3rd party servo drive - UMI-7764 connections

[MlleDonna]

I am using a UMI-7764 with the PCI-7340 board to control a servo motor.

The motor, MPJ0922D3E, is from Parker Automation and uses the Aries servo drive AR-08AE.

I cannot tune the motor in MAX or make it spin in the Interactive mode.

In MAX, I am not getting any signal from the motor:

i.e. the step response of the motor obtained in all attmpts to tuning was a flat line at 0 count, while the command is set to 4000 counts.

In the 1-D interactive mode, all commands to start motion always lead to a following error and the axis is stopped. The motor never moved, even on the rare occasions when the LED tells me that the move was completed.

I know that the motor is not defective, because when the Aries drive is connected directly to a RS-232 9 pin port on my computer, I am able to tune it and have it spinning at different speeds using the drive software from Parker.

For now, I think the problem is faulty connections between the drive and the UMI.
The connections that are currently in place is attached.

Any help is appreciated!

Donna

[Brian_Beal]

One of the first things to check is to see if you are getting the encoder signal back to the controller.  Spin the motor by hand and see if the counts change in MAX.  Also make sure your limits are disabled in MAX, since you are not using them.  I am not familiar with the Aries drives, but at some point you must configure it to use the analog input for torque or velocity reference.

[MlleDonna]

When I spin the motor by hand in both ways, the counts in MAX only changes in one direction. i.e. Spin clockwise, counts decrease; spin counter-clockwise, counts stay the same.

I made sure via my direct connection to the drive software that the drive is in torque mode and I disabled the limits.

However, the auto-tuning still gives a flat step-response. In 1D interactive mode, I can now start the motor, but the counts do not change while it is spinning(it's spinning counter-clockwise) and I have to manually stop the move.

Thanks for responding.

[MlleDonna]

Also, one turn of clockwise spinning only decreases the counts by 5, while the encoder resolution is set to 8000 couts/RPM (post-quadrature).

[VGA_CD-ROM]

Hi Donna,
I would first check your encoder wire connections to make sure they are secure. I assume you also have your Axis Configuration in Measurement & Automation Explorer (MAX) set to a Servo Axis with Encoder feedback.

After you've checked those, what happens if you try switching your A, A-, B, and B- wires around to various positions?

Do you then still only get counts in one direction (be it clockwise or counterclockwise) and not the other?

As for the encoder resolution counting, the setting in the Encoder Settings page is what you provide based on either what the motor claism to be the actual counts per revolution, but it will not change the actual interpretation of encoder ticks. 

[MlleDonna]

Hi Vijay,

I have rewired the encoder cable and the position counts is working properly now: increase of 8000 for one clockwise rotation.

But the main problem remains:

When I auto-tune, MAX always freezes when the dialogue box show "calculating optimal control loop parameters" and the computer must be restarted. The step-response is still flat. Moves initiated in interactive mode is immediately aborted, the axis stopped because of a following error.

The system behaves as if the analog output is not reaching the servo drive. Is there anything to look out for while making the following connections?

Analog Output ---- AIN+

Analog Output Ground --- AIN-

[VGA_CD-ROM]

Hi Donna,
I see you have an issue open by email with another one of our application engineers int he Motion group. I've been working directly with him on the issue and one or both of us will get back to you when we have some more information. We'll probably need to escalate this further to R&D.

[MlleDonna]

Here I am, two weeks later with a happy ending.

Here's the summary of my adventure with the servo motor setup:
In chronological order,

Lousy encoder response?
-->Improper shielding of noise. 
Solution:
Change cables.

MAX unable to autotune/move in interactive mode?
Approach: Changed axis from 1 to 2.
-->Problem persists.
Looked at the drive, DISABLED before commanded to run and ENABLED when axis is killed.
-->Inhibit Output signal polarity.
Solution:
Need to have low inhibit output signal when drive is active/inhibit command is inactive.
Set UMI Switch S1 to HIGH ACTIVE.

Strange motor behaviour?
vibrating, shaking, wouldn't move when connected to controller, capable of spinning if rotated by hand.
-->Controller saturation/Integral windup
Solution:
Change PID control loop parameters. Kp=1, Kd=ki=0

Axis immediatedly killed in interactive mode for following error?
-->torque offset mismatch between drive configuration and controller configuration
Solution:
Set the same zero reference voltage in both devices.

Voila, the motor is ready to be tuned.