18 / 52
DRIVES, MOTORS + SWITCHGEAR
T
he effect of different load inertias is illustrated with experiments
using real life components. In addition, tuning the controller is
covered in detail. Previous articles described systemtesting, servo
operating principles and a practical mechatronics case study. This article
describes the effect on the system when the load inertia is changed.
Equipment used
For the test, a brushless motor was mounted on a chassis and coupled
to a Techlyn brushless drive. Test response was measured using a
Personal Computer (PC) coupled via a serial port to the drive. Step re-
sponse was displayed while the tuning parameters were interactively
changed. Tests were carried out on the motor with no load, as well as
with flywheels dimensioned to provide load inertia of 3 X and 10 X
motor rotor inertia. Conventional wisdom states that inertia matches
of up to 10:1 are acceptable, with the optimum point being at 3:1.
Our experiments sought to see if this was indeed true. Our drive is
controlled by a Microchip PIC18F2331 microcontroller. This device is
highly optimised for motion control with all time intensive functions
being performed by hardware registers. Moves are commanded via a
serial port or step and direction digital inputs. The step and direction
inputs make the system resemble a step motor system (with encoder
position maintenance). The PC test program sends move commands
and graphs the response with respect to time.
Tests
The test motor had a rated rotor inertia of 0,23 kg.cm². Two flywheels
were machined with inertias of 0,69 kg.cm² (3:1 inertia match) and
2,3 kg.cm² (10:1 inertia match). In total, six tests were performed
and graphed.
What do the parameters on the graph mean?
Ideally, the July 2015 Electricity+Control article (Brushless servo
operating principles) should be to hand [2].
Figure 2
• On the lower left is the Servo Parameters Pane
• Kp is the Position Gain
• Kd is the Derivative Gain
• Ki is the Integral Gain
• Il is the Integration limit
• The top half shows the rotor position with respect to time
• At the lower right is the Configuration Pane which is self explana-
tory. The horizontal dotted lines are the position thresholds which
are used to measure times. The rise time for instance is the time
measured fromzero time to the first crossing of the lower threshold
PID Servo Control
These are the control filter parameters. The output to the motor
amplifier is the sum of three components:
• One, proportional to the position error providingmost of the error
correction (Kp)
• One, proportional to the rate of change in the position error which
provides a stabilising damping effect
Matching
load and servo motor sizes
Glyn Craig, Techlyn
This article, the fourth in the series, deals with stability in practical systems.
Figure1: The
mechanical part of
the test rig.
Electricity+Control
July ‘16
16