20x20 square

[jburnett83]

We have been cutting a 20x20 inch square off our SS2
Left top side to bottom right is 28 1/4
Right top side to bottom left is 28 5/16
What is the best way to get this back into square? We have tried doing the Axis Calibration.

[FabLab Wageningen]

Could you put your .CRV online?
8 inches more than the 20 planned / designed seems too much for me to be a pure machine issue.

Marcel.

[TReischl]

Could you put your .CRV online?
8 inches more than the 20 planned / designed seems too much for me to be a pure machine issue.

Marcel.

Reread the post. He is talking out of square measurements, not length and width measurements.

[TReischl]

There are a few ways to square a machine. You are not "calibrating" the machine. Calibration is determining how far the machine moves not how square it is.

The way I do it is to machine an edge along the axis that is fixed to the machine bed. Probably the Y axis on your machine if it is a gantry. Then place a sheet of thin mdf, like 1/4 against that newly machined edge that is secured to your spoil board. Cut a slot along the X axis. Flip the board over, place the same edge against the reference surface. Machine another slot. Measure the distance at both ends between the slots. That will be double the error. Adjust your machine accordingly. Have no idea how the Laguna sets the axis, some machines use two switches, others use hard stops. You adjust the axis that is fixed to the machine base.

The longer you make those slots, the more accurate you will be when you adjust the machine.

This technique is a riff on checking a draftsman's square.

[FabLab Wageningen]

Could you put your .CRV online?
8 inches more than the 20 planned / designed seems too much for me to be a pure machine issue.

Marcel.

Reread the post. He is talking out of square measurements, not length and width measurements.

eh, Ted,

If you plan 20 by 20 and get 28something by 28something, then it seems not just a small 'square' issue to me. A good 40% more than you plan seems veryveryvery odd to me.

Marcel.

[rig gap]

28 plus (28.288") is the distance across the diagonal corners for a 20" square.

[TReischl]

Could you put your .CRV online?
8 inches more than the 20 planned / designed seems too much for me to be a pure machine issue.

Marcel.

Reread the post. He is talking out of square measurements, not length and width measurements.

eh, Ted,

If you plan 20 by 20 and get 28something by 28something, then it seems not just a small 'square' issue to me. A good 40% more than you plan seems veryveryvery odd to me.

Marcel.

Need to read more carefully Marcel. He wrote:

Left top side to bottom right is 28 1/4
Right top side to bottom left is 28 5/16


He is checking for out of square by measuring across corners. I have also included a sketch of how a draftsman checks squareness of a triangle. This same method can be easily adapted to checking squareness of machine movement. It is extremely accurate. On my machine I cut those slots about 4 foot long. then I measure between them. If I get it adjusted to say .004 of an inch total, that means my machine is square within .001 over 12 inches. Mine is actually way better than that since I kept at it until it was right about.001 over the 4 feet.

I am sticking with the Pythagorean Theorem on this one:

a^2 + b^2 = c^2

Or for those trigonometrically impaired:

Screenshot 2024-11-28 151930.png

Check for square:

Screenshot 2024-11-28 152354.png

[Tailmaker]

That looks like you have an angle error of the gantry of about 0.12 degrees or 0.04" (x) on per 20" (y) gantry movement.
It is not much and you may be able to square that out with proper homing of both "x" drives, depending on your machine.

I have a similar issue with my machine but the machine is quite rigid so the gantry linear bearings would bind if I try to do that.
Fortunately, my Planet-CNC controller allows for the application of a coordinate transformation matrix to correct for such things.

Another option *might* be to tweak the post processor to apply such a coordinate correction but I don't know enough how to do that in a PP.

[GEdward]

This machine is adjustable from what I have been able to find in the forums but Laguna has NO documentation in their manuals to explain how to go about truing the X Y squareness or the Z axis perpendicularity for that matter. The gantry is adjustable by loosening the gantry bolts from the gantry carriage on either side. I do not know if there is a mechanism by which forward and backward adjustment can be done in a controlled manner but I suspect not. Regardless, you will need to loosen completely one side and loosen then snug the other side to allow the gantry to pivot on the carriage of that side. Then you will be able to nudge the completely loosened side in the direction you need to make the correction. A dial indicator is a big help in this process. It will allow you to know exactly how much you have moved the gantry; especially if you have done the math an know how much the gantry needs to move in order achieve squareness. Then make sure that as you tighten the gantry carriage bolts, to leave the indicator in place to be sure that the position is preserved. I would probably use two indicators, one on each side, to make sure nothing shifts during the tightening process. Or at least if something does shift, you will know by how much and in which direction.

Ed

[FabLab Wageningen]

Need to read more carefully Marcel. He wrote:

Left top side to bottom right is 28 1/4
Right top side to bottom left is 28 5/16

I apologize, Ted,

You are right. I totally missed the concept 'diagonal'. This is probably why I am allowed to teach statistics but not trigonometry (*).
Your expanantion makes total sense, but the thought of a square that is not square is too much for me on a Friday evening.

Marcel.
(*) and all rest assured: I will not do so.

[TReischl]

.......I do not know if there is a mechanism by which forward and backward adjustment can be done in a controlled manner but I suspect not. Regardless, you will need to loosen completely one side and loosen then snug the other side to allow the gantry to pivot on the carriage of that side. Then you will be able to nudge the completely loosened side in the direction you need to make the correction. A dial indicator is a big help in this process. It will allow you to know exactly how much you have moved the gantry; especially if you have done the math an know how much the gantry needs to move in order achieve squareness. Then make sure that as you tighten the gantry carriage bolts, to leave the indicator in place to be sure that the position is preserved. I would probably use two indicators, one on each side, to make sure nothing shifts during the tightening process. Or at least if something does shift, you will know by how much and in which direction.

Ed

All excellent advice. I do the slot thing I mentioned above because it gives an idea of how much I will have to move the machine. The amount is about 1/2 the difference of the distance between the slots.

What I eventually did is similar to what Peter Paquelli (CNCNutz) did/does. Place mechanical stops on the end of each axis. To square the machine just turn the power off and move the gantry until it contacts both hard stops. Works great and never goes out of adustment, like switches can do.