Archive for March, 2008

Today I connected the motors to the V90 machine, connected the motors to the HobbyCNC controller board and started EMC2. In the EMC2 interface AXIS the cartesian coordinate system is shown along with the position of the bit, illustrated with a cone:

I connected the motors to the board so the following coordinate system would be used on the machine. I picked this system because it seems customary to call the longest axis the X axis. Plus this system makes the most sense to me. However, note that other people may swap the X and Y axis from what I am using here:

Jogging each axis causes the cone (representation of the bil) to move in the display, and at the same time I observed how the machine moved. In AXIS a positive movement in any direction moves the cone towards the letters ‘X’, ‘Y’ and ‘Z’. However on the machine every axis moved in the opposite direction. This meant that I needed to reverse the direction of the motors.

The motor direction is reversed by swapping the ‘A’ and ‘a’ wires. Note that this may not be the case for all motors. Check the documentation that came with your contoller board to be sure.

Once the motors had been reversed the machine movements copied the movement of the cone in AXIS.

The next step was to make sure the machine was moving roughly the correct amount each time. I calculated a configuration (essentially the steps needed to move one inch) for each axis. On the V90 the X and Y axis are 5 threads per inch, but the Z axis is 12 threads per inch. More on how this calculation is performed in a later post. Plugging these values into the EMC2 configuration should do the trick.

I then zeroed the position of each axis in AXIS and used a ruler to measure the location of the Z axis assembly from the side of the gantry, which is the distance the Y axis is moved in from the side. I then jogged the Y axis until the display showed it had moved +0.5 inches. Remeasuring confimed that the Y axis had indeed moved half an inch.

These steps were repeated for the Z axis.

To be sure I also created an EMC2 configuration for the V90 in millimeters and repeated the test. This time I jogged each axis +0.1 x 10 times, to move the axis one millimeter.