Axis parameters, axis monitoring and axis calibration

Determination and setting the most important axis parameters: maximum velocity, following error monitoring, end position monitoring and those for dynamic behaviour and calibration.

Interface

The axis parameters are accessed via the System Manager.

General axis parameters

A variety of further parameters can be adjusted on the global tab for the axis. These determine the behaviour of the axis. A range of monitoring functions and status signals are also controlled here.

You can specify the level to which all commands sent to the axis can be limited with the maximum velocity. You should set a value here of about 90 to 95% of the maximum velocity of the drive. The remaining control capacity is required for the controller in order to ensure correct positioning.

When an NC interpreter is used, certain commands will cause the axis to be positioned using an adjustable control for the high-velocity travel. Not more than about 90% of the maximum velocity of the axis should be chosen here.

The min and max manual operation velocities can be freely chosen, and largely depend on the ”confidence” of the user. In practice it has been found appropriate to set these velocities with a ratio of about 1 to 5. The values must, of course, be smaller than the maximum velocity.

The reference travel velocity in the positive direction is only effective on axes with incremental encoders. This is the velocity with which the axis moves when looking for the calibration cam. The direction is set on the encoder's incremental tab.

The reference travel velocity in the negative direction is only effective on axes with incremental encoders. This is the velocity with which the axis moves when leaving the calibration cam and when looking for the synchronisation pulse. The direction is set on the encoder's incremental tab.

The positive / negative direction pulse widths specify the distances to be travelled by the PLC when the axis starts in the corresponding movement modes. Fuller information is found in the PLC's function block descriptions.

You can activate and set the position of a movement limit (software limit switch) both below and above.

The following error (the difference between the set and actual positions) is continuously determined for axes with an analog tab (i. e. not for high/low-speed or for stepper motor axes). An adjustable threshold for monitoring these values can be activated. It is possible here to specify a filter time which may prevent an alarm being signaled if the threshold is briefly exceeded, for instance, during acceleration or braking.

Two independent windows can be specified around the target position of a movement (target position windows). Any size can be chosen for the windows. This makes it possible to create a signal for the PLC at any desired distance from the target. One of these windows also offers a filter time. The signal is only generated if the axis remains continuously within the window for whatever period is set there.

It is possible for high/low speed axes to approach each destination always from the same direction, and having covered a minimum travel distance. This allows a greater precision to be achieved if the axis does not behave symmetrically, or if there is mechanical play. This behavior can be activated, and the travel distance can be specified. The arithmetic sign of the specified distance determines the direction.

In servo axes "reverse backlash compensation" can be used to compensate for mechanical play in the axes. The value entered here as the "reverse backlash" specifies the distance by which a target is exceeded. The arithmetic sign of the value specifies the direction of the overshoot. If the reverse backlash is positive, the target is only overshot when moving in a positive direction. Movement in the direction of smaller positions is not affected. Negative reverse backlash results in the opposite behavior.

Axis dynamics

On the dynamic tab for an axis, values for the acceleration, deceleration and jerk are to be set. Two methods are available: direct entry as a numerical value, or indirect specification via the run-up time and the characteristic acceleration curve.

The values to be set here are the parameters to be observed by the set value generator. Drive manufacturers often mention values that would result in very hard axis positioning behavior if they were to be set here. This is because the values mentioned represent the limiting value of the motor's or of the power section. Usually, however, the machine's behavior would then be very dependent on the load. The result of this is considerable variation in the following error, which causes the position controller to become active. However, the axis is hardly able to react to the controller output, because it is already operating at the limit of its current.

The "correct" values for an axis depend heavily on the desired behavior and on the properties of the drive technology and of the machine. They can only be determined through test runs. Values should be tested as they are increased step by step, observing the following error when starting and stopping. It has been found in practice that the numerical value for the jerk should be somewhere between twice (for processing axes) to ten times (for transfer axes) greater. For indirect adjustment, these correspond respectively to soft and hard adjustments.

Use the values for the acceleration, deceleration and jerk from TwinCAT Scope for the adjustment. Whether the values for the acceleration A+ or deceleration A- are really achieved depends on the critical jerkJ±, and on the ratio (A+)²/V or (A-)²/V, where V represents the set velocity. Check, in particular, whether the actual acceleration value can follow the set acceleration value (and the same for the deceleration).

The effects of incorrect cycle time settings

Two errors result from incorrect cycle time settings, i.e. a difference between the set cycle time (NC task; SEC task; online; cycle ticks in ms) and the real cycle time taken by the SEC task:

Velocity pre-control: if the real cycle time TC_eff is greater than the set cycle time TC_saf the time axis becomes stretched out (scaling TC_eff/TC_saf), and this implies a non-linear transformation of the setpoints. Every calculated set velocity is maintained (TC_eff/TC_saf) times longer than initially calculated. The integrated effective target velocity reaches a target position which is (TC_eff/TC_saf) times larger than the calculated one.
Position control: since the target velocity and the set position do not match (v(t) ≠ dp(t)/dt), the position controller must work against the pre-control. The resulting behavior depends on the proportionality factor, the size of the cycle time error setting, and the controller resources (reference velocity). In any event there will be a lag error in those phases where a force is present, and large overshoots.

Delay between pre-control and position control

The purpose of the delay generator is to compensate for the effects of a system-related delay time (hardware/communication) in the P part of an position controller between the output of the set velocity value and its effect on the positioning system. Instead of the output velocity v_o(t) as a function of the velocity pre-control v_g(t) and of the P part K_v (p_g(t) - p_i(t)), where p_g(t) is the position setpoint and p_i(t) is the actual position value,
v_o(t) = v_g(t) + K_v ( p_g(t) - p_i(t)) + ...
 the position setpoint is output subject to a delay period of delta, so that
v_o(t) = v_g(t) + K_v ( p_g(t-delta) - p_i(t))+ ...

Parameter: delta delay time in s >= 0.0

Parameters for axis calibration

These settings are only necessary for axes with incremental encoders. So that the actual value system in such axes can be adjusted to the machine's reference system, a calibration travel is started. A number of parameters and signals are required for this.

When the calibration travel of an axis is started, it begins with the velocity reference travel in a positive direction (the global tab for the axis) mentioned above. The actual direction is specified by the encoder's incremental tab. The axis stops when the signal bit for the calibration cam is set in its NC-PLC interface. This cam allows a specific pulse to be selected on rotating encoder systems with one synchronizing pulse per rotation.

The axis now starts again to leave the cam. It now executes the velocity reference travel in the negative direction (global tab for the axis) until the signal bit is cleared. Without changing its velocity, the axis now continues until the encoder's synchronous latch has asserted. Only now does the axis stop.

The distance covered between the assertion of the latch and the location at which the axis of comes to a halt is determined. The total formed by adding this distance to the reference position (incremental tab of the in encoder) is set as the new actual axis position. This procedure allows the synchronous pulse from a selected rotation of the axis to be assigned to an absolute position in the machine.

The calibration is started by means of <F9> under Axis Online in the System Manager .