Sainsmart 3018 making mistakes when cutting helical ramp cuts (I J K errors?)

[superrabbit_5]

Hello, 

  This is quite possibly a ID-10T error but here goes: 

When cutting helical ramp cuts on my 3018 cnc the machine starts fine, cutting in the expected downwards spiral, however it will occasionally (with repeatability) make mistakes with the IJK code. 

Rather than cutting a downward spiral, the cnc will attempt an 8 shape. once making a mistake it will return to the intended spiral. It is as though it is using the XY Coords from the previous line. Normally this has happened well into the project - destroying my stock 😞  but the latest bit of code it happens almost immediately so I thought I would seek help.

I have posted the code below and it shows up fine in any program I simulate it on so I think the error is in the controller missing or misreading lines but I am relatively new to this. 

G90 G94
G17
G21
M03 S1000

(line above has been changed as my controller seems to prefer this format to the standard Sxxx M3)
G54

Z1

(line above added as sometimes initial positioning assumes 100% flat stock and leaves marks)
G0 X106.598 Y36.962
Z2
G1 Z1 F900
G3 X104.521 Y25.077 Z0.338 I-1.039 J-5.942 F433.3
X106.598 Y36.962 Z-0.324 I1.039 J5.942
X104.521 Y25.077 Z-0.985 I-1.039 J-5.942
X106.598 Y36.962 Z-1.647 I1.039 J5.942
X104.521 Y25.077 Z-2.309 I-1.039 J-5.942
X106.598 Y36.962 Z-2.971 I1.039 J5.942
X104.521 Y25.077 Z-3.633 I-1.039 J-5.942
X111.592 Y31.019 Z-4 I1.039 J5.942
X99.527 Y31.019 I-6.033 J0
X111.592 Y31.019 I6.033 J0
G1 X111.593 Y31.429 F900

The mistake is equivalent to: 

G90 G94
G17
G21
M03 S1000
G54

Z1
G0 X106.598 Y36.962
Z2
G1 Z1 F900
G3 X104.521 Y25.077 Z0.338 I-1.039 J-5.942 F433.3
X106.598 Y36.962 Z-0.324 I1.039 J5.942
X104.521 Y25.077 Z-0.985 I-1.039 J-5.942
X104.521 Y25.077 Z-1.647 I1.039 J5.942

(i have edited the above line to represent the mistake)
X104.521 Y25.077 Z-2.309 I-1.039 J-5.942
X106.598 Y36.962 Z-2.971 I1.039 J5.942
X104.521 Y25.077 Z-3.633 I-1.039 J-5.942
X111.592 Y31.019 Z-4 I1.039 J5.942
X99.527 Y31.019 I-6.033 J0
X111.592 Y31.019 I6.033 J0
G1 X111.593 Y31.429 F900

Some clarification points

-Z axis operates correctly throughout

-The cut starts with a spiral downwards climb cut as expected. 

- I have had this problem repeatedly (with repeatability) and it is not apparent (to me) what causes it. Previous code with helical ramp cuts have operated fine until, in one case at the end of a 15 minute pocket cut, suddenly making this exact same mistake.

- probably not relevant but I am using a 12.7mm (1/2 inch) straight cut double fluted bit 

- I am using an offline grbl offline controller as my laptop isn't going near the dust factory that is my workshop! 

Could be a controller problem? 

Thanks a lot in advance for any help. 

[seth.madore]

What post processor are you using, and can you share your Fusion file?
File > Export > Save to local folder. Return to thread and attach the .f3d file in your reply

[superrabbit_5]

Hello,

thanks for your reply. I am using GRBL, I have included screen shot of post settings and the fusion file. I have stripped away the other cuts etc for clarity but I was not using the other cuts when the error occured.

 

If they might interact in some way i can send the whole file. 

Thank you

There are two

'Warning: Lifting retract height to safe height.'

errors in the tool path but i have struggled to remove them and they did not seem relevant in the simulation. 

[seth.madore]

And to confirm, you ARE using a GRBL controller?

[superrabbit_5]

yes! Definitely GRBL! 🙂 

it is a VERY cheap controller, so i am somewhat suspicious of it, but it seems to run fine with everything else.

[seth.madore]

Is the part you shared also one and the same with the code snippet above, or is that from another project?

[superrabbit_5]

The code was definitely from that project, although i have been playing around with cuts to try and fix the problem so apologies if I sent the wrong version. The same error occured with every version of the cut i tried. always with the helix ramp cuts. 

I have narrowed the problem down to the offline controller. I ran the program from my laptop without a bit installed  and it ran perfectly. Sorry for wasting your time looking for coding problems, I really appreciate the time and effort you have gone to though.

FIX:

For others who run into this problem I am running a sainsmart 3018 clone, the controller is pictured below. I have been unable to fix the problem using the controller unfotunately. loading the Gcode into candle and running the code from a computer did work. 

If anyone has run into this problem and has a solution I would love to hear it but in the meantime it is not the end of the world to set up a workshop computer. The benefits of running code from a computer outweigh the issues with setting one up for me. 

Thanks again for your help.

[jtaylorisu]

I’m running into the same problem with my offline controller and also found out that running the same code through candle works just fine.

When using the offline controller, it will work fine for a while, then mess up the Z-height on some portion of the tool path in a pocket. It screws up the z height in the same xy locations on each pass. Then eventually the machine stops before program complete. 

Anyone else have this issue and found a solution?

[jtaylorisu]

Found another post stating that the arcs overload the 3018 offline controller.  I unchecked the allow helical moves in the Fusion G-code post process window.  Now all moves are x/y/z commands and it ran with no issues.