Consistency with cuts

Hi again!
So I’ve started making some cuts but struggling with the consistency. I’ve a gcode that has some small pockets but the machine is not cutting well.

I’m finding that after calibration (fitness of 1.05nnn) I’m getting situations where the cables appear to loosen. I’ll reset the machine (full power down, retract, extend, apply tension) and it will work for part of the cut, then cables loosen and system will stop. I’ve had this occur 5 times this morning.

And suggestions as to what is going on? Pictures and gcode attached.

Thanks,
frank.

warre_box_1x10_Full.gcode (147.3 KB)


1 Like

Bar is working on some things, watch for a mid-week firmware update.

David Lang

2 Likes

You might try increasing the current threshold and let us know if that helps.

2 Likes

I’ll push it up to 1500 and see what happens.

1 Like

Updating to 0.72.1 might help too. Generally working on making the belts tighter is my current focus :slight_smile:

1 Like

maybe try lowering your plunge rate, see if that helps?

1 Like

Also what kind of bit is that you have installed?

1 Like

1/4″ Carbide Compression Bit from cncbits.ca

1 Like

hmm. I never used that type before have you? Looks like they require a dialed in toolpath and other considerstions to use correctly over a more simple endmill.

1 Like

Bit of an update - 0.72.1 appears more stable; Not encountering the same issue as before. The cables appear to be a lot tighter than previously. Cut is running now - hopefully I can share a successful run later!

Thanks all for the suggestions and Bar for the FW update!

No luck - made some good progress then went awol… will try again tomorrow.

[GC:G1 G54 G17 G21 G90 G94 M3 M9 T0 F300 S24000]
[GC:G1 G54 G17 G21 G90 G94 M3 M9 T0 F2050 S24000]
[MSG:INFO: Reset during file job at line: 894]
[MSG:ERR: Emergency stop. Update function not being called enough.1002ms since last call]
No heartbeat from machine in 10

I think I’m still on 70, but something I’ve noticed is that bottom right likes to develop slack while jogging.

It seems to resolve this while running G code, getting tighter as it runs until it’s not noticeable anymore.

I’ve taken to telling it to apply tension before performing cuts.

I did 3 cuts yesterday where I had it take 4 passes for cutting profiles and couldn’t see any wobble in reproducing the path with each pass.

I’ve got the retraction power set to 1700.

How many passes are you telling it to do each cut in? It may be more accurate if you remove less material at a time.

1 Like

And from what I read about compression bits the depth of cut needs to avoid the transition area of the bit or you might get instability, especially in a ~floating sled.

So maybe your toolpath is crossing into that region where your bit flutes change from upcut to downcut.

chatgpt–>You typically want to set the depth of cut so that it extends slightly beyond the transition area of the bit. This ensures that the part of the bit with downcut flutes is actively cutting into the bottom surface of the material, while the upcut portion handles the top. This setup maximizes the compression effect, reducing tear-out on both the top and bottom surfaces of the material.

For most compression bits, the transition area is usually located around the middle of the flute length. Therefore, the depth of cut should be at least equal to or greater than half the cutting length of the bit to fully engage both cutting actions. However, specific dimensions can vary based on the bit design and manufacturer recommendations, so checking the specific bit’s specifications or consulting with the manufacturer is advisable for optimal results.

Just did a test with .77 and looks like I’m getting the same issues as posted above.

First plunge for the cut w 1/8 compression was .125inch.

Is the sled staying flat as you cut? (and are the belts tight enough to not
sag?)

if you can get a video as this is misbehaving it would help.

David Lang

Good new / Bad New:

  1. Good: retracting and extending then tensioning without a bit (or z axis up) seemed to help plunging accuracy.

  2. Bad news: once i ran my gcode it 1% well but then started a second plunge and ran off the map!

  3. Bad news: I hit pause before it got too far, then I asked it to pull up on the z axis and it just DROPPED while pulling up. Check out the cut getting less deep as it drops.

4: Good news: i caught it and it stopped before if completely fell of the ‘map’.

  1. Bad news: this was the second time this happened today.

I popped in the serial data if that helps anybody with more savvy than me.


Maslow-serial (7).log (5.1 KB)

that is one of the bugs that were supposed to be fixed in .78 (which was buggy,
so use .78.1)

David Lang

1 Like

Yeah. I tried that to start and it had some issues. Might try again.

1 Like

Good news. I was able to get v.78.1 working but I can’t seem to figure out the plunge issue.

It seems to find it’s location just slightly above the cutting area, plunge and then drop into the pocket. See photos above and video below.

The video stinks bc my had shakes but if ya watch carefully it will plunge and then move down when it reaches its depth.

The command was just to jog to the location and plunge a few mm.

Any ideas??