Not cutting square

[Marc Nicoloudis]

Good day Everyone!

Having only cut smaller more intricate designs with my laser, I was unaware that it is currently cutting/lasing out of square. Measuring corner to corner over the entire size of the table(26x20) it is 1/8" out of square. Laser is an Pinnacle Accuris model 30watt.

I have moved the machine a couple of times, but have always been careful to secure not only the laser engraver, but the gantry inside as well. All of the linear slides have the original silicone "Unbroken" marking points on the screws, telling me that they have not been previously adjusted.

So my main question is, could this be any type of a software issue, or is this wholeheartedly a nuts and bolts problem? Any ideas before I start adjusting manually? Thanks for your help! Marc

[Kev Williams]

nuts & bolts. Misaligned mirrors can cause it. Misaligned gantry setup can do it.

to quick-check the gantry (my way, anyway ;) ) First, it can be assumed (sorta) that the Y rails are reasonably parallel to each other, or the X rail would bind up somewhere in it's travel.

That said, get a carpenter's square, move the X rail down to where you can look straight down at one of its straight edges. Put the square along the right Y axis rail, and move it to where you can look straight down to see if the X rail is straight with the square. I check with the square both ways (short and long sides inverted). Move the lens head out of the way and repeat on the left side.

If it's out, adjust-- if it's good and straight, I'd start checking mirrors...

[Marc Nicoloudis]

Kev,
Thanks buddy. I loosened up each screw that aligns y rails and shifted them over just a bit. Corner to corner is now equal(based on a tape measure). Willing to bet that it is still off a bit. I will read up on adjusting mirrors. I was under the impression that mirror alignment was more for aligning a precise beam focus, rather than beam placement on the material? Thanks, Marc

[Kev Williams]

If the beam is angled, the point where it hits the mirrors will continually change. However, I've found that our lenses are pretty forgiving. They will focus the beam pretty close to the same place regardless of where the incident beam hits it. By pretty close, I'm talking .010-.020". But that's MY machine. My mirrors are totally out of whack, the poor beam is all over the place on the lens mirror. But most of my laser jobs aren't all that critical, and those that are, I compensate for.

My point is, if you're off 1/8", I doubt the mirrors could even miss that far and still work...

[Marc Nicoloudis]

If the beam is angled, the point where it hits the mirrors will continually change. However, I've found that our lenses are pretty forgiving. They will focus the beam pretty close to the same place regardless of where the incident beam hits it. By pretty close, I'm talking .010-.020". But that's MY machine. My mirrors are totally out of whack, the poor beam is all over the place on the lens mirror. But most of my laser jobs aren't all that critical, and those that are, I compensate for.

My point is, if you're off 1/8", I doubt the mirrors could even miss that far and still work...

Gotcha.
Ok, the other question I had, is whether I can adjust the steps of the stepper motors through the software like I can with my CNC router? Doing some tests today, when cutting a 5" square, it's coming out at 4.984". Now I can adjust for this in Corel, but might be nice to set it once instead of every time I draw. Thanks for your help, Marc

[Kev Williams]

Not sure that's possible in the software, although it might be possible to adjust the X-Y controllers?

[Brian R Cain]

I can't speak for the Pinnacle but with ULS machines there is a dowel pin on each of the Y-axis rails. When you pull the X rail up to them it should touch both dowels. If it doesn't, slacken off the screws on the Y-axis motor coupling so you can pull the X-rail into place and re-tighten the coupler. I would imagine that something similar exists on your machine.

[Michael Reilly]

Are you measuring something you've cut or the table? I know on our Epilog, the rulers were not in alignment with the gantry. We used the laser pointer to adjust one ruler to be parallel to the gantry, and then used a framing square to adjust the other ruler. There was also the potential the gantry wasn't square. We were instructed to move it to the back and there was a point that should touch the back wall. If it wasn't even on both sides, we could loosen the belt and skip teeth on it until it was. On our Trotec, we could cut a 1" square and find that the width was not the same as the height. Trotec could never explain it, but when buying our Vytek lasers, they explained that the laser is actually an oval shape and if it's not rotated just so, then the kerf will be slightly different when traveling in the X vs. the Y axis. While some of that can be overcome in good design of the machine, some things are inherent in the physics of it. As long as the amount that it's off is the same, it can be compensated for in the design. Though we don't cut industrial parts, we have tried to get things as accurate as possible and with each new machine, we get closer and closer.

