Use laser to align XY axis

[Mr.Chips]

I have seen somewhere that builders have used a laser to insure that the X axis gantry is built to be square with the Y axis rails.
My Y axis (48" extrusion) rides on 1" linear rails, there are two trucks 7" apart (on each end) bolted to a rectangle of 1/2" 6061 Alum (truck plate). The X axis 72" extrusion is bolted to the truck plate by six pieces of 1/2" 6061 alum.
I'm positive either one end or the other truck assembly's is not square to the X axis (gantry).

I need to tear it down and refine the alignment to the X and Y axis, a laser comes to mind but have no idea as to how to do that, but am open to other methods as long as the results are a more precision machine.

Here are a couple of photos of my machine construction.

Similar Threads:

• x axis grd align failure!
• K40 laser will align but not cut or engrave
• Need Help!- laser mirror align problem, OR NOT
• Need Help!- Align beam in laser engraver
• Auto align B-axis (rotary axis) using probe

[routalot]

The laser alignment sounds interesting.How do you make sure the laser is perfectly in position and then how do you adjust the machine to get things right?I suppose it would be useful to know what level of accuracy you require.For most woodworking projects we don't need the level of precision that a chronometer maker would find essential.

I don't see a ballscrew on the other side of the machine,is there one?It might be that for a modest misalignment a tweak to the home position of each side might be a way ahead.If you had a 24" engineer's square and could seat it securely it might give an indication of how things are at present or you could lightly cut the largest rectangle the machine is capable of and compare the diagonals.For really accurate alignment of machine tools I believe they use laser interferometers and look for discrepancies of the order of one or two wavelengths and that really is a bit ambitious for a hobbyist.

[peteeng]

Hi Chips - A machinist level will pick up that discrepancy. A digital "box" angle protractor is pretty good as well. Peter

https://www.ebay.com/b/Machinist-Lev...81/bn_55190684

https://au.banggood.com/Digital-Leve...r_warehouse=CN

[Mr.Chips]

Yes there are two ball screws.
And yes I don’t know how they users lasers to do alignments, but asking how.
It’s not the vertical I’m trying to correct, so a level isn’t needed.
This is not dealing with cutting a square part. Or squaring at the hole position.

When I made the piece parts to each end of the X axis, the cradle things, they bolt solidly to the trucks without any adjustments. So there was an error in one or both of the X axis cradle/trucks they are not perpendicular to the Y axis rails. How do I know this?
If I loosen the mounting bolts to the right side Y axis extrusion to the base, with the rail bolted securely to the extrusion and drive the gantry to the back of the CNC and mark the side of the Y extrusion in the back and the front, then drive the gantry to the front the extrusion twists slightly sideways. This is because the Y axis and X axis are not perpendicular. When the Y axis extrusion is bolted securely it cannot shift and I see this in a change in the amps that are drawn as it meets more resistance in certain sections of the Y rail as it struggles to overcome the misalignment.
This is what I want to correct. If your using a machine with V bearings running on angle iron rails there is a lot of free play so you don’t notice a problem like this. But since I am using linear rails on all axis this is a very rigid and unforgiving system and it shows up.
Am currently redesigning the X axis cradle and Truck mount to be able to make a angular adjustment to correct the out of square condition between the trucks and X axis. .
The new version will have two layers, the bottom layer bolted to the trucks, and the top layer that holds the cradle they will have a pivot centered under the center of the X axis extrusion, and slotted hole in the top layer for adjustment. Will also change the mounting to the Y axis extrusion to accommodate a angle change. I should have made it this way in the first place not realizing that a degree or two over 72” results in the problem I’m experiencing.

[machinedude]

i think if you got a machinist square and indicate one side straight attached to the table and use the other edge of the machinist square to check the other axis this would give you a good idea of the machine squareness. the machinist squares are precision ground and should be pretty close. a 1-2-3- block or a 2-4-6 block used the same way would probably give good results. what you want to look for on the indicator is the amount of taper over a set distance. so if you are checking over say a 6 " length you are not going to want to see very much movement on the indicator. probably .001 or less. but this needs to be based on your actual needs so you could have more tapper if the accuracy is not to demanding. the longer the distance you are able to indicate over is going to give better results so keep this in mind as well.

