BenJackson wrote:dirktheeng wrote:Also, the machine will change by about 0.0012" over 96 inches for every degree F shift in temp.
Easy to compensate in software.
This got me thinking about corrections in software that could be made.
What if we could set up a very simple measurement of the linear accuracy of the mill that is very cheap and easy... say imaging a line laser beam down the length of an axis. It should be very easy to apply a transform in the machine interpretor (mach 3, emc, etc) to compensate for that error. This, of cource, would only work for regular errors... random error cannot be compensated. Then the expensive rails would not be necessary. you could use a rail that had a 1 inch bow in it and it wouldn't make much difference. You could also theoretically compensate for temperature and machine expansion and all that.