4th Axis Incomplete Finish Cut

Topics related to wrapped rotary machining in Aspire or VCarve Pro
HBruns
Posts: 12
Joined: Tue Jan 15, 2019 9:00 pm
Model of CNC Machine: Axiom AR6 Pro+
Location: Texas, USA

4th Axis Incomplete Finish Cut

Post by HBruns »

I cannot get the FINISH tool path to complete a cut.
The machine (Axiom Precision AR6) finishes the file, yet the piece is not done.
Everything looks great in the software (VCarve Pro 10.5), yet I have never had a piece that was completely cut.
I have re-downloaded & re-installed the 4th axis post processor, and I am sure to select it when saving the files.
Am I doing something wrong?
The image shows the most progress I have had yet -

Image

User avatar
TReischl
Vectric Wizard
Posts: 4083
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 Build 48X36X10 RP 2010 Screenset
Location: Leland NC

Re: 4th Axis Incomplete Finish Cut

Post by TReischl »

Can't see your image, sorry.

Check where you are setting your Z zero, most of us set it to the centerline of a rotary project. Not the surface.
"If you see a good fight, get in it." Dr. Vernon Jones

User avatar
adze_cnc
Vectric Apprentice
Posts: 1656
Joined: Sat Jul 27, 2013 10:08 pm
Model of CNC Machine: AXYZ 4008
Location: Vancouver, BC, Canada

Re: 4th Axis Incomplete Finish Cut

Post by adze_cnc »

TReischl: Probably browser related. Image: http://www.hbruns.com/images/TMP/unnamed.jpg

HBruns
Posts: 12
Joined: Tue Jan 15, 2019 9:00 pm
Model of CNC Machine: Axiom AR6 Pro+
Location: Texas, USA

Re: 4th Axis Incomplete Finish Cut

Post by HBruns »

Z-zero isn't the issue (ran into that already and fixed the issue).
The part is correctly carved, but the carve ends before the part is complete.

kyeakel
Vectric Apprentice
Posts: 99
Joined: Tue Oct 28, 2014 3:42 pm
Model of CNC Machine: Camaster Stinger Shopbot PRT 9648 STV480
Location: Curwensville, PA

Re: 4th Axis Incomplete Finish Cut

Post by kyeakel »

Can you run the code in a g-code simulator? There are some online. This would tell you if the code is correctly generated from the post processor. If not, I’d look at the code with a text editor and verify that the A axis commands are calling for a full revolution.
CAMaster Stinger 1
Shopbot PRT 9648
STV4800
Aspire
Cut2d,Cut2dLaser

User avatar
TReischl
Vectric Wizard
Posts: 4083
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 Build 48X36X10 RP 2010 Screenset
Location: Leland NC

Re: 4th Axis Incomplete Finish Cut

Post by TReischl »

Just a WAG, but I would first guess that your A axis motor is stalling.

As I look at the picture (thanks Adze) I note that the cut is tapered. Narrow at the large diameter ends and wider at the smaller diameters. When the tool is out at a large diameter it has a lot of leverage on the motor, not so much when it starts getting down to smaller diameters. I would expect it to look a little more ragged, but that is my guess. This would only be possible if you were cutting around the piece, not along it's length.

Highly doubt it has anything to do with the software.
"If you see a good fight, get in it." Dr. Vernon Jones

gregk
Vectric Staff
Posts: 260
Joined: Mon Mar 05, 2018 12:34 pm
Model of CNC Machine: None

Re: 4th Axis Incomplete Finish Cut

Post by gregk »

If you're using a thumb drive to transfer G-code to your machine, it may be worth checking that the file was fully written to it. The finish toolpath is most likely way longer than roughing toolpath and if you didn't use 'safe remove' on your thumb drive maybe it didn't write complete file.

