Hello all, I’m posting on behalf of my work crew. We recently picked up a Monport 40W Pro with the integrated air assist and Lightburn license.
The initial setup wasn’t too terribly painful, but I’m stuck trying to figure out why we keep getting “wiggles” in the x-axis direction of text when engraving. I’ve cleaned and tensioned the mounts and rail for the x-axis as well as checked mirror alignment and also tensioned the y-axis belts. The wiggling appears to change with the speed as well - faster = longer waves, slower = smaller - which made me think mechanical but everything feels pretty smooth, especially along the x-axis.
Thanks for the response. The odd thing is that the vertical and diagonal lines are all straight and crisp, it’s only horizontal lines. We went through and aligned the mirrors as well.
We did have issues moving the tube to center the laser in the mirrors but that doesn’t seem to change this either. I did see that some people have issues with the mount being flimsy. Mine is no exception but the extra mounting holes were never tapped
So I just went through and checked the mirrors as well as performed another alignment. Everything appears as tight as can be and the laser centered. I’m still having the minors waviness on the horizontal lines shown in the original post.
Are there any settings within Lightburn that might affect the laser that could cause this? I did notice that the PWM frequency $33 can only be set to 1, even though it gives hz as the unit. I can’t change this to any other value.
What is the pitch of the waviness? Can you measure a difference in either pitch or amplitude at different speeds?
If there is a mismatch somewhere between GT2 and MXL belt/pulley I wonder if it might cause something like this. On a K40, it should have MXL belt and pulley; these are 0.08" / 2.032mm pitch, vs GT2 2mm pitch.
You can get wavy type lines on a scan pass that happens on a laser head direction change if there is any play in the system. But the waviness in that case is usually limited to the ends of the scan pass. I guess it could be longer over short scan distances where any induced oscillations don’t have time to dampen out.
Looks like the monoport pro has a linear bearing for the x-axis. Does the linear bearing seem pretty solid if you wiggle the end of laser head in the y-direction?
I think you’re mistaking a Y problem because you see it in the X direction, but the variations are in the Y direction.
I agree with @mcdanlj that it’s a mechanical issue and likely related to the Y axes.
Anything in the optics path or in the mechanical parts can cause this. Something as simple as a grub screw can do this. Speed changes and variances in the output are usually because somethings loose and you’re running different acceleration times… all exacerbate or change the results.
This is likely kHz, at a pwm of 1 it would be visible on your output with a machine like yours. The fastest you could change the output would be once a second, which would be pretty useless.
To the best of my knowledge, @cprezzi implemented $33 and it’s actually Hz in his implementation, which defaults to 5000 (5kHz)
I’m guessing it’s a placeholder on a board that doesn’t actually do PWM?
@NickTom you might share exactly what board they have put into this machine, and what firmware you see running on it in the configuration. Monport don’t help us out much here with details; mostly they just spam us mercilessly from time to time, wasting copious amounts of moderator time removing their spam.
What’s been mentioned so far has been checking for loose mirrors and mirror mounts, checking for a loose lens in the laser head, checking for loose mechanisms like X carriage and Y rails and loose belts.
@jkwilborn hit on one which hadn’t been mentioned yet and that is checking for loose drive gears on the stepper motors by seeing if any of the grub screws are loose.
As far as I know, the Monports I’ve worked with do produce a pwm signal. I have no idea about the firmware range or available settings as they don’t talk about these.
The last couple years of Monport they use L as a laser enable via a switch and wire the pwm to the IN terminal.
So after doing some more diagnosing today, it is in fact a mechanical artifact. The problem now, though, is that it’s due to the thin gauge sheet metal they use to mount the laser to both the linear rail and the x-axis belt.
It did a series of tests with different speeds at different belt tensions and it showed that the belt needed to be much more loose than I would expect it to be (as an avid 3D printer user) in order to minimize the waviness. Unfortunately it’s still evident at, would I would comfortably have as the loosest tension.
This picture doesn’t quite show it, but the mount to the linear rail wraps around the front and bottom of the gantry to attach to the belt. Because it’s thin gauge, you can easily push the whole head forward and back.
Maybe you can find a more light weight laser head unit so there’s not so much mass getting jerked around. Russ Sadler’s design comes to mind but don’t know if you can still buy it. He did post the files so you could make it if you are good with metal tools and sites like Send-cut-Send can cut the flat stock at good prices.
Although I know about some of Monports goofs and I’m not a fan of theirs, but I find it hard to believe they are shipping a new version or design of the head/rail/belt system that it would allow it to vibrate.
No one, I know of, would accept that from a new machine, no matter the cost, even cheap.
As a new machine from Monport, I’d be burning up the email with them to get them to fix it. If you’re having issues, then there are a lot of others having the same issue.
I’d also expect this on the Y axes only, as yours shows, but I’d expect a damping type of wave (ringing) produced.
Unfortunately Monport’s support requires pictures of the nameplate (not really sure why) but this K40 is in a secured area, so that can’t happen. I’m trying to come up with a creative way to get through to them, though.
As for the mount itself, I haven’t seen anything that could replace this version, but it also doesn’t seem like a “mainstream” platform. I modeled up a quick and dirty idea of what it looks like. the bottom 90-degree angle is where the belt attaches and pulls the head left/right. This is also where that vibration is coming from when the stepper is pulling it along.
Could you get a piece of aluminum angle(90deg) material and sandwich the existing area where the belt mounts? Better yet an aluminum “Z” bracket.
BTW, I’ve seen K40’s with a linear rail added and all but the ends of the original stock was left. ie the square rail guide was all there was and the carriage held the mounting bracket and laser head.
I would also look to see that the belt teeth mate very well with the geared rollers on both ends so that those vibrations are adding to the harmonics observed.
Russ Sadler addressed this by flipping the belt over so the smooth back side of the belt rolled on flat rollers. Just to removed the vibrations he saw in his cuts.
There are youtube videos of this too.