Still need some help finding out what's going on ...

@Manon_Joliton
Ok you may not be able to see the arching.

Try this:
Turn the lasers power to a low setting. Preferably so it just barely marks paper, run the same test job.
We are trying to see if the movement error is caused by the LPS.
Listen to see if the whistling noise is less or gone?
Does the movement error still happen?

During the video, i’m cutting paper so it’s only a 11.5% power
It’s whistling the same way at 40% and at 11%
Weird thing, at 11.5 the paper is cut, at 11.4, the paper isn’t even marked

Trying to figure a way to elliminate the LPS as the source of the problem.
Since this started have you had to increase power to get it to cut as you did before?

@donkjr yes, i had some power loss but I just had to check the alignment and then it get better, think it’s just about rhe alignment

@Manon_Joliton if the power setting is too low the tube will not ionize. Hence why it cuts out at 11.4%.

@Nedman ok so it’s normal if it seems not to work at all at 11.4 but works perfectly at 11.5 ?
Also, is it normal that when it cuts, it takes little time to reach it’s maximal cutting power ? Like on 3 cm it cuts at low power abd for the rest of the shape, it goes with a more powerful cutting

@Manon_Joliton Yes, there is a lower power threshold were the laser will not ionize. The exact number depends on the LPS and the Tube for each machine. If you are running the laser at the edge of ionization the laser output can fluctuate. So it’s not recommended to run that low.

Also, I looked at your machine parameters and the only thing I see is that, typically, the thread priority is set to “lower” with the “anti-disturb” checked off. I don’t think it will make a lot of difference, but those are the normally recommended settings.

Ok, i’ll change that tomorrow but I don’t think it would change anything ^^
What speed do you usually use ? Cause when I used it the first time it was set to 500mm/s and it seems very high to me but as it was working i didn’t change it

The max speed is fine at 500mm/s. That fast is only used for large engraving areas. Some people would say that the motors are only rated for 350mm/s but I haven’t had any problems personally running up to 550 mm/s. For cutting, the speeds are usually between 4-25mm/s. For engraving I typically run anywhere from 300-500 mm/s.

Ok, but cutting at 500 or 250 isn’t a bad thing for the machine right ?
'Cause I usually cut my veneer wood at 16%- 18% power and 250mm/s and if i make it slower it would burn the veneer

Typically the biggest issue at faster cutting speeds is that you get momentum bounces of the head as it changes directions which can mess up the cut line. It can also cause more wear and tear to the machine. I typically cut wood veneer at 15% power and 15 mm/s without problems. Of course my tube is old and is less powerful than new tubes. Your problem with burning is more related to the fact that you aren’t using an air assist to knock down flames.

@Manon_Joliton @Nedman

Warning: this will make your head hurt :(…

Sometimes it helps to look closely at the image to see if it is telling us something.
Don’t know if it will help but I slowed down the video and created a pictorial story board for the movement which is attached below.

I did not at first notice that the squares are drawn in two different rotations which makes the drawing error look different but it is actually not.

Observations:
…The Y stroke is offset in the -x direction only during a vertical stroke
…The -X direction error always occurs on a downward Y stroke
…On the upward Y stroke [after a -X error] the error is fixed after the Y stroke starts
…The error initially occurs after the first Y stroke.
…The error does not occur during a change in direction
…There is a pause when the error occurs
…The whistling noise only occurs on Y strokes.
…The amount and position of the error is the same in the CW and CCW direction of drawing.
…The amount and position of the error and correction is the same on both of the squares drawn irrespective of the position on the bed.
…The error is repeatable at any location on the bed with two different drawing objects.
…The error is only in the -X direction
…Something is smart enough to correct the error on the subsequent Y move.
…The error is repeatable and predictable.

Since the error is repeatable and then it gets corrected mid Y stroke I don’t think this particular problem is caused by:
…LPS NOISE
…Mechanical error
…Optical errors or alignment

These seem more likely:
…This error is in the source file
…There is a bug in the firmware
…The controller is not working properly (low probability)

Does this look right and make sense?

Its hard to imagine what is causing a Y move to pause and then move in the -X direction without it being told to do so.

@Nedman can we get @Manon_Joliton a known good source file?

@donkjr yeah this sums up well the situation, except sometimes (under 10% of the time) the problem disappear (for the exact same shape in the same series) I think it depends of if it starts with a vertical or horizontal movement or maybe if it starts at the middle of a side idk

@Manon_Joliton I think our next step is to somehow verify that the source file you are using does not contain these errors.

@donkjr ok, how shoul i proceed ?
About this, once I was making some marquetry with the laser for jewelry and in the design, the laser used to cut lines that i didn’t create on the corellaser file

We need to find someone with a stock K40 that can send you a corel test file that works. Then you can try it to see if its your machine or your source creation. Unfortunately I cannot since my machine is modified.
Another thing you can try is to use K40 Whisperer to drive your machine and see if a simple box works properly using it.
http://www.scorchworks.com/K40whisperer/k40whisperer.html#download

He can send me the file with the squares and I can take a look at it. If I don’t see anything I can convert it to jpg and send it back to see if it makes.difference. .

I tried to do a 30mm square using k40 whisperer but I still have the same ‘loops’ and my square remains as a rectangle :3

@Manon_Joliton
What do you mean by “loops”. Does the squares marking look exactly like the one your posted in the video?

Please send +Ned Hill your squares file and lets see what that does however I am going to guess it will mark properly.

Looking back over your posts you posted:
“Yesterday, i checked the tension of the left and right ‘roll’ (i don’t know how it’s called?), could you help me find out what is happening ?
I noticed the rotary pole in the front of the machine didn’t turn ‘round’ because of some screw that were missing on what i think is a motor ? (The little black cube that make the rotary pole rotate)
so i fixed the screws but it still doesn’t work…”

Can you post pictures of this shaft and its motor and coupler?

In the X dimension. Have you checked the belt tension and the shafts coupling to the x stepper motor? Similiar to what you did on the Y?

Here is an idea for some more tests while we are waiting for @Nedman results.

Horizontal draw and Y move tests:
Mark a horizontal line 4 inches long in the x direction.
Space down 2 inches in the Y direction and mark another 4 inch line
Space down 2 inches in the Y direction and mark another 4 inch line

Vertical draw and X move test.
Mark a 4 inch veritcal line at the left edge
Move 2 inches in the Y and mark antoher vertical line

Post videos of each pls.

Still looking for clues to what is happening :(!

@donkjr this is what wasn’t fixed well last week so I screw it back
This motor make the right and left belt to turn and before I fixed it, the shaft didn’t turn round but had a twisted movement
I’m gonna post the vids as i did last time and share you the link

Also, how should I sent the square file to Ned ?
Sorry, I saw your post th is morning, idk why i didn’t receive any notification about it