[Marc Nicoloudis]

I can't speak for the Pinnacle but with ULS machines there is a dowel pin on each of the Y-axis rails. When you pull the X rail up to them it should touch both dowels. If it doesn't, slacken off the screws on the Y-axis motor coupling so you can pull the X-rail into place and re-tighten the coupler. I would imagine that something similar exists on your machine.

Brian,

No dowels that i can see. Because of the way it was askew, I was able to loosen two of the screws that hold the the y rails to the cabinet, and shift both of the rails a tad left, and then re tighten. Am getting closer to square the more I work at it.

[Marc Nicoloudis]

Michael,
I am cutting the largest square possible on my machine(26x20), then measuring corner to corner. The laser pointer idea is good, and i will try it out, but I have trouble trusting my eyes for alignment.

I wonder how cutting a 1" square would give you much inclination as to the squareness. Doesn't seem as though there is enough travel to find anything but the most obvious of misalignment. Seems like I have been doing more maintenance on my machines than cutting lately!

[Richard Rumancik]

. . . Ok, the other question I had, is whether I can adjust the steps of the stepper motors through the software like I can with my CNC router? Doing some tests today, when cutting a 5" square, it's coming out at 4.984". Now I can adjust for this in Corel, but might be nice to set it once instead of every time I draw. . .

Marc, check if there is any provision in the driver to fix the error. I have a similar problem with my GCC Mercury in that it cuts smaller in the x direction than it should (0.2% short). There is a driver option for the Mercury to tweak the scale, but unfortunately for me the resolution of the correction is too coarse to be helpful. I have to tweak in Corel which is quite honestly a pain as it means keeping a master file (to scale) and a "compensated" file which will cut to the correct size. I would not recommend this work-around if there is another way.

I looked at an Accuris manual and I see something called "shape adjustment" in the driver. It allows setting a "scale" value in the x and y axes. But it does not explain much, nor does it indicate if this is for raster or vector correction (or both). It also allows a password protection for this function. Maybe check with your dealer or contact the factory to get an explanation and see if this is a possible solution.

I think that a better place to allow the correction would have been in firmware (not the driver) but GCC did not provide for that option. I'm not sure if any other manufacturers do. Conceptually it is such an easy thing to implement but you have to remember that these engravers were not really designed to make accurate parts. If you are making a sign or a nametag, who cares about a 0.3% dimensional error? But for those of us who are trying to manufacture parts, it is a big deal.

[Marc Nicoloudis]

Marc, check if there is any provision in the driver to fix the error. I have a similar problem with my GCC Mercury in that it cuts smaller in the x direction than it should (0.2% short). There is a driver option for the Mercury to tweak the scale, but unfortunately for me the resolution of the correction is too coarse to be helpful. I have to tweak in Corel which is quite honestly a pain as it means keeping a master file (to scale) and a "compensated" file which will cut to the correct size. I would not recommend this work-around if there is another way.

I looked at an Accuris manual and I see something called "shape adjustment" in the driver. It allows setting a "scale" value in the x and y axes. But it does not explain much, nor does it indicate if this is for raster or vector correction (or both). It also allows a password protection for this function. Maybe check with your dealer or contact the factory to get an explanation and see if this is a possible solution.

I think that a better place to allow the correction would have been in firmware (not the driver) but GCC did not provide for that option. I'm not sure if any other manufacturers do. Conceptually it is such an easy thing to implement but you have to remember that these engravers were not really designed to make accurate parts. If you are making a sign or a nametag, who cares about a 0.3% dimensional error? But for those of us who are trying to manufacture parts, it is a big deal.

Richard,
Yes thank you. Exactly what I was looking for. I pulled it up in the driver area under advanced, but I cannot enter any information. I will do some checking, might need to find a password listed somewhere. Contacting the distributor is almost always of no help. Will keep in touch on this, Thanks, Marc

[Richard Rumancik]

Marc, you might need to get the password from the manufacturer - for some reason they felt that users shouldn't normally touch this. No idea why. Probably any technician could get you the default password.

One other point - to determine a scale factor I suggest measuring the distance between scribed lines, not the edges of a cut shape. This eliminates kerf variability. I use a 24" machinists scale but a 12" one might do. Longer is better for accuracy. I scribe light lines say 23" apart in Corel and then measure the actual distance under magnification (use a loupe to measure accurately).