Ghosting in raster image scan direction

Hi, I am using a Sculpfun S10 to engrave plastic coated aluminium business cards with LaserWeb4. When using the raster feature I get strange ghosting (attached image) in the direction that the scan travels (horizontal in this image).

I ran a testcard (also attached) and picked the settings from that (ignore the bottom right… I scratched it with my nail to see if the residue came off easily). It did strike me as odd that high power/slow was less effective than low power/mid speed, but I went with it anyway as I know the test card is good (running the same gcode on other materials results as expected).

Settings are:
Laser Power Min 0
Laser Power Max 0
Laser diameter 0.08mm
Passes 1
Cut rate 3000
Trim pixels Y
Join Pixels Y
Burn White Y
Overscan 0

Can anyone point me in the right direction to solve this?

Thanks


This has been typical of a mechanical problem. If the head can wobble at all, this can happen. And I think others have sometimes thought that they didn’t have that problem, but then discovered that they really did. :grin: So I’d suggest looking carefully for any looseness.

3 Likes

you might get some hints as to what is going on if you zoom in as close as possible and look at the image formation.
Post picts.

2 Likes

I concur with @mcdanlj ; This looks like slop in the axis.

  • If it were a Gcode issue you would see it in the LW preview.
  • It’s not due to missed steps since the error does not ‘accumulate’ over the course of the raster.

The thing I would check first is the gear that drives the Belt for the affected axis:

  • Check that the Grub Screws holding it to the Motor shaft are secure.
  • This looks very like the effect you get if the grub screw is not properly gripping the flat on the stepper motor shaft!

Also check belt tension, and check the carriage wheels are correctly engaging the rails and cannot ‘rock’ as the direction changes.

3 Likes

Does it only do this in raster mode?

I cannot exactly tell from the pictures;
Which direction is the error, looks like x to me if that is the orientation of the sample in the machine.
If its x that is interesting because I think the x on this machine is a direct drive lead screw.

I would print two test pages; one with horizontal lines spaced apart and one with vertical lines spaced apart.

1 Like

No, it’s a belt. The SculptFun is just a nicely built and anodised example of a the classic Opensource 5040 (etc) machines. My ‘cheap and cheerful’ 5040 is basically the same, but the SparkFun is a more ‘polished’ kit, and better documented :slight_smile:

The rod + couplers you can see in pictures under the X axis is the Y axis drive rod carrying the single Y axis motor drive across the bed to the other side from the motor. Both axes are belt driven, two on the Y and one for X.

Which is a good point, there are three drive gears that need to be checked. And the coupler (to the right under the X beam) needs checking too.

1 Like

Engrave-able materials are like that, You just need enough power to vaporise the top layer, more power just starts heating and damaging the substrate, blurring the lines without any advantage.

1 Like

Hi all,

Thanks for the advice everyone. @mcdanlj helped me focus on looking into mechanical problems and it turns out it was a (hidden) loose screw on the X axis. It was a screw that secured the belt to the carriage. A tiny amount of slop in it was manifesting as backlash in the engraving. Taking the Laser module off, tightening the screw and replacing the module sorted it!

@easytarget thanks for the advice on engraveable materials, useful to know.

5 Likes

Surprised it does anything at all with these power settings.

:smile_cat: