What's new
What's new

Laser cutting badly in one direction?

Winterfalke

Stainless
Joined
Mar 26, 2011
Location
Huron
So I have this POS chinesium CNC laser gantry unit with an IPG YLS2000 and a Raytools head, and I'm running out of ideas. It is consistently slagging the cut in 10ga hot roll steel, but only in the upper left quadrant when cutting a circle. For the most part it cuts fairly clean, but every so often it just starts slagging with no apparent change in parameters or setup, and nothing I do will change it. It always generates slag in the upper left quadrant, cutting either direction, and sends a shower of sparks across the top of the part toward the 7-8 o'clock direction. In the other three quadrants it cuts clean with no sparks on top of the sheet. I recently dissassembled the entire laser head and went through it, and the laser cut considerably better for a while, but the issue is back with a vengence. Even when cutting well however there was always a hint of sparks and slag in that one direction. I have checked the perpendicularity of the head to the table, the levelness of the gantry itself, the laser concentricity to the nozzle (as close as I can make it with a microscope to check with), and the shape of the laser hole (cut half way through a piece of packing tape, fairly cylindrical). I am leaning towards either the gas jet is blowing to one side for some reason, the beam is not exiting perpendicular to the laser head, or the capacitance follow is varying based on direction (maybe a signal interference issue?) Is this an issue anyone has seen before?
 
Post a picture of the machine. What is the Beam Delivery on the Machine? Fiber Cable or Mirrors? A picture is worth a million words.
 
MTB is Dade Heavy Industries, doubt you have ever seen one. I doubt they had ever made a laser machine of this type before. It certainly doesn't seem that way from the general lack of usable common sense features, like having the controller know which table is in the machine or not putting gussets completely blocking the dust collection passages. Everything we have is from these same idiots, and it's all garbage. The laser unit itself is an IPG YLS2000, a 2000w fiber laser running to a Raytools focus head.
 
MTB is Dade Heavy Industries, doubt you have ever seen one.

Thus why a picture would be worth 1000 words.

Nozzle Centering??? But you say it is in 1 quadrant of the machine???

Is there something in that area of the machine causing the "machine" to go out of square with the head if the head is squared in another location?
 
Thus why a picture would be worth 1000 words.

Yea, it won't let me upload pictures from work, says something about not having permission. The nozzle is as centered as I can make it with a 35x microscope to inspect the tape. I don't think it's anything to do with machine level as it does this anywhere on the table, either table, cutting in both directions. It only seems to be as it is traveling around a circle in that upper left quadrant. The funny thing is it is cutting with almost no slag today, with no changes. The sparks are still there in the one location, but barely noticable. It's just so inconsistent that I can't ever troubleshoot it day to day.

The focus assembly looks a bit like this one BT24S SERIES 4KW Laser Cutting Head - RayTools AG, except it's labeled as a BMH110SA1AP which I couldn't find.
 
Don't know if you can do it with a fibre, but with co2 its kinda a common maintenance practice to take a unfocused beam shot into a lump of acrylic, the pattern it burns is very specific and it lets you check for a evenness in the circle of the beam and its energy the way few other tests will.

Being only one quadrant, its got all the hall marks of a nozzle centring issue or a beam issue (one of the fibres dead???) Do you know how to do a shot with a bit of tape on the laser nozzle, its old school but oh so effective for undefeatable proof of beam centring!

Am making the assumption the nozzle has been changed to a new one and is unquestionably not at fault??

The fact that it went away on a full rebuild then returns kinda points to something gradually moveing - failing, don't know enough about fibre beam paths to make a valid comment other than knowing the beams carried on multiple fibres and making the assumption that if thoes fibres were failing on that quadrant would kinda give you a gradually worsening issue? Then again if they were failing can't see how the maintenance would have fixed - helped it.
 
Depending on your table you can do a mode burn with all of the optics attached.

It sounds like a centering issue to me, and I'm not even sure how you're checking that with a microscope. On our CO2 lasers we adjust it on the fly by adjusting the centering towards the direction of the bad cut.

Do you have any acrylic you can pierce with nitrogen? It will make a little cloud around the pierce that should be exactly even around it. If it's not even adjust the centering.

I'm sure it's different with fiber vs CO2, but I don't know if it is different enough for the heads to have different problems.



Is the head perfectly symmetrical? On our old CO2s we can rotate the head any way we want, which is nice seeing where the issue is coming from. If the problem rotates in the head, it's usually a lens alignment or nozzle issue.

And an easier thing to check: is your nozzle round? If the machine has been crashed the nozzle can deform, giving a bad cut in one direction.
 
I'm not even sure how you're checking that with a microscope

It's not complicated, I put the smallest nozzle we have on the laser (0.8mm), stick a piece of packing tape to it, do a low power pulse, then stick the tape over the hole in a 1-2-3 block and examine it under a microscope. That way I can measure the actual concentricity between the hole and the nozzle imprint.

