deltaa
-
Posts
12 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Downloads
Posts posted by deltaa
-
-
I used the latest version of the spreadsheet(v2.8) to generate the data.
I changed Ixx71 and Ixx91 to match with the rest of the motor settings (which derived from DLS). Ixx69 was generated from the spreadsheet.
It does not look to me like you followed the steps in the manual properly (make sure you look at the latest revision on the website). But also, your settings are not consistent with the spreadsheet either:
- Ixx71 = 4096, this is using the older micro-stepping technique
- Ixx69 = 4!
- Ixx91 = 0!
It would be helpful to take a look at all of your I variable settings (global), and not only motor 1's.
-
The motor I used is VEXTA PK264-03B.
Step Angle
1.8 °
Connection Type
Bipolar (Series)
Current per Phase (A/phase)
2.1 [bipolar (Series)]
Resistance (Ω/phase)
1.26 [bipolar (Series)]
Inductance (mH/phase)
2.4 [bipolar (Series)]
I used the latest Geo Brick LV Setup Assistant to generate the settings(attached). With the setting (I169=4), the motor quickly developed following error and stopped. When I increase I169 to over 100, the motor can move continously (although not very smooth, may need further current loop tunning).
Please use the "Geo Brick LV Setup Assistant" Excel spreadsheet to generate all the proper settings for your Geo Brick LV and your motors.
You can get the "Geo Brick LV Setup Assistant" file from the filedepot: http://forums.deltatau.com/filedepot/download.php?f=Turbo%20PMAC/Tools/Geo%20Brick%20LV/Geo%20Brick%20LV%20Setup%20Assistant%202.8.xls [FILE REMOVED]
-
Hi,
I need help on how to properly setup Ixx69 on a Geo Brick LV controller to drive a direct micro-stepping motor. If I follow the instruction in pg.158 of the Geo Brick LV user manual, or use Geo Brick LV setup assistant, both give a small value of Ixx69 (under 42), when the value used to drive the motor, following error will be quickly developed. I'm wondering if I missed something here.
Thanks
-
Setup the test according to the suggestion, ran several tests and checked ADC readback. When motor is running at a lower speed, the performance is acceptable, with the increase of the speed, more vibration introduced and ADC B readback became distorted, any suggestion? Thanks
Setup: 2048 steps/full step, motor current around 3amp RMS.
I122=100, looks good and feels acceptable
I122=200, looks ok
I122=300, looks and feels bad
I122=300, difference Geo Brick
2 suggestions:
1. Use the Geo Brick LV Setup Assistant Excel spreadsheet available through FileDepot Link (in the menu bar of the forum) in order to get 2048 microsteps per full stepper of the motor.
2. Reduce the I7004 and I7104 (Pwm Dead-time) to a value of 4. This should reduce your zero-crossing distortion.
-
-
Geo Brick LV to drive a two phase stepping motor in direct micro-stepping mode with 32 micro steps per full step.
Current waveform distortion noticed around zero-cross area as shown in the attachment.
Motor current: 5 A peak to peak
Frequency: 20 ms/div
Any idea on how to correct the zero-crossing distortion? thanks.
-
Thanks Steve, both deadband setting and j! solved the problem.
-
The PID gains are
I430=4096;Motor 4 PID Proportional Gain
I431=0;Motor 4 PID Derivative Gain
I432=16384;Motor 4 PID Velocity Feed Forward Gain
I433=10000;Motor 4 PID Integral Gain
I434=0;Motor 4 PID Integration Mode
I435=4096;Motor 4 PID Acceleration Feed Forward Gain
I436=0;Motor 4 PID Notch Filter Coefficient N1
I437=0;Motor 4 PID Notch Filter Coefficient N2
I438=0;Motor 4 PID Notch Filter Coefficient D1
I439=0;Motor 4 PID Notch Filter Coefficient D2
Do I need to tune the position loop for an external pulse/direction setup?
By the way, when I jog to specified position using #4j=***, the problem does not happen.
It only happens when using jog stop command j/ when the motor is moving.
-
System config,
==
External micro stepping amplifier
Two phase stepping motor
Geo Brick LV drive channel 4 output Step/Dir signal
==
Parameters used,
==
I403=$3504
I404=$3504
I8003=$C78018
I7040=8
==
Use jog command #4j+ or #4j- to start motor, motor runs without problem, then #4j/ to stop jogging, after the motor stops, it's still vibrating. Use scope to check the pulse and direction signal from Geo Brick LV, these two signals still have outputs. Anyone has similar experience? What could be wrong?
Thanks
-
The mircostepping uses a simulated feedback in the Turbo PMAC's memory and doesn't use any of the DSPGate1 ASIC functionality which position compare is one of them. If you want to do a position compare, the best way is to set up a virtual motor using an internal pulse and direction on one of the available channels on one of your DSPGate1 ASIC(s) and set it as a salve to your microsteped channel and perform all the position capture on the virtual axis.
I hope this helps.
Trying to use motor #1 for direct micro stepping and motor #4 as the virtual motor with internal pulse & direction wrapped back, with the following setup, both motors work fine individually.
; ECT for motor #1
I8000=$6800BF
I8001=$10020
I8002=$EC0009
; ECT for motor #4
I8003=$C78018
;virtual motor #4
I400=1
I401=0
I402=$7801C
I403=@I8003
I404=@I8003
I405=@I8002
I406=1
I7040=8
I7046=3
When trying to move motor #4 along, position compare works no problem. When enable position following on #4 and moving motor #1, position compare does not work. It seems motor #4 does not "follow". What could be wrong?
Thanks
-
Anyone has experience on how to setup position compare function for Geo Brick LV direct microstepping mode without using encoder? Thanks.
Setup Ixx69 for direct micro stepping
in Turbo PMAC
Posted
I used the latest version of the spreadsheet(v2.8) to generate the data.
I changed Ixx71 and Ixx91 to match with the rest of the motor settings (which derived from DLS). Ixx69 was generated from the spreadsheet.
I attached the screenshot of the spreadsheet and the configuration file for your reference. Motor is setup on axis #1.
Thanks.