In control systems that process one block of cnc code at a time, i.e. one coordinate at a time, it is not possible to end the movement on a given vector at a speed other than zero. This is because the driver does not analyze data on the successive vectors after the current execution. Not knowing what move will be next, he must stop in order to start the next move after downloading the next block.
This results in a situation in which the movement along the tool path is interrupted even though the successive vectors are, for example, tangent to each other. In tool paths where long vectors dominate, it does not matter much, because when moving along such a vector, the machine has a long enough path to reach the set speed. The time after which the machine reaches this speed and whether this speed can be reached at all on a vector of a given length, depends on its value and the set acceleration.
The problem appears when the vectors are short enough that the set speed cannot be reached on them. In this case, the average feedrate is much lower than the commanded speed. This results in a significant reduction in machining efficiency and, moreover, due to frequent stops with accelerated tool wear caused by frequent changes in cutting data.
This problem is especially visible when working in HSM ( High Speed Machining) consisting in working with significantly increased cutting speeds. In this technology, the feed rate is greater than the speed of temperature propagation in the workpiece, which results in almost all of the energy accumulated in the removal of the chip is ejected with it. As a result, the tool and material heat up less during cutting than with conventional cutting.
In order to stay ahead of temperature propagation in the material and at the same time keep the chip thickness at a safe level, the spindle speed must be increased accordingly. So large that in hard materials at low feed rates (below the HSM) it would overheat and damage the tool.
The use of HSM technology on machines with such a control system is not possible because frequent stops of the tool in the material result in its frequent overheating, which results in very fast wear.
To eliminate these problems, the machine should, as far as possible, maintain the feed rate at the level set by the operator. The maximum speed in the node between the vectors should depend on the angle between them and the shape of the tool path that these vectors represent.
The solution may be to analyze more than one vector at a time, which will result in a non-zero velocity value at the tool path nodes. Unfortunately, we cannot analyze only one forward vector, because it (one) can be short enough that it will not be possible to reduce the velocity along its length to the value of its limit at the end of the vector.
It is therefore necessary to iterate successive vectors and to modify (boost) the initially zero velocities in the nodes between the vectors that will meet the assumed speed limits in the tool path nodes, and at the same time limit the time in which the tool moves at a speed lower than the set speed.
The developed method was called Dynamic Vector Analysis ™, and its implementation was a complete success. With a complicated tool path of several tens of thousands of vectors with a total length of approx. 20 m, with a set feed speed of 100 mm / s and with Dynamic Vector Analysis on, the working time was less than 4 minutes, while with the analysis turned off, it was approx. 20 minutes.
This colossal difference in machine operation time allows you to achieve significant benefits when performing work that uses complex tool paths such as machining injection molds, dies, dies, casting models, dies and other tools.