Tips for squaring your laser?

[Clark Pace]

So my laser is pretty good, but could be better. I have 2 lasers both china based. On older laser (2013) is a little out of square. Do any of you have good tips for getting is as close as possible?

[Gary Hair]

First you need to align the axes to the rulers, then check for 90 between the two. The exact adjustment for your machine will determine how to go about this. For my Trotec there are several screws that hold down the rulers through slots. To align it I can move them out of the way and cut a piece of thin acrylic along the X and Y, then move the rulers into position against the acrylic. This will get the rulers aligned with the beam. Next you need to check if they are 90 to each other. That adjustment will be different on most machines so you'll have to figure that part out yourself.

[Rich Harman]

When you think you have it squared, vector the largest rectangle you can fit on your machine and measure the diagonals, corner to corner. If they are not exactly equal then it is not square. The rectangle needs to be as big as possible so that the error is apparent, the larger the rectangle the larger the difference will be in the diagonal measurements for a given amount of "out of squareness".

I was able to adjust mine by loosening one side of the torque bar where they join in the middle (the Y axis is driven by a motor in the middle of the machine) so that one side of the Y axis could be adjusted independently. This works well because it is a clamp and not a grub screw; it can be very finely adjusted.

[Kev Williams]

my way: Pretty much what everyone else says above, with one addition:

Assuming the Y axis supports are parallel, and they should be else the Y gantry would bind up somewhere between top and bottom-- Run the beam-to-rulers test on the Y axis FIRST, since the Y rails should already be parallel.

Once the Y scale is positioned correctly, do like the others say :)

[Jerome Stanek]

Remember that the difference in the measurement is to be cut in half to get square. So if you measure 1/8 out you have to move just 1/16 to get square.

[Rich Harman]

Remember that the difference in the measurement is to be cut in half to get square. So if you measure 1/8 out you have to move just 1/16 to get square.

It does not work that way. For example, if it was one degree out of square, the difference in diagonal measurements would vary greatly depending upon the size and the proportions of the rectangle. A four inch square may show a 0.050" difference, while a twenty inch square would show a 0.25" difference. You cannot determine how much to move based upon the diagonal measurements unless you also factor in the size and aspect ratio of the rectangle - and also the distance between the Y axis rails (which will be larger than the largest rectangle you can vector).

What you need to do is mark the current location somehow, make a small adjustment then run the test again. It is an iterative process.

[Jerome Stanek]

It does not work that way. For example, if it was one degree out of square, the difference in diagonal measurements would vary greatly depending upon the size and the proportions of the rectangle. A four inch square may show a 0.050" difference, while a twenty inch square would show a 0.25" difference. You cannot determine how much to move based upon the diagonal measurements unless you also factor in the size and aspect ratio of the rectangle - and also the distance between the Y axis rails (which will be larger than the largest rectangle you can vector).

What you need to do is mark the current location somehow, make a small adjustment then run the test again. It is an iterative process.

Figure it out if you move it by one half it will be square. I have been doing this with squares from 1 inch to houses. the diagonal will always be the same

[Rich Harman]

Figure it out if you move it by one half it will be square. I have been doing this with squares from 1 inch to houses. the diagonal will always be the same

Sure, when framing in carpentry you can physically move the frame so that you remove one half the difference by simply skewing. That same corrective movement does not correlate to the orthogonal adjustment that needs to be made to one side of a Y axis on a laser.

For example, if you are 1/2" off in a diagonal measurement when squaring up a window, that does not mean you simply shift one corner to the left or right by 1/4". The math does not work that way.