2021 Curve 5 Axis Tool Path

[Corey Hampshire]

Hi guys,

I sent this question to my reseller yesterday, but haven't heard back  from them yet. I am trying to move over to 2021 from 2019 and am seeing an issue with Verify and Simulation. Please see  the attached video.

As you can see the tool is coming in from the wrong side of the curve. Also it seems to move to the origin when the op is complete. When I run just this toolpath alone in back plot, the tool comes in backwards on the approach and then flips to the correct orientation. Has anyone seen this before? Is it a setting that I am missing? 

Thanks in advance!

[Chally72]

The move-to-origin issue in simulation when the op is complete is a known issue and is either fixed in 2021 Update 1/Update 2 (I forget which, shooting off the cuff right now).

 

I'm guessing this path is locked to 4 axis in TAC. There are some toolpath transform issues in Verify/Simulation in 2020/2021 that occur when picking that "4-axis" dropdown from the TAC page in multiaxis toolpaths. Your reseller should be able to give you some more specific information here on exactly what's going on. An interim solution (which I know may very well not work for your file here) to specifically solve this simulation issue is to set Tool Axis Control to 5 axis, but then tweak the method of control such that it is still 4 axis motion. (IE, To chain with the chain as the centerline of the part, etc....   Again, the trigger here is choosing 4 axis in the TAC dropdown. I'm not sure of the status of those issues, and asking your reseller for updates/resolution is the best way to keep continued attention on them.

I think I've seen this file or a near-identical file a few times already this year and I know it has all the triggers to hit these obscure 4 axis issues, including toolpath transform/plane issues.

 

 

[crazy^millman]

Got a sample file you can share? I have tons of Curve 5 Axis toolpaths in 2021 and not had that issue. Have you checked your backplot setting and made sure they match? You don't have a vector picked the wrong way if using vectors? A surface reverse if using surfaces? Have you tried other 5 Axis toolpaths on that section and do they behave the same? 

[Corey Hampshire]

Thanks guys. 

Dylan,

I have updated to Update 1 so it is probably going to addressed  in update 2. You are correct in thinking that I have it tied to 4 axis. I am following a wireframe curve and pointing to the origin point to maintain tool orientation. 

Ron, I have double checked everything I can think of in 2019 to 2021 as far as settings. I may have missed a checkbox though. I will work up a watered down file and share it. I learned this trick from you so it is pretty fitting that you responded. It has always worked great and I use it to miss some of our fixturing. I thank you for all that you share here!

This is not a huge deal for me. I just skipped it in verify and simulation as it is tried and true. I also double checked the posted code and it is correct also. I can get by without too much of a hassle.

File will be coming shortly.

[Corey Hampshire]

Hopefully this link works!

https://drive.google.com/file/d/1k1z8rluERGThrE_HaLWO_Yu-8KkgfWv4/view?usp=sharing

[Chally72]

Just to clarify- this is a simulation-only issue and does not affect posted code. Backplot should be showing correct motion in all scenarios.

[Corey Hampshire]

The posted code is correct. If you run just that op in backplot the tool comes in incorrectly (upside down) but as you step through it, it flips the correct way and runs the rest of backplot correctly. If you run all three ops together, the orientation of the tool remains correct the whole time.

I don't know if that information will help or not, but figured I would pass it along.

[crazy^millman]

Have to watch out for the ref points in that toolpath. They will also mess with the backplot, but they do throw in indexes that are present on the screen and NC Code so need to make sure you have a method to verify the difference between them or be aware of the possible issue when running programs on the machine. 

Here is what machinesim shows with Y axis checked at the end of the motion. The toolpath without it respects the normal direction of the toolpath and all is good. With it checked it decides the WCS control the direction the ref point.  Same toolpath without ref points is the 2nd picture no collision respecting the last vector. 

 

 

[Corey Hampshire]

Update on this... I updated to Patch 2 this morning. First thing I did was try these tool paths. In backplot, the tool still comes in upside down when you run just the curve 5 axis path. 

In verify, the move to origin is gone as promised. However, the tool wants to run upside down still. 

Sending this off to my re-seller this am to get sent up.