I think that o am probably going to run open loop with the best servo drive tune due to the fact that the motion is smoothest. I can absolutely tell you that in full closed loop at this point in time that there is too much noise transferred into the machine from the control managing the following error corrections. what is causing it to me could be a number of things, as you stated, drives, trapezoidal acceleration, bandwidth, etc.. I'm still new to all of this and still learning, so each day brings more Information to work with. oddly enough, Clint has been running machine with a completely untuned drive since day one and if it is anything like mine the max following error was Astronomical.. yet he managed to make parts that were to tolerance. a mystery to me.
I have the accelerated feed forward in the drive set to zero because it made things worse. the position feed forward is set at 100% and position gain at Max but once position feed forward is at Max the gain plays less of a part it would seem. changes to it seemed to make little difference unless they were really big.
I have some direction from PCW at mess to try using the encoder as a low bandwidth feedback loop similar to a linear encoder, I'm researching now how to configure that in the Hal to test and will report the results! thanks for you input, very insightful!
Chris
Sent from my XT1585 using Tapatalk