Hello,
I have Co2 laser, with large working area (2,5x1,25m) and I have strange problem.
I need it to cut straight lines but the result is not satisfactory.
I am attaching picture to see what our laser cuts are looking like.
It`s with ruida controller and my x/y axis is software calibrated.
So where could be the problem?
Thank you in advance!
Youâve exactly identified mechanical backlash, where the axis does not retrace the same path starting from different directions.
You have a very very large laser and are looking at fairly small errors. It may not be possible to achieve the accuracy you want from that machine.
Things that go wrong with desktop machines also apply to yours:
The small scale of the backlash error means you must examine and verify each of those items well below the level of a glance and saying âLooks good to me!â 
Set up a simple test pattern exhibiting the problem when burned into cheap material like a cardboard carton, then use it to check your progress at each step.
Start at the axis motors and work your way outward, checking every mechanical part for stability and proper operation. Ask how anything that touches anything that moves can contribute to the problem, then verify that it does not.
If you find nothing reduces the error after youâve checked everything, then you have probably found the limits of your hardware.
Do you have tabs enabled in the layer?
Hello,
I Just tested changes to the settings for the motors but no change in the result. I changed them drastically to see what would make an impact, but to no avail at the moment.
I feel like the motors are losing steps, but Iâm not sure and thatâs why I started there.
I think you have other issues. The lines appear to âlineâ up. When it loses steps it will never recover, if you have standard motors. This looks as it itâs part of the artworkâŚ
Is it possible to post the artwork? (.lbrn2 file)?
Did you check for the tabs being enabled�
If that were the case, you would surely have already seen it in more complex patterns.
Reducing the speed & acceleration by large factors will completely eliminate that as a problem, unless the axes have mechanical binding ⌠which you will discover and fix as you search for the backlash.
Try a circle test. If the start and ends of the circle donât meet then itâs surely backlash. Run it as slow as you like to remove the concern for losing steps as that wonât affect backlash.
So I am attaching pictures and the test file that i am using (.lbrn2)
And thank you guys for trying to help me! I appreciate that!
There is a line missmatch that I want to fix. It`s like 1mm or sometimes more.
The template lines are straight but the result is not as good as i want.
Oh. I think I see it on the horizontal line. I thought you were referring to the vertical line.
Was this done all in a single job?
Yes, its done all in a single job.
I tried to do in 2 jobs - the same results.
Iâd suggest dramatically reducing acceleration settings on the controller and reducing the Cut speed settings to see if the issue goes away. If it does, then you know itâs an issue of missing steps and can then adjust accordingly.
Interesting that nothing changes when I dramatically reducing acceleration settingsâŚ
What was the setting before and what did you set it to? Did you also reduce speed? 50 mm/s is quite fast for the types of movements here. And Iâd expect your gantry to be quite heavy.
I tried with many variations on acc settings.
I am cutting these templates on plywood 15mm, on 2-3mm/s and the missmatch is still thereâŚ
As long as nothing changes with the settings it leads me to think that it is a mechanical problem.
I donât want to experiment with the parts on the machine, but apparently I will have to.
By way of logic, if it is from the mechanical part, it will do it constantly and everywhere. I donât know how right I am, but couldnât the problem be the belts - if they are not tightened evenly or if they are looser than they should be?
A loose belt will prevent the machine from moving when the motor reverses direction. After the motor turns enough to retighten the belt by taking the slack out of it, the axis will start moving. The result will look like âlost stepsâ, but is really due to âlost motionâ.
Thatâs why belt tension is at the top of my list of things to check.
You are looking for a tiny amount of slack in a very long belt. If you can find the manufacturerâs recommendations for adjusting the belts, that will eliminate a lot of guessing as to how tight they should be; whatever advice you may find for a desktop machine will be entirely inappropriate for yours.
With the caveat that mechanical motion requires the same forces applied to produce the same result.
A slightly loose setscrew holding a pulley onto a motor shaft may stick in one position until a âfast enoughâ reversal shakes it loose and jams it in another position; until that happens, it will seem to work perfectly.
Given the low speeds you have tried, Iâd check the belt tension first. If adjusting the tension doesnât fix the problem or at least change the symptoms, then itâll be time to look for other problems.
Did this thing ever work correctly?
If so, how long and when did it failâŚ
I think we are missing some important information on the history of this machine.
