72
The smoothing filter can reduce machine impact by smoothing the command. The Pos_Filter_N parameter
define the time constant of this filter in ms. Normally, if the machine system oscillates when it starts and
stops, a larger Pos_Filter_N is suggested.
Notch filter
The notch filter can suppress resonance by reducing gain around the resonant frequency.
Antiresonant frequency=Notch_N*10+100
Setting Notch_On to 1 turns on the notch filter. If the resonant frequency is unknown, the user can set the
maximum value of the d2.14 current command small, so that the amplitude of system oscillation lies within
an acceptable range, and then try to adjust Notch_N and observe whether the resonance disappears.
Resonant frequency can be measured roughly according to the Iq curve when resonance occurs on the
software oscilloscope.
Table 7-5: Notch filter list
Panel
address
Internal
address
Name
Description
Default
Range
d2.03
60F9.03
Notch_N
Used to set the frequency of the internal notch filter to
eliminate mechanical resonance generated when the motor
drives the machine. The formula is F=Notch_N*10+100.
For example, if mechanical resonance frequency F=500 Hz,
the parameter setting should be 40.
45
0-90
d2.04
60F9.04
Notch_On
Used to turn on or turn off the notch filter.
0
:
Turn on the notch filter
1
:
Turn off the notch filter
0
0-1
7.3 Factors which influence tuning results
The control command is created by the upper controller (e.g. PLC):
The control command should be smooth as much as possible, and must be correct. For example, the control
command should not create the acceleration commands (inside the position commands) that the motor
cannot provide. Also, the control command should follow the bandwidth limit of the control loop.
The machine design:
In the actual application, performance is normally limited by the machine. Gaps in the gears, soft connection
in the belts, friction in the rail, resonance in the system – all of these can influence final control performance.
Control performance affects the machine’s final performance, as well as precision, responsiveness and
stability. However, final machine performance is not only determined by control performance.