Thanks for the feedback. I have not used backlash compensation. The reason is that I thought that using a closed loop would take care of the need to estimate a backlash value (since the linear scale gives the actual position). I'll try increasing the acceleration to allow the PID to compensate the backlash faster.. and will read about backlash compensation on the axis section.
The idea of my project is to add linear scales to the worse possible X and Y mill and get good results out of it. I could have use better components (like antibacklash circular balls screws) but I'm trying to figure out if by attaching a couple of simple linear scales to the mill I can reduce most of the mechanical issues with a closed loop. I'm thinking about writting a HAL component that will detect a backlash condition (sign change on the PID output) and stop all the axis movement until I can compensate the backlash on the axis with problems.. and then continue the movement. I've tried doing something similar measuing the error on the PID and if this error was greater than a setpoint value stop the other axis until they both catch up.. but this did not work great. IT ends up moving one axis and then the other most of the time. -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/machinekit/dad7d129-841f-4859-b513-102ac61dbbd3%40googlegroups.com.
