Post-Processor bugs (Fusion360)

Hello!

First post on the forum and as many else I’m so excited to get going with my machine! :star_struck:
For now, I have been making a few parts to get to know the machine.

I first used Easel to get a Gcode as fast as possible. This is the result ;

Since then I have been experimenting and making nothing seriously. But in my time playing around I have encountered two issues.

  1. When making holes with fusion it starts to almost work against itself. This leads to skipping steps and misaligned milling patterns (the circle at z= -5mm are not aligned with the circle at z= -10mm)
    Have tried to use the OF post-processor, GRBL, and easel. Easel works.
    So the machine is working, therefore the next suspected is the code/software…

Does anyone have any clues?

  1. My holes are not the correct size.
    I tried to fit a 22mm bearing into a pocket, but the pocket is too small. I’m guessing that has to do with fine-tuning of the machine? I have yet to follow this post made by@cyberreefguru
    (Profile - cyberreefguru - Onefinity Forum)
    Testing and tuning OF for accuracy
  • noen of the F360 or Easel files made the hole bigg enough(sligthly under)

Here’s a little video;
(in the next post due to: “Sorry, new users can only put 2 links in a post.” :slight_smile: )

Here’s the video link ;

Have you updated to the latest firmware? The vibration issue should be fixed with the firmware update.

I See!
I read the topic about the new firmware and thought “Vibration issues, I aint got no vibration issue!” :sweat_smile: :grin: But I clearly might have misunderstood what they meant with vibration issue!

I will try the new firmware and check in here later on ! :+1:t3: :crossed_fingers:t3:

Let us know if the update fixes it. Mine was doing the same thing, but I ‘fixed’ it by slowing the speed way down and I was able to finish my wasteboard. I haven’t updated my firmware yet so can’t speak from personal experience if the firmware resolved the issue.

Yes it did! The 1.0.6 beta software fixed the issues i documented in post nr.2

Now I need to fix the accuracy, problem nr 2.
Will follow the topic “Test and tuning” and see how it turns out then!

1 Like

Good tip! I Will measure and see when I do the testing! :+1:t3:

So I’m a bit lost at the tuning for my cnc…
I tried to follow the guide from @cyberreefguru as mentioned in post 5.
This did not help me achieve higher accuracy.
I found that my machine wasn’t completely squared up, so that got taken care of.

If somebody got a hint at what I could have a look at next up I would appreciate it! thx! :slightly_smiling_face:

Time for an update;
After spending quite(!!!) some time trying to tune-in the cnc, I’m not sure where to go.

What we have found out so far ;

  • When traveling 800mm its off by less ~0.1mm (Neglectable and to small to measure very accurate on the ruler?)
  • When making a small square(45mmx45mm), it makes the same amount of error, apx. 0.5mm…
    • Measured in at (
  • When making a larger rectangle(20mmx140mm), it makes the same amount of error.
    • Measured in at (
  • it does not seem to add up the offset for each cm traveled. (For instance, 1mm error for each 10mm traveled → 100mm traveled = 110mm measured)

Here is a video and some photos. Maybe you see something that would be of importance?


Here you can see the minor offset it makes after traveling 80 cm. (started at 1cm)

Picture of some of the tests. on the left is a F360 with the new post-processor. You can see the designed paramertes in the square. The rectangle is design to be 140mm * 20mm.

The video might be pointing to the problem. That seems like a lot of movement on the dial indicator when you flex the table top. I had some issues with mine initially from not enough rigidity of the table top. The machine could be ‘torquing/winding’ the table on direction change?

In my case it was laziness and I had not put enough screws in securing to bottom MDF. My top is 8020 aluminum frame, braced with I beam to prevent sag, partially torsion boxed, 3/4 MDF on top of that, another layer of 3/4 MDF that is a bit bigger than the footprint of the machine. And a 3/4 spoilboard on top of that.

Once I banged more screws in and got rid of the flex in the top I was good to go. Also, put lags in the feet that went through 2 layers of MDF instead of just 1.

2 Likes

The take flex seems sketchy at best. Have you checked to ensure all the guides/bearings are tight? If you push on the axis is there any slop (likely had to feel at that small of a shift)

Agree on the flexing tabletop!

  • Have to sort that out in the future some time…

But with such low feeds the dial indicator should have gone back to zero although the table can flex a bit, or am I wrong?
I had the majority of my body weight on the table when that flex occurred!