I think my first post was misunderstood or poorly stated by me. Sorry about that. What I meant is that if I move lets say the X axis forward 3 times at 5mm intervals (it moves from 0.000 to 15.000)... if I then move back again 3 times at 5mm intervals I don't get back to 0 due to backlash. I'll post my backlash numbers once I have access to the machine again in a couple of days, but its not 5mm backlash.
I'm not working on a commercial project or intend to use this mill for anything other than this research. I'm doing a university project where the goal is to automate a manual mill (with really bad screws) using linear scales and steppers. My goal is to come up with a setup/config or something useful (hal module) that would kept 2 axis syncronized to compensate for terrible backlash. I'm trying to use as much of what linuxcnc/machinekit has built it to address this problem, and if that is not enough try to modify it (machinekit) to get better results. I'll publish my numbers in a couple of days (backlash and result after the acceleration). I record the encoder values into a sepparate file and I generate a BMP with all the encoder pairs to keep a consistent test environment for changes. On Tuesday, July 23, 2019 at 2:56:34 PM UTC-3, justin White wrote: > > I can’t say for certain but I don’t believe that the backlash comp on one > axis effects the other it shouldn’t really have to. Is there really that > much backlash that you think it should allow the other axis to slow down or > stop? I read your first post not to mean that you had 5mm backlash, I just > assumed that was the arbitrary number you were moving the axis. If it > really has that much I think you should do something about the hardware. > .030” would be extreme backlash imo, if we’re talking about close to .200” > I wouldn’t bother trying to compensate for that. Running a mill with that > much backlash is going to be a horrible experience with chatter and > accuracy. > > Otherwise if your acceleration is setup properly it can easily wring out > the backlash before your other axis moves enough to cause arcs to be > misshaped. Depending on how bad it is you could see a minor step in the > work but at that point i think you should tighten up the machine. -- 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/e7828eff-3fad-43cf-b769-965610f1af34%40googlegroups.com.