Greg K

HBruns
Posts: 12
Joined: Tue Jan 15, 2019 9:00 pm
Model of CNC Machine: Axiom AR6 Pro+
Location: Texas, USA

Re: 4th Axis Incomplete Finish Cut - FOUND A BUG

Post by HBruns »

I'm pretty darned sure I found the problem.
Short answer - There is a bug in the software.

Longer answer -
The tool path that produced the incomplete part is a FINISHING RASTER tool path with a 0-degree raster angle.
What SHOULD happen is the Z-axis lowers to the cutting depth and the A-axis rotates the part 360-degrees while the Z-axis changes cutting depth.
After a full rotation the Y-axis increments the step-over, the A-axis rotates while the Z-axis does its thing.
There is supposed to be no Y-axis motion except for the tiny step-over increments every time the piece does one full rotation... this is for the entire finishing tool path.
- AND THAT IS OBVIOUSLY NOT HAPPENING.

I created a new tool path - another RASTER FINISH tool path with a 45-degree raster angle.
This tool path finished correctly while the same tool path with a 0-degree raster angle runs off the rails.

I did copy a newly-calculated finishing tool path (0-degree raster angle) to the controller.
It behaved exactly the same as when run from the USB stick (U drive), so that has not been the problem.

User avatar
Rcnewcomb
Vectric Wizard
Posts: 4704
Joined: Fri Nov 04, 2005 5:54 am
Model of CNC Machine: 24x36 GCnC/WinCNC with ATC
Location: San Jose, California, USA
Contact:

Re: 4th Axis Incomplete Finish Cut

Post by Rcnewcomb »

I’d recommend sending you file to support@vectric.com so they can look into it.
- Randall Newcomb
10 fingers in, 10 fingers out
another good day in the shop

User avatar
JoeBlow
Vectric Apprentice
Posts: 94
Joined: Wed Nov 21, 2018 11:19 am
Model of CNC Machine: AxiomAR6Pro

Re: 4th Axis Incomplete Finish Cut - FOUND A BUG

Post by JoeBlow »

Having the same machine and rotary setup as the OP made me very curious about this. I have my rotary orientated along the Y axis and have not done a finish toolpath with a raster angle of 0. I use a raster angle of 90. I drew up a pawn, modeled with a 2 rail sweep and toolpathed with a finishing raster strategy set at 0 degrees. Sure enough it cut just fine. Rotated 360 degrees, Y moved according to my stepover and then rotated 360 degrees and proceeded to cut. About every 2-3 rotations Z would raise up and lower back down before commencing the next 360 degree cut. Otherwise all seemed normal.

Not a particular strategy I would choose going forward as I think it would be more prone to losing steps.

As far as why this might be occurring...could be due to Axiom rotary users needing to change the RT Type parameter in the Rich Auto controller.
Beautiful Things Function Better

User avatar
TReischl
Vectric Wizard
Posts: 4083
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 Build 48X36X10 RP 2010 Screenset
Location: Leland NC

Re: 4th Axis Incomplete Finish Cut

Post by TReischl »

It would be nice if the OP posted at least a portion of the G code. Watching something run and then declaring there is a bug is not very "scientific".

A look at the g code will confirm whether the software is outputting full rotations around the axis or not. It is very easy to read. Just look at the A commands, they should go up, then down, up, down, etc. The max would be 360 and the min 0.

Myself, I am very cautious about declaring something to be a bug. Keeps me from having to wipe egg off my face a lot.
"If you see a good fight, get in it." Dr. Vernon Jones

HBruns
Posts: 12
Joined: Tue Jan 15, 2019 9:00 pm
Model of CNC Machine: Axiom AR6 Pro+
Location: Texas, USA

Re: 4th Axis Incomplete Finish Cut

Post by HBruns »

http://www.hbruns.com/images/TMP/3DFini ... ter425.mmg
http://www.hbruns.com/images/TMP/unnamed_01.jpg


G-Code file -
http://www.hbruns.com/images/TMP/3DFini ... ter425.mmg
"Finished" piece -
Only the top of the chess pieces crown finished, and one line down the length (y-axis). everything else was done by the roughing toolpath.
Image

The roughing toolpath had no problem.

