Problem with CNC mounted laser

I can’t get my CNC mounted Jtech laser to engrave properly without making it painfully slow. I’m getting this odd result where it look as if it’s shifted. This is on the CNC axis not the rotary. I’ve tried overscan up to 8% and as low as 0 which has its own issues. The only thing that seems to fix it is dropping the acceleration on the CNC to 200mm/s, CNC is capable of 1500mm/s. Painfully slow at this acceleration rate.

My previois CNC did a decent job at 500mm/s acceleration and 8% overscan, I know this quite high but seemed to work well at that.

You sure you are using the correct units? 1500mm/s is 90,000 mm/m, which is stunningly fast for any frame design laser. I really doubt any CNC machine, with screws instead of belts, can approach those speeds.

Try setting your units to mm/m and see if your results make more sense. Engraving speeds up to 4000-6000mm/m is more typical. And much lower than that using a chuck rotary, and even more less using a roller rotary.

1500mm/s2 is the acceleration setting for the CNC. The max speed is 15000mm/min. Engraving settings are quite dependent on the acceleration settings, smaller stuff will never come close to top speed with low acceleration so power needs to be decreased.

My typical Lightburn settings with my old CNC using same laser and rotary were between 45-65% power depending on color of tumbler at 6000mm/min and 0.08 interval.

Yes, when you include units, the numbers make more sense.

Back to the doubled image…

  1. Assuming the double is along the Xaxis and it is rotating the Yaxis.
  2. You have backlash in the Xaxis.
  3. You have Bi-directional turned on.
  4. If you turn off bi-directional, it goes away.
  5. You have a loose coupling or pulley on the Xaxis drive.

If you have screw drives instead of belts, it has to be the motor to screw coupling.

Don’t just tighten the set screws. Rock the shaft back and forth to find the bottom of the flat on the shaft.

It really looks like the laser is actually turning off in the areas that are doubling on then points. If you look at the O’s there’s also a small section that isn’t getting engraved.

I don’t think this is a hardware issue on the CNC as I don’t see anything out of the ordinary when carving. It’s an Altmill so is ball screws and linear guides. I had this thought as well and went to an excessive amount of overscan (15%) which should rule out a coupler issue.

For some unknown reason there is no flat on the shafts. Kind of ridiculous.

A closer look reveals the M is painted in two positions, one coming and one going. See the image…

Turn off bi-directional so it only burns in one direction.

Ball screw machines come loose too. Mori Seiki, Bridgeport, Haas, Cincinnati Mills, etc.

Dude you’re my hero today!

Except now I question what the hell is wrong with my CNC..

I would love to take credit, but it came from the Forum. Users helping users.

Now you just have to figure where it came loose. :nerd_face:

So far I’ve found nothing. I’ve tested X movement with a dial indicator and it’s within 0.01mm for accuracy of movement and direction changes. Checked couplers, checked ball screw nut is tight in gantry block.

In the meantime I’ve applied a 0.2mm line scan offset and it works well in bidirectional.

The CNC only has maybe 50 hours on it. Only thing that comes to mind is ball nut needs to be greased even though the spec is 140 hours. But would have expected to find some error with the dial indicator.

There’s a difference between the gentle motions we all use when adjusting the machinery and the rather abrupt accelerations / decelerations applied during normal operation.

It’s entirely possible for the thing to work perfectly while you’re testing it and still have backlash when it’s running.

One interesting case involved a loose focus lens flopping around inside the laser head:

There are so many places for a little bit of slack to creep in … :grin:

Just to rule some obvious things out, this test pattern helps identify mechanical problems:

Backlash.lbrn2

Scale it uniformly to fill the platform and run it as fast as it will go in Line layer mode with Enable optimizations turned off and power set to mark a sheet of cardboard. Any differences from the design will be informative; a crisp photo will let us look over your shoulder.

1 Like

Just had a vision… As the trolley with the laser module attached whips back and forth, the gantry moves side to side slightly due to flex. You would never notice this with a spindle motor mounted on the trolley.

If you are thinking not my machine, I was looking at my CNC machine with the traveling table when the thought occured. The trolley is bolted to two vertical stanchions. You did mention 500mm/s, which is 30,000mm/m and considered blazingly fast for non-galvo lasers.

@MikeyH the 500mm/s mentioned is acceleration setting not engraving speed. Engraving speed was set at 6000mm/min.

The CNC manufacturer is saying this is a laser issue that it can’t keep up to the speed of the CNC.. this is laughable.. even at a 1khz I feel like the laser can easily fire fast enough for this speed.

In any event I applied a 0.2mm line scan offset and the engravings come out well. This is a backup for my galvo that had a source problem and had to be shipped back China. As long as it can do a few jobs if necessary I’m happy.

Thanks for your consideration and thoughts on the issue.

Cheers
Mike

Got it! Those are both conservative numbers. So it is traveling at a walk.

Even if it could not, it would leave a dotted line. In high speed mode, one of mine is rated at 30,000mm/m, so maybe they are thinking about a laser pointer. I hate responses that are so far off that I am not ashamed to use the word “stupid”.

That is what is really important. thanks for letting me know.

Pedantic: that’s 500 mm/s² or 500 mm/s^2.

Props for including units, but the “squared” part is vital! :grin:

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.