programing for backlash problem

[charansak 0]

Daer all l have any problem about backlash from machanic. l used FX3U + MR-J3-A servo amp program positioning Absolute but the position drift ,checking machanic loose from "0" to target position 500 running to 499 but back to "0" was over to-1 and shift every time. how to add pulse to DDRVA to target position and zero position? have any program sample? Thank you very much. cha

[Crossbow 295]

Using a pulse output command there is no feedback to the controller, so the controller has no way to know that the position is not where you requested it. If your concern is physical backlast then my recommendation is to write code to track which direction the servo moved last time, and when it switches direction add a desired number of pulses to the destination before you execute the movement.

[charansak 0]

Thank you very much crossbow Using a pulse output command there is no feedback to the controller, so the controller has no way to know that the position is not where you requested it. : The feedback to the controller was encoder from servo motor HF-SP 152B If your concern is physical backlast then my recommendation : Yes correct to write code to track which direction the servo moved last time, and when it switches direction add a desired number of pulses to the destination before you execute the movement. May you explain more or exsample for them , how to add insert to DDRVA function ?

[Crossbow 295]

You don't add to the command. You store the command destination in a register and modify the register value before executing the command.

[charansak 0]

so May you sample or more detail to explain Crossbow.Please.

[Crossbow 295]

I do not have time to write a sample. But I think my description was pretty complete. it should not be a problem for you to write some code to determine when a direction change occurs and add a number to a register.

[Carples 0]

I'd imaging even if you did what crosbow says over time backlash will increase requiring constant adjustment if positioing is critical. Add a 0 position sensor, or directly measure position with feedback attached to moving part not the motor. Standard mechanical solution to backlash is travel past 0 to say -5 then back to 0 always meassure in the same direction, this will produce accurate positioning. Make return distance 501