Accuracy of Square

[Close Work]

What are the travels of your router? Checking squareness of travel on a machine is a different question than checking the squareness of a machinist's square. The straightness of travel of your slides is part of testing the geometric accuracy of your machine. Roll, pitch and yaw are also errors in the slide travel other than straightness and squareness that affect the accuracy of your machine. A small square made perfect doesn't seem adequate to test the accuracy of your machine. You might make up a square of some kind from ground stock as square as you can make it that is as long in both directions as your X-Y travels and then take it to someone with a coordinate measuring machine large enough to measure it and give you some "calibration" on that tool which you could then use in evaluating your machine. If you are routing wood and the machine is small the 3-4-5 technique and cutting test parts may be all you need. Rout a square part or a cube part and test it by some of these self-checking methods mentioned. The errors in the machined part are the sum of all the errors in the machine operation. A standard test part is a good thing to have for occasionally evaluating and proving a machine.

 

[gentle]

erection of a perpendicular line - all you need tram bar

see http://www.nvcc.edu/home/tstreilein/constructions/lines/line1.htm for proper steps. you can make this as long as you would like thus no need for square guess work

 

[LIR&F]

Once again Forrest thank you for the definitive answer.

 

[Marty Feldman]

Your link gave me "Page not found".

see http://www.nvcc.edu/home/tstreilein/constructions/lines/line1.htm for proper steps. you can make this as long as you would like thus no need for square guess work

 

[67Cuda]

I am just finishing up a CNC router and want to square it up. I do have a machinist square, but I am suspect if it is very accurate. How can I test?

thanks
d1

How accurate does it need to be for a CNC router?

Tom

 

[Forrest Addy]

This reply is far too late and far too obvious. I'm embarrased not to have thought of it sooner.

Make a cutting test piece having sufficient number of easily measured features (circ. interpolated holes. grooves, ledges for example) to demonstrate positioning accuracy, linearity, axs perpendicularity and so on. The test piece (some stable easily machined material - MDF comes to mind) can then be measured bore to bore to feature. The machined piece can be inspected separately and calculations performed to determine feature location, squareness, etc. Linearity can be determined by machining one edge, rotating the part 180 degrees, dialing in the ends and scanning the length with the indicator to read double the linearity error. squarenes by machining the edge and rotation the part 90 degree etc also a doubling test.

This demonstrates point to point strock removal, rigidity, positioning, linearity, squareness, cutter comp, etc. and is a practical test of the machines actual performace as opposed to quantified data collection.

Satisfactory completion of a cutting test piece exercising all machine functions have been open and shut proof of final acceptance for all machine tools since day one and many clever, easy to execute teest piece designs have evolved over the years.

 

[chuckey]

This reply is far too late and far too obvious. I'm embarrased not having seen this thread before.

Roughly cut out two identical squares (or triangles) of decent plywood. label each one on the same face. screw them together with the labelled faces upwards. Now with your typical setup, machine one corner to a right angle, by running the cutter down the two sides. Unscrew the two pieces, stand them on a flat surface with the two machined edges together (one sheet will have to be reversed), use a feeler gauge to find out what the gap is and if its at the top or bottom. Voila! - you are calibrated, should be good for .003 over 2' - good enough for woodwork?
Frank