V2-10 appearing to drift off G-code Path, and then correcting

Hello Penta Forum,

I am writing here today because I have encountered a problem with my V2-10 appearing to drift off of the g-code path which has been given to it. This results in cuts increasing in stepover as the machine cuts, as well as small notches being left in our part when the machine “corrects”.

When looking at the mirror in the kinetic control simulator, I would watch as the tool center would drift off of the g-code path, while occasionally correcting itself and returning the center of the bit to the g-code path.

I was using G54 offsets which were all set to 0. The tool length offset was -78.27mm.

The program was made in Mastercam using an opti-rough tool path which was then converted to 5 axis using the convert to 5 axis operation.

A photo of our piece of stock is attached where the “correction notches” circled in red are shown:

I would have included the G-code in this post however since I am new to the forum, I am unable to attach files to this post. I would’ve also included a photo of the G-code in the Penta Machine Co simulator, however I am a new user and can only attach one photo.

Thanks for reading through this post and let me know what you think.

Hi @cbosenberg, I’ve made you a basic user so you should now be able to post your G code and any other images you think would be helpful.

Hello John,

Thank you!

Here’s a photo of the toolpath in the Simulator:
toolpath

and here is the G code (fair warning, it is a larger file):
Nozzle-Top5Axis-V2.txt (10.8 MB)

Thanks @cbosenberg! The tool path lines are fairly close together compared to your tool diameter. Are you certain the simulator is showing the tool drift? Generally, when a program seems to drift over the course of a program, its due to excessive feed rates and the stepper motors are skipping steps. This wouldn’t present itself in the simulator, though, because the control isn’t aware that steps are being skipped.

Is there a specific line number where it’s happening?

Hello John, thank you for your reply.

There doesn’t seem to be a specific line where this is happening, it appears to be happening randomly.

I’ve attached a modified version of your file (I also renamed it to .ngc so the simulator will pick it up automatically when you click the load in simulator button). The only difference is the comment line that describes tool 6. I set it to an engraver tool so the tip of the tool comes to a point. This should allow you to better visualize whether the tool is following the tool path when in mirror mode.
Nozzle-Top5Axis-V2.ngc (10.4 MB)

Have you tried slowing down your feed rate to see if that stops the drifting? Also, it looks like with the tool length and work offsets you provide the program wouldn’t run. Are you sure those are correct? Have you modified any of the configuration on your machine?

Thank you!

to answer your questions at the bottom of that last message:

I just did a run with the feed-rate slowed down and we still ran into the problem with the machine stepping over.

As for the tool length offset, there was a miscommunication between me and my co-worker and the Y offset in G54 was set to -7.412mm. Apologies for this, this allows the machine to run.

On the last run with the feed-rate slowed down, I managed to get a video of the step over happening. Again, when you look in the simulator this “step” doesn’t show up in the path.

Does the simulator show that movement in mirror tab? If it’s possible, can you capture a video of what the simulator shows when on the mirror tab when this occurs? Please display the G code at the same time by clicking the Toggle G code button on the right hand side of the simulator.

I would definitely recommend spacing your points out more. It will make your program smaller and faster to simulate. It looks like the control may even be struggling to keep up, so it may make your program run faster on the machine as well.

Sounds good, we will try spacing our points out more.

Here is the video of our program in the mirror mode:

The movements where the tool returns to the path are reflected in the machine itself.