User avatar
TReischl
Vectric Wizard
Posts: 4083
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 Build 48X36X10 RP 2010 Screenset
Location: Leland NC

Re: 4th Axis Incomplete Finish Cut

Post by TReischl »

Took a long hard look at the g code file.

It is commanding full rotations of the A axis. It steps its way around changing Z as it goes. I could not find anything wrong with the file at all.

You have something else going on at your machine. I know you do not want to hear that, but it is what it is.

Edit: There is something a little odd going on in that file. The Z axis is up and down a lot with little movement on the A axis

N24420G1A140.518Z12.621
N24430G1A140.342Z12.899
N24440G1A140.166Z13.034
N24450G1A139.814Z13.197
N24460G1A139.463Z13.295
N24470G1A138.935Z13.360
N24480G1A88.813Z13.360
N24490G1A88.285Z13.295
N24500G1A87.934Z13.196
N24510G1A87.582Z13.033
N24520G1A87.406Z12.894
N24530G1A87.230Z12.474
N24540G1A87.054Z11.916

Notice that at about 139 degrees to about 89 degrees one cut is made at a depth through about 50 degrees of motion. That pattern repeats itself consistently throughout the file. I would guess the jittery Z motion is the result of using an STL file? The Z motion is very small usually about a tenth of a millimeter (.004 inch).

What I am wondering is if the control is getting over run. I would be sending this file to the folks who sold you the machine and ask them to give it a go.
Last edited by TReischl on Tue Mar 16, 2021 12:58 am, edited 3 times in total.
"If you see a good fight, get in it." Dr. Vernon Jones

SteveNelson46
Vectric Wizard
Posts: 1296
Joined: Wed Jan 04, 2012 2:43 pm
Model of CNC Machine: Camaster Stinger/WinCNC
Location: Tucson, Az.

Re: 4th Axis Incomplete Finish Cut

Post by SteveNelson46 »

I have made many, many rotary projects using Aspire and if there is a flaw in the software I have not discovered it. It appears to me that you are using a flash drive that is either flawed or does not have the capacity to hold the complete g-code file. There could also be a problem with the imported .stl file. I have found that most chess .stls are really make for 3d printing and are prone to problems like these. This is my latest rotary project that I made just yesterday. It's a knob for a tray for Michael's Book Pile Project. It measures 9/16" W. by 7/16" H and is all one piece. Including the mounting stem. I could have just bought one but what's the fun in that?
Attachments
Knob.jpg
Steve

HBruns
Posts: 12
Joined: Tue Jan 15, 2019 9:00 pm
Model of CNC Machine: Axiom AR6 Pro+
Location: Texas, USA

Re: 4th Axis Incomplete Finish Cut

Post by HBruns »

TReischl -
The red section you highlight is to make the dimples in the crown of the chess piece.
Most of the file deals with this short section.
When it is done with that section, the piece becomes symmetrical around the Y-axis.
At line 5516 the code changes to a Y&Z step, then a full rotation, then a Y&Z step, and it rotates back.
This repeats until the Y-axis increments to a value of 96.272 at the end of file (line 6514)
That 96.272 is in mm I assume, which corresponds closely to the height of the piece.
Each Y-step is 0.159 (mm I assume again) - this corresponds to the ~0.006" step-over value for the tool.

To my uneducated-in-G-code eye, it looks like the G-code is OK.

The issue is that the machine is not rotating the A-axis in the sections where the piece is symmetrical around the Y-axis.
This is also happening at the very top of the chess piece where there should be a small ball.

Image of incomplete cut -
www.hbruns.com/images/TMP/unnamed_01.jpg
Image

Preview Image of what the piece is supposed to look like -
www.hbruns.com/images/TMP/QueenPreview.png
Image

Post Reply