[peteeng]

Hi Mr Chips -To square my machines I use the best, longest square I have. This is placed against the master rail. I use an aluminium extrusion to extend this across to the other rail. I then use a plumb bob hanging across the gantry rails and check across the gantry. I have used a steel rule at times as well. I repeat this from the secondary back to the master (at rail ends and in middle). Then I use the machine to mark out a 1m x1m square and check lengths and diagonals. Adjust as needed, patience is needed... Peter

[machinedude]

actually a Granite Tri Square is what you see on the high end machine tool for calibration but they are not very realistic in price for the home builds.a very small one seems to be around around several hundred dollars.

[JerryBurks]

I used grid printed vellum paper (like this: https://www.cutting-mats.net/clearprint-vellum-paper-6488.html) taped to my table. This is very precisely printed with a rectangular pattern.
Then I attached a cheap USB microscope camera to the z-plate, scanned the x-axis to be parallel and then the y-axis to check perpendicularity.
I can not adjust this mechanically on my machine but found out I was 0.0025" (y) per inch (x) out of perpendicular and could correct that with my machine controller software.

Edited to say: Whoa, just saw that the paper on my link is super expensive. I paid maybe 20 bucks for it but forgot where I bought it 8 years ago....
Anyway, just google for Fade-out vellum paper.

[Mr.Chips]

Sorry guess I’m not explaining the real underlying problem, it is out of square but not correctable by turning the left or right ball screw to correct the problem. I

Below is a photo looking down on the XY point and an exaggerated drawing showing that the physically the Y axis rail is not perpendicular to the X axis extrusion and rail.

Currently this is a bolted situation without any adjustment, I’m in the process of modifying the left and right XY connection making it adjustable to hopefully solv my self caused problem.

When I get this finished I will be able to make the X and Y to be perpendicular to each other and square, that’s when all the tips on “squaring” will come into play. Thanks for all your input.

[machinedude]

from what i can tell it looks like i thought as you explained before. so if you use some form of square and an indicator that will show you where you need to make the adjustments and what direction things need to be moved.so i'm guessing the adjustments you plan on making are on the gantry cross beam to bring one side over to where it needs to be so its square to the table axis.i would think even a simple framing square could get you in the ball park as long as the quality is there on the framing square. with the way things are made even those should get you to +/- .010 or so i would think.

[routalot]

So basically,opening the holes a touch and then carefully re-assembling?A bit of work,but compared to building a CNC machine from scratch its not that big a deal.Good luck.

[Mr.Chips]

Routalot, when I did the build about 11 years ago I didn’t plan for adjustments, and the way it’s bolted together I would have to enlarge some holes and guess on the amount to move then reassemble it all and if it wasn’t right do it all again.
By making the modifications I’m planning I can adjust it without taking things apart.
Got the two plates blanked out and a pair of gantry supports made to support the gantry to get it up off the rails and not have to disconnect a lot of wires and such while I work on the XY connections.

[machinedude]

i think your slots should be arcs rather than straight if you want a lot of movement. it looked like they were straight slots at a quick glance? thought i seen some flat head bolts with counter sinks on the original so those should be something else since they force the location.

i think with the straight slot your pivot will be limited to the slot clearance which might be enough to get things straight but you could need more? the arc would not be hard to figure out. the radius would be from the center line of the pivot to the center line of the slot. it would basically be a slot following a bolt circle diameter. that would let the pivot point get move travel for adjustment.

[Mr.Chips]

i think your slots should be arcs rather than straight if you want a lot of movement. it looked like they were straight slots at a quick glance? thought i seen some flat head bolts with counter sinks on the original so those should be something else since they force the location.

i think with the straight slot your pivot will be limited to the slot clearance which might be enough to get things straight but you could need more? the arc would not be hard to figure out. the radius would be from the center line of the pivot to the center line of the slot. it would basically be a slot following a bolt circle diameter. that would let the pivot point get move travel for adjustment.

They are Arcs, and centered off the pivot hole. Just hard to see at the angle.
Originally I had planned on one hole on each end of the truck but now I have decided on two bolts on each end.

[machinedude]

could not really tell from the picture but since they are you have it covered already. just wanted to point that out in case they were not. like you said you don't want to have to tear the machine down a bunch of times to get things fixed.

[Mr.Chips]

Just finished cutting the two tops to the XY axis cradle modification 9 minutes per piece at 35 IPM 0.010” per pass with 1/4” aluminum bit.
I made the arcs longer to be able to put two bolts at each end clamping securely so it holds the adjustments.
Now thinking if there is anything else I need to cut before I tear it down. More holes get drilled in the plates but that’s a job for the drill press.