Random cut quality

[Txchevy]

Well after almost 6 months I've finally worked out all the kinks on my diy build. I made my first cut and everything was working pretty well not perfect but not horrible either. Then I cut a simple 3x3 square with 1.5 circle to begin tuning and everything goes crazy. It cut pretty well the first time and then got better after I slowed down the internal circle. Then the 3rd 4th and 5th cut everything goes to crap. The inside circle is crisp and smooth but the outside square has two sides that have weird squiggles in then and 2 straight lines. I can't for the life of me figure out why. The gantry is square that thc appears to be working. Then the 7th cut it works great and then the 8th it goes back to squiggles. They are always in the same place and the same pattern. Any ideas?

Table specs: percision plasma 5x10 gantry, candcnc bladerunner kit with 620 oz in. steppers, power max 45 with hand torch for now.

Cut specs:
1st cut:
16ga
350 imp
pierce height: .15
cut height .06
no separate layers

2nd 3rd 4th
16a
350imp outside
120imp inside
ph .15
ch .06

5th 6th
16a
325imp outside
120imp inside
ph .15
ch .06

7th 8th
16a
325imp outside
120imp inside
ph .15
ch .09

sq hole below a messed up piece


messed up part


It almost seems like the thc isn't reacting quick enough and draging on the plate, or maybe I need to reference the z after each pierce.

Any suggestions ya'll?

[I Lean]

Does it do the same thing with the THC off during the cut? Easy way to tell if your THC is causing anything...

I sometimes see similar squiggles on two opposing sides of squares, if I have a high cut speed. I attribute it to flex in the gantry...when it changes direction, the torch vibrates a little bit, which translates to wiggles in the cut. It goes away if I slow down the cut.

[Dustin407]

350 ipm.... Thats good for well built machines but for most of us thats not gonna happen.... Slow Down... Try your same cut file with THC off and at 180 ipm.... guaranteed greatness.... The snake lines are from movement in the z head when you change directions in shakes and shows up on the next cut path... I can tell you cut counter clockwise b/c the snake lines are after it makes a left turn and then fades out straight over the 3" distance. Did you notice your 1.5" circles look pretty good because its a smooth cut and not changing directions fast so the z head didnt move.... I try not to go over 200 IPM on any part, I threw my cut spec book away along time ago...lol

[Txchevy]

Does it do the same thing with the THC off during the cut? Easy way to tell if your THC is causing anything...

I sometimes see similar squiggles on two opposing sides of squares, if I have a high cut speed. I attribute it to flex in the gantry...when it changes direction, the torch vibrates a little bit, which translates to wiggles in the cut. It goes away if I slow down the cut.

Good call! I remembered that the z axis crashed into a box i had on the table when it took off looking for a home switch on the x axis. The problems all started after that. I tore down my z axis and found that the linear rail bearing block has 3 bearings that ride against a steel rail on each side. A 3rd bearing is in the middle and can be moved left or right to take up the backlash via set screws and then locked in place. I also noticed one of the fixed bearings was .015 out of plane. This allowed .015 backlash on the x axis and .023 on the y axis. more than enough to cause problems with the 6 inches of travel on the z axis.
I tightened every thing back up and made a few test cuts. Looks like everything is back to where it should be. I'll post pics in the morning when I tear it back down for paint.

Thanks for the help.

[Txchevy]

350 ipm.... Thats good for well built machines but for most of us thats not gonna happen

Hmm... Now that I've got the z axis tightened back I have had no problems running at 350 imp at 45amps. I guess I did something right for once