User`s manual

Startup and Tuning
GFK-1742A Appendix D Tuning GE Fanuc Digital and Analog Servo Systems D-5
D
Tuning the Velocity Loop
The proper method to tune the velocity loop is to separate the velocity loop from the position loop.
To achieve this separation, a method must be used to directly send velocity commands without
using the position loop control. The DSM module has several modes that will allow the user to
send a velocity command directly to the velocity loop. Two methods are as follows:
Method #1:
The
Force Digital Servo Velocity
%AQ immediate command (34h) will send a velocity command
directly to the velocity loop. This command is different from the
Move at Velocity Command,
which uses the position loop to generate the command. This is important since we do not want
the position loop interacting with the velocity loop at this point in our tuning process. The
Force
Digital Servo Velocity
%AQ command allows the user to generate a step change in the velocity.
The velocity command step is then used to generate the velocity loop step response. The user
should note that when a velocity command step change is performed the acceleration is limited
only by the bandwidth of the velocity loop. In some applications this can cause damage to the
controlled device due to the high acceleration rate.
Method #2:
In some applications, method #1 introduces too large a shock to the device under control. In these
cases, another method to generate a velocity command is needed. The method requires that the
user set the position loop to an open loop configuration. The position loop is set to open loop by
setting the
Position Loop Time Constant
to zero and the
Velocity Feedforward Gain
to 100
percent. You can then use the
Move at Velocity Command
or a motion program to generate
velocity commands to the servo drive. The first parameter that needs to be adjusted is the
Velocity
Loop Gain
. The parameter adjusts the velocity loop bandwidth. As a starting point use the
following formula (also reference the Velocity Loop Gain Section):
Equation 1
Velocity Loop Gain =
J
J
Where:
J Load Inertia
J Motor Inertia
l
m
l
m
⋅
=
=
16
The
Velocity Loop Gain
calculated in equation 1 in many cases will not need to be altered.
However, due to the application (for example, machine resonance) the value may need to be
adjusted. To tune the
Velocity Loop Gain
the following procedure can be used:
1. Choose the method to introduce velocity command to the velocity loop. Method #1 and
Method #2 (above) are examples of methods to perform this task.