Nel Pretech’s Dave Nelson gives tips on CMM programming. In this installment, Dave walks us through the Cosine error that can occur; what it means and how to account for it.
A coordinate-measuring machine, or CMM, is a powerful piece of technology that can play a number of roles in the commercial world. Used to take highly-accurate measurements of three-dimensional objects, a CMM is able to create plans and drawings of parts and other items that can then be recreated in a manufacturing environment.
While there are different styles and types of CMM’s, the basic idea remains the same from one machine to the next. CMM’s generally work on three axes – known as X, Y, and Z – and record coordinates for all three axes to accurately measure an item. The part of the machine that traces the item to records its dimension is known as the probe. There is contact probing, which involves physically touching the item at a collection of points, or non-contact scanning, which uses lasers or white light to measure the item in question.
Obviously, there is a lot of impressive technology at work for a machine to be able to take precise measurements of a three-dimensional object. However, like all machines, CMM’s require a significant amount of human input to work correctly and complete the tasks they are designed for. A CMM programmer is tasked with creating programs that will instruct the CMM on how to take measurements of a specific item. Without these programs guiding the CMM’s actions, the machine would be useless and unable to produce the data that it needs to gather.
A big part of what a CMM programmer needs to do is properly calibrate the machine so the measurements are as accurate as possible. By testing on items of known-size, the programmer can dial in the CMM to make sure the margin for error is squeezed down and the final measurements can be trusted. Technical knowledge, along with experience, are key to performing this process correctly.
One of the most common causes of error in the CMM process is known as the Cosine Error. The cosine error relates to the calibration of the probe itself, and how it interacts with the item that it is measuring. The video included in this blog post does a good job of describing the cosine error in simple terms.
Basically, any probe with a small round tip is prone to the cosine error when the probe approaches the item from a non-normal (or non-90*) angle. During calibration, the CMM learns what size the round tip is on the end of the probe, and then takes that measurement into account while gathering the data. When the probe is moving in a 90* manner from the surface of the item that it is measuring, there is no problem and the CMM is able to make the calculations correctly.
However, if the probe approaches and contacts the item from a non-90° angle, the difference in vector will change the calculations that need to be done and the resulting measurements could be slightly off. In a world where every small error is magnified, it is crucial that CMM programmers understand what the cosine error is, and how to avoid it.