I don't have another nozzle that size yet, still on order, but I have tried the next size up and it pretty much does the same thing but worse. It looks clean under the microscope, but I don't think that means much to a supersonic gas stream. And I have finally verified that it IS NOT only that one quadrant, if I cut a circle in the other direction it sometimes does the same thing on the other side, but not as noticable. So it is still very directional, and not tied to any location on either table, but it is more so on the one side. So yea, I'm agreeing that it looks like a beam shape, density or perpindicularity issue, but I don't know how to test for it. Am contacting IPG to ask for information on doing a mode burn, I think that is one of the diagnostic procedures I've been looking for (It might have helped if the idiot machine installer gave us any real training, or even spoke english.)

And thank you for the suggestions and comments, you guys are much more helpful (and I'm fairly sure much more knowledgeable) than the MTB.
 
You want to do the tape on the nozzel your cut with, your not looking at the hole in the tape, your looking at were the hole is in relation to the imprint left of the nozzel hole in the tape, You want them as near ceter as you can see under say 2-3x magnification.

Its not about the gas stream at this point, its about haveing the nozzel smack bang middle of the laser beam. Like that when you cut the center point of the gas stream is right over the center point of the laser beam.

You still need to do the acrylic block shots , that will show you the true shape of your beam - how its energy density is. should be a certain shape in cross section but above all it should be even all the way around. if not you have problems these are bad, can't find a good example

mode_burn_misaligned_laser.185211122_std.jpg
 
Finally got pictures to work...
IMG_0530.jpg
IMG_0231.jpg
Here's a test burn halfway through a piece of packing tape, looks pretty cylindrical. Hard to get the full depth of focus putting an iphone up to the microscope eyepiece.
IMG_0491.jpg
 
Yea, I get the part about having the beam centered on the laser, I was also trying to do double duty to get some idea of the beam shape and density. I've been trying to keep it as centered as possible, it's super touchy, I don't think the optics are entirely square or concentric. And wow those mode burns are sooo much more useful lol. Definitely need to do that. So I imagine a good one would be a cylinder with even rounded corners at the bottom?
 
No a good one for a co2 is more like a donut has been pulled out, you don't need energy in the middle of the beam, you want it more around the edge but above all you want it even all around, other wise it effects things as you cut in different directions. Ripped off google, but B is what your seeking. Though ideally with less peripheral damage,

Laser-intensity-profiles-of-the-laser-beam-burned-in-a-Perspex-block-A-unfocused.png
 
Once centered though it should stay that way unless your changing nozzles. If its not you need to work out the why. Not been around a fibre enough, but if there like CO2's the output lens is some what a consumable and gradually distorts do to atmosphere, contamination, heating and such. Left too long that will effect cut quality.

Lasers are finiky things, may pay you to find a good laser tech and pay them to come in and go over the basics with you about the beam output end. Yep they may be no expert machine wise, but the knowledge you can gain in a hour or 2 will help you a lot for a long time to come.

FYI i have never seen a part shot on packaging tape, only holes. how many watts does it take to do just that?
 
FYI i have never seen a part shot on packaging tape, only holes. how many watts does it take to do just that?

I think I was at 8% power and as quick as I could pulse the button on a 2k laser, but with the chinese software I'm not even sure if that is accurate. And yea, I've been pushing to get a real expert in here for weeks. I've built few lasers in college, but it's not really comparable experience lol.
 
Yeah im not knocking you on the expert bit, i see it all the time in the print world too, companies spend big money on equipment then nothing on the basic staff training. Then they run it so in efficently for years losing so much extra what could profit or quality improvements its unreal.

With competent people, a couple of hours with some one that truly knows the process - tricks generally pays off big time. Normally its the same kinda employers that insist on highly qualified people that skip the training the worst too.
 
Winterfalke,

I suspect something is loose in the optics allowing your nozzle centering to go off in a quadrant. A loose optic may shift position due to the X-Y acceleration. This will off-set the beam centering location with respect to the nozzle center then fall back into place when the acceleration changes direction or drops to “0”.

Check the fiber cable connection to the Raytools cutting head. The fiber cable bayonet lock ring should be tight. Lean on the fiber bayonet in all directions. There should be no movement where it inserts into the Raytools head.

The next thing to check is the Raytools head. Leave the fiber cable connected but remove the head from the Z-axis . Give it a good shake next to your ear when the room is completely quiet. Nothing should rattle.

Why the ziplock bags?

Adama,

The acrylic plastic trick will not work with a fiber laser. The beam is a much shorter wavelength than a CO2 laser. The plastic is a good window for a fiber laser.
 








 
Back
Top