Alignment question - laser beam or x/y chassis adjustment...

[Craig_Johnson]

I printed a page limit grid with black dots on the 4 corners of my work area and one in the center.
For my system the x axis is left/right and y is front/back with the origin in the upper left.

I assumed I had a stepper motor setting that was off. The two settings were within 1% but not the same number.
After checking the x axis I came up with a better number, and the y axis seemed to work with it just fine.

The problem is the spotting light is on target for 0,0 and the back right dot but front right dot is missed. The spotting light is 2mm outside my working area.
The 'box' is not square with the angle at 0,0 being over 90 degrees and the back right under 90 degrees.
The spotting light is about 1mm right of the center dot (this is with the origin and back right dots in alignment)

Do I need to adjust my mirrors or could the motorized frame be out of square slightly? Which is more likely?

I adjusted the mirrors before, the light doesn't hit in the center of them and is closer to the edge than the center for all but the first one. It still works though.

Any suggestions?

[parsifaldruddle]

Hi Craig,

Sounds like you're pretty much 'there' and can probably tune the discrepancies out w/ offsets/axis settings. What's the size of your x-y stage? What sort of beam combiner are you using to add the spotter? (I noticed you said 'spotting light' and am assuming a low mW coloured laser added to the beam path with a combiner)

The displacements sound interesting because: The angles given - (0,0) > 90 and (x-max, 0) < 90 - would indicate a left skew at y-max. However, the spotting dot positions of ((0.5*x-max) + 1mm) at center (0.5*x-max, 0.5*y-max) and (x-max. + 2mm) at (x-max, y-max) indicate a right skew at y-max. Shouldn't have both.

I would use the laser at low power to mark out an actual square or rectangle and take measurements from that (larger is better). A nice fat circle in the middle might be handy as well. Your spotting dot may not be hitting the same point as the CO2 dot. You can then retrace the same figure with spotter only and get a better idea of the situation.

I would check square of the x-y stage - diagonal measurements should be close as possible. The more square your frame is mechanically, the smaller the correction factors required w/ your controller. Get your frame, and hence the elements of your 'y' axis, as square as possible. Okay to measure face to face distances along the y as a further check. Next check that your 'x' frame/beam to main frame spacing is as equal as possible at (0,0) and (x-max, 0). If it is off a bit, you might fix it by loosening the y-belt pulley on the shaft on the 'out' end, nudging the 'x' frame into position & holding/clamping it there (use of identical thickness spacer blocks is handy), then re-tightening the 'y' pulley on the shaft. Single stepper dual belt 'y' like their belt tensions reasonably balanced.

What diameter are your CO2 and spotter dots and how concentric are they? Procedure to set them I've seen most commonly is: Place CO2 markable target as far away as possible and burn a beam diameter dot as a target. Turn on spotting laser and adjust for center of target. Farther away is better - if they're centered well at 10m they should be cozy along the beam path.

You say your system is working so unless any of your optics mounts are heating up - mirror frames, head housing, air assist tip - mirror alignment is probably not an issue for you.

Anyways, good luck - you'll get it dialed in.

Pars

[Craig_Johnson]

Turns out the frame is square but the arm is off about 1.5mm, making it lower on the right side. Perhaps I'm calibrating from the wrong side - I should align the calibration sheet on the left instead of the top. I'll try that next time I work on it. Maybe if I align the beam again... but the spotting light is the problem, and the beam is right on the light!

Closeup of beam off by 2mm right

Calibration Grid showing beam off by 2mm right

There doesn't appear to be any way to adjust the arm

Rail looks sturdy

Close up of rail

[Craig_Johnson]

I aligned the grid this time using the dots on the left instead of along the top. The spotting light is then 2mm below the lower right mark. This would be consistent with the cutting I've seen with it. Next I adjusted the stepper motor for the y axis (I think they are switched) so that the dot aligned exactly on the right two dots and discovered that the lower left dot was 2mm short of the mark. So I recorded that number and adjusted the stepping motor to the left dots and got a lower number. By averaging the two numbers for the setting I get a better cut - 1mm short on the left and 1mm over on the right. The arm is clearly crooked, I've measured it with a square. The frame is OK though. Next step will be to adjust the tension on the belt to see if that will fix the problem, the adjustment screws are under the glass tube and will require disconnecting it. That will be a March project.

[Tech_Marco]

Craig:

Please update what you found as another folk seems having similar issue. I'm thinking that it is caused by the unbalance position of the X guider. Lower it down the X bar to the bottom then measure both side of Y see if any distance difference between those


Marco

[Craig_Johnson]

I think my laser tube is dying so I'll be replacing it after March 3rd. The only info I could find for adjusting the arm is to tighten the belt, and those controls are under the tube, I can't get to them without tearing the back apart. Brute force may come after that doesn't work. There isn't any flex in the sliding clamps on those tubes.

[Craig_Johnson]

The tube died and I took the time to replace it today. After removing the tube I see this:

There are 4 screws holding the plate in place, the plate must come out. Fortunately they have nuts with washers and locking washers on the back. Sure glad they didn't drill and tap the hole for easy access, I planed on killing the whole afternoon on this project, and shortcuts just won't work. After I get the plate out I see the belt adjustment for left and right:

Seems the only thing I can do is adjust the belts, so I loosen them both up all the way. There is about 5mm wiggle in the arm at this point. If I tighten the left side up and not the right its still off by 1mm (better than the 2mm its been off).
I put tension back on both belts. Can't hardly move it, so I loosen it up. Alignment still off. I grab each end of the arm with my hands and try to bend it into the right place. Then I hear a pop, or more like a thunk. I check the square against the arm:

It looks perfect. I tighten the belts to a little more that it was, but light enough that it still moves easily. Then I put it back together, and oil the rails. When I add the new tube, I try the Teflon knot suggestion I found in the owner's manual (FSL build):

Finally I do a calibration test:

Everything seems to be right on now. Maybe need to make some minor adjustments to the stepper settings, but its very close.
I suspect that the right side was off by 1 notch in the belt. This may have been due to the previous owner. He would send it out of range and you would hear this horrible sound of the notched belt slipping. Something I avoid.