Hello guys!
I have a strange marks on engraved surface. As you can see there are overburnt blops/points that are out of the desired engraved surface, which destroys the whole look of the engraving.
Can you provide settings where the issue is visible?
That can help who are able to guide you.
$$
$0=10
$1=25
$2=0
$3=4
$4=0
$5=0
$6=0
$10=1
$11=0.010
$12=0.002
$13=0
$20=1
$21=0
$22=1
$23=3
$24=75.000
$25=3000.000
$26=250
$27=3.000
$30=1000
$31=0
$32=1
$100=80.174
$101=53.422
$102=1701.023
$110=24000.000
$111=12000.000
$112=500.000
$120=3000.000
$121=3000.000
$122=80.000
$130=603.000
$131=1203.000
$132=79.600
ok
Thats my settings, but it occurs all the time. Is it possible, that Arduino is too slow to react and it creates problem?
I don’t meant those settings but they are useful too.
I mean Lightburn layer settings to those who can help can check speed, passes and other variables.
I know arduino but not controlling laser machines so, i’m not sure if the arduino is too slow or not.
My sight is not that trained yet. 
Your $100 $101 $102 values are different. Usually $100 and $101 are equal. How did you get those? Check that.
Likely is your issue.
These are settings i used to engrave that numbers:
10 / 150
means
% of power / speed mm/s
This is set for 10% power and 150 mm/s speed:
So you suggest to decrease the max speed of X and Y gantry?
Speeds in GRBL are set to mm/min, I don’t think that
24000/60 = 400 mm/s
and
12000/60 = 200mm/s is fast for CO2 laser.
However, I will check that.
This settings:
$100
$101
$102
are used to determine how many steps stepper motor needs to do to move the gantry by 1 milimeter.
In many CNC machines X and Y gantry are on the same size pulleys and that’s why theese values are equal.
My machine uses GT2 T20 pulley on X axis and GT2 T30 pulleys on Y axis, so that values can’t be the same.
If $100 and $101 were the same on my machine and I tried to cut square piece, my outcome piece would be a rectangle.
This information was not available for us before, so it makes sens those values are different.
As this is for a CO2 machine and my knowledge is almost 0 about it, I leave any advice to those who can realy help you instead of guessing.
I,ve checked how it will engrave with only one-sided mode
and it looks like the laser is pre-firing or giving full power when it is enabled.
This is how it looks:
Maybe it is a problem with electronics and PWM signal isn’t stable?
ps. I’ve also tried engraving it perpendicular to the plywood knots, on the photo is parallel engraving.
Can you post the GCode of your project for us to check?
Here it is:
leafDefect.gc (70,7 KB)
Maybe unshielded steering cables are the issue…
ps. I’ll soon try to change the arduino and check if it would work and add shielded microphone cables to isolate signal from EMI.
I changed Arduino and it didn’t work. Still the same issue.
Shielded cables also changed nothing.
Did you try 1.1h?
Your CO2 issue seems related to LPS but try first the latest GRBL 1.1h. Easy upgrade with LaserGRBL. Copy your GRBL settings first.
I’ll try and send feedback. I also suspect that it is LPS fault. I’ve heard that newer versions of Windows or faulty USB port can somehow disturb the signal.
Check, firmare change to 1.1h doesn’t change anything. My Arduino is connected to 3.0USB port, I switched it to 2.0USB, checked and changed transfer mode to synchronised, also checked, nothing changed
I bought wave analyser, but it hasn’t been delivered yet. I’ll check the PWM signal from Arduino to LPS to ensure, that it is not Arduino fault.
If nothing is visible, it will either be the fault of the tube’s power supply, the tube itself (is there a chance it could be the fault of the 100W power supply and the 80W tube?) or the operating system/computer, I have to dig out the old laptop and check😁
I found an interesting relationship…
USB 2.0 port, ver 1.1h.
In the buffered mode, defects are separated by approximately 0.5 mm and most often appear in pairs, although there are also series of 0.5 mm intervals.
So we have 2 dots with a distance of 0.5 mm, then, for example, a distance of 1 mm or 0.8 mm or another distance and then again there is a pair, or very rarely a single dot.
Without wave analyser it’s like reading the tea leaves…
Without a doubt, a problem that needs to be analyzed in detail.
What counts is your sense of humor, otherwise you would have already solved the problem with a liter of gasoline. 
Or a gallon if you live in the states.
It appears as if it’s sticking at full power for a short time. It’s rather unlikely that the controller is doing this, but it’s definitely a hardware failure of some kind.
I don’t know how you would set up a scope to even check this… what I mean is what would you trigger on?
I’d guess the laser module, rather than the control board… IMHO…
It would help if you could figure out if it’s the leading or training edge…
