Kollmorgen Support Network

To better serve our Kollmorgen users with questions too complex to be addressed in this space, we made the painful decision to close this Community forum.

Please submit your question using the contact us form to reach our application engineering team.

For the immediate future, you can still access and read past posts.

Thank you for understanding and for participating in the community over the years.

Linear Motor Analog Control | 07 Jun 2023 | |

Linear Motor Analog Control

Hello everyone, 

Thank you to Michael_Brown455 and jcoleman02 to reply to my previous posts. We were having issues with using the AKD driver with ICH Linear motor and as soon as we operated in Analog mode in the velocity loop, we started having lound noise and vibration from the motor. We tried a lot of steps which did not work and at the end, I ended up resetting the driver setting to factory default and this issue was solved.

We are having some more problems now. So we are using the ICH22-100 Linear motor and AKD driver AKD - P 012 07 - NB CC - E000. We have a linear application and the motor is coupled with Linear guides on 2 sides. We are using Galil motion controller to establish a position loop(outer loop) and operating the drive in Analog velocity mode. We are using an quadrature incremental encoder for position feedback with 2.5um resolution.

The Galil Motion controller has analog output ( -10V to +10V) range with 12 bit resolution which is used as analog input for the AKD.

The problems we are facing now is unnecessary disturbance when we try to run a sin wave, we see noise and resonance at different frequency values. Everytime we change the frequency of the sinusoidal profile, we notice resonance at different frequency values. This can be seen in the pictures where the red FFT profile represents the input profile and the blue represents the output FFT we get from the motor. This perfomance for the linear motor is worse than our conventional ballscrew systems that we were using which were way less disturbance prone than this. Is there something we can do to improve the perfomance, any suggestion is welcome.

We also tried to tune the Master control loop(position loop) but it did not make any difference. The problem we think is with the drive or the motor settings or the motor itself here.

In the pictures attached, the acceleration has units of g and displacement has units of mm.

We also contacted our local Turkey Kollmorgen support but they were of no help. 

Thank you

P.S. I cannot comment on the posts I made on Kollmorgen so I have to create a new post every time, can I change this from somewhere? 

[email protected] (190.11 KB)
[email protected] (165.76 KB)
LomaProfile.png (404.8 KB)

Comments & Answers

jcoleman02 | Wed, 06/07/2023 - 21:46

Tamim,

I'm not sure how much I'll be able to help you on a forum like this.  I recommend contacting our Applications Engineering group in Turkey for assistance.  Contact Us | Kollmorgen

From your description, the cause of the problem could be any number of things.

EMI - Make sure all your cables are properly shielded and ground the shields.  Make sure the analog input wiring is twisted-pair and shielded.  Follow the recommendations in the articles on this site: Grounding and Shielding Landing Page (Prevent Noise Issues) | Kollmorgen

Tuning - The resonance could be due to improper tuning.  Since the velocity loop is in the drive and the position loop is in the Galil, you must tune both to make sure they are balanced.  You can use the Performance Servo Tuner in Workbench to autotune the velocity loop in the drive, but you will likely need to make manual adjustments to the velocity gains while tuning the position loop in the Galil.

Cables - The motor power and feedback cables for the linear motor and linear encoder will move as the motor moves.  Make sure to follow proper bend radius limitations.  If you are powering the drive with 400V/480VAC, putting the motor power cable and the feedback cable together in a cable track may not be ideal.  The cables should be separated by several inches at least.  Also, make sure that the cables you are using are rated for flexing.  If the cables are not flex rated, as the cables flex/bend, the shield moves and can open up allowing EMI.

Encoder resolution - The 2.5 micron  AquadB encoder is relatively low resolution.  2.5 microns = 12800 lines per 32mm (1 pole-pair) = 51200 counts (post quad). That is not the worst resolution, but it is relatively low compared to very high resolution encoders.  The resolution affects the overall performance.

Since you are running the drive in Analog Velocity mode and closing position loop in the Galil, you are using the drive's Emulated Encoder Output (EEO).  The EEO resolution setting in the drive needs to match the expected EEO signal resolution in the Galil.  The EEO resolution needs to be set as high as possible without exceeding the maximum frequency limit.  A low EEO resolution is the same as using a linear encoder with a low resolution.  It will give poor performance and make it difficult to tune.

I can't really address any possible errors of mechanical installation or incorrect drive configuration in this forum.

About this Question

Tamim
Taxonomy: