW: that looks like a trigger and it may be better served with an index as opposed to a rotary path. I try to view the rotary path as more of an organic or symmetrical toolpath.

[magicniner]

BurrMan,
Thanks for the help and excellent suggestions, the problem part is a beavertail which I currently manually machine using hand ground custom cutters, multiple fixtures and a rotary table - image attached.



I'll have a look at the file you've very kindly posted and see what I can manage,
Regards & Thanks,
Nick

Edit - Having tried a variety of settings in Base Point and just moved the ignored section about I looked Base Point up in help, presumably this is written for someone better than I as it doesn't explain what changes the setting might make and why, I've attached an image of the Help entry for base point in case anyone might be able to help me understand what how the setting affects tool paths.

[BurrMan]

Ah, the ol' it's a beaver tail and not a trigger trick, eh....

That picture has me agree with the rotary more too......

Let me know, and we'll see if we can get some input from BobCad. Thinking about the toolpath and results more, I'm leaning towards the toolpath looking at surface normal, but also accounting in the rotation axis, which can have the normal not read well when coming from under the zero line. They "could" probably look at this.

ANyway

[magicniner]

I think I'll have to wait until Al gets back from the big machining show he's at this week for a definite answer on this.
At least I can take a break from it until then ;-)
Regards,
Nick

[RAF.]

Nick

Would it not be possible to do this part indexing to 4 faces in 3D it still could be a walk away totally automated part I'm thinking.
Might be able to just index the problem areas and 3D cut.

I guess I'm thinking from the cheap seats at the price of 5th axis

RAF

[magicniner]

RAF,
Yes, good ideas, I've had a look at indexing and a fast & dirty test suggest it should work, not as elegant as Rotary though.
Indexing for 3 sides and rotary with limited angular travel, with the part shifted so the compound curved face is fully above the axis, might fit the bill.
I've got to get some proper paid work done though!
ATB,
Nick

[RAF.]

So what is going on is that it would not be possible to mill a single throw crank shaft with counter weights?

This is the best I can do for now to build a Crank with 4th axis standard with a single rotation axis, 5 inch stroke with a 5.2 in Rod Journal:nono:
Hopefully there will be a fix for this.

[BurrMan]

Is there something wrong with the rotary toolpath I posted? It did the double undercut......

[RAF.]

BurrMan
Maybe I'm not understanding the axis points.
Is there a way to under cut by using the main journals in my post as the rotation point or will the stock physically have to be moved on the forth axis to make the Rod journal smaller?
Am I missing something here?

Thanks